chore: bump version to 7.0.6

- Fix incorrect package names (@ccw/cli, @anthropic/claude-code-workflow) to claude-code-workflow
- Add workflow examples page with 8 practical artifact samples
- Optimize sidebar navigation with emoji icons and second-level grouping
- Add usage examples to commands and skills documentation
- Update qa/issues.md with current status
- Sync Chinese documentation changes

.claude/skills/team-planex/SKILL.md

@@ -1,155 +1,105 @@
 ---
 name: team-planex
-description: Unified team skill for plan-and-execute pipeline. 2-member team (planner + executor) with wave pipeline for concurrent planning and execution. All roles invoke this skill with --role arg. Triggers on "team planex".
+description: Unified team skill for plan-and-execute pipeline. Uses team-worker agent architecture with role-spec files for domain logic. Coordinator orchestrates pipeline, workers are team-worker agents. Triggers on "team planex".
 allowed-tools: TeamCreate(*), TeamDelete(*), SendMessage(*), TaskCreate(*), TaskUpdate(*), TaskList(*), TaskGet(*), Task(*), AskUserQuestion(*), Read(*), Write(*), Edit(*), Bash(*), Glob(*), Grep(*)
 ---
 
 # Team PlanEx
 
-2 成员边规划边执行团队。通过逐 Issue 节拍流水线实现 planner 和 executor 并行工作：planner 每完成一个 issue 的 solution 后立即创建 EXEC-* 任务（含中间产物文件路径），executor 从文件加载 solution 开始实现。所有成员通过 `--role=xxx` 路由。
+Unified team skill: plan-and-execute pipeline for issue-based development. Built on **team-worker agent architecture** — all worker roles share a single agent definition with role-specific Phase 2-4 loaded from markdown specs.
 
-## Architecture Overview
+> **Note**: This skill has its own coordinator implementation (`roles/coordinator/role.md`), independent of `team-lifecycle-v5`. It follows the same v5 architectural patterns (team-worker agents, role-specs, Spawn-and-Stop) but with a simplified 2-role pipeline (planner + executor) tailored for plan-and-execute workflows.
+
+## Architecture
 
 ```
-┌──────────────────────────────────────────────┐
-│  Skill(skill="team-planex", args="--role=xxx") │
-└────────────────┬─────────────────────────────┘
-                 │ Role Router
-         ┌───────┴───────┐
-         ↓               ↓
-    ┌─────────┐    ┌──────────┐
-    │ planner │    │ executor │
-    │ PLAN-*  │    │ EXEC-*   │
-    └─────────┘    └──────────┘
+┌─────────────────────────────────────────────┐
+│  Skill(skill="team-planex", args="需求描述")  │
+└──────────────────┬──────────────────────────┘
+                   │ Always → coordinator
+                   ↓
+          ┌──────────────┐
+          │  coordinator  │  Phase 1-5 + dispatch/monitor commands
+          └───┬──────┬───┘
+              │      │
+              ↓      ↓
+     ┌──────────┐  ┌──────────┐
+     │ planner   │  │ executor  │   team-worker agents
+     │ PLAN-*    │  │ EXEC-*    │   with role-spec injection
+     └──────────┘  └──────────┘
 ```
 
-**设计原则**: 只有 2 个角色，没有独立 coordinator。SKILL.md 入口承担轻量编排（创建团队、派发初始任务链），然后 planner 担任 lead 角色持续推进。
-
 ## Role Router
 
+This skill is **coordinator-only**. Workers do NOT invoke this skill — they are spawned as `team-worker` agents directly.
+
 ### Input Parsing
 
-Parse `$ARGUMENTS` to extract `--role`. If absent -> Orchestration Mode (SKILL.md as lightweight coordinator).
+Parse `$ARGUMENTS`. No `--role` needed — always routes to coordinator.
 
-Optional flags: `--team` (default: "planex"), `--exec` (execution method), `-y`/`--yes` (auto mode).
+Optional flags: `--exec` (execution method), `-y`/`--yes` (auto mode).
 
 ### Role Registry
 
-| Role | File | Task Prefix | Type | Compact |
-|------|------|-------------|------|---------|
-| planner | [roles/planner.md](roles/planner.md) | PLAN-* | pipeline (lead) | **压缩后必须重读** |
-| executor | [roles/executor.md](roles/executor.md) | EXEC-* | pipeline | 压缩后必须重读 |
-
-> **COMPACT PROTECTION**: 角色文件是执行文档，不是参考资料。当 context compression 发生后，角色指令仅剩摘要时，**必须立即 `Read` 对应 role.md 重新加载后再继续执行**。不得基于摘要执行任何 Phase。
+| Role | Spec | Task Prefix | Type | Inner Loop |
+|------|------|-------------|------|------------|
+| coordinator | [roles/coordinator/role.md](roles/coordinator/role.md) | (none) | orchestrator | - |
+| planner | [role-specs/planner.md](role-specs/planner.md) | PLAN-* | pipeline | true |
+| executor | [role-specs/executor.md](role-specs/executor.md) | EXEC-* | pipeline | true |
 
 ### Dispatch
 
-1. Extract `--role` from arguments
-2. If no `--role` -> Orchestration Mode (SKILL.md as lightweight coordinator)
-3. Look up role in registry -> Read the role file -> Execute its phases
-4. Unknown role -> Error with available role list: planner, executor
+Always route to coordinator. Coordinator reads `roles/coordinator/role.md` and executes its phases.
 
-## Input Types
+### Orchestration Mode
 
-支持 3 种输入方式（通过 args 传入 planner）：
+User provides task description.
 
-| 输入类型 | 格式 | 示例 |
-|----------|------|------|
-| Issue IDs | 直接传入 ID | `--role=planner ISS-20260215-001 ISS-20260215-002` |
-| 需求文本 | `--text '...'` | `--role=planner --text '实现用户认证模块'` |
-| Plan 文件 | `--plan path` | `--role=planner --plan plan/2026-02-15-auth.md` |
+**Invocation**: `Skill(skill="team-planex", args="<task-description>")`
 
-## Shared Infrastructure
+**Lifecycle**:
+```
+User provides task description
+  -> coordinator Phase 1-3: Parse input -> TeamCreate -> Create task chain (dispatch)
+  -> coordinator Phase 4: spawn planner worker (background) -> STOP
+  -> Worker (team-worker agent) executes -> SendMessage callback -> coordinator advances
+  -> Loop until pipeline complete -> Phase 5 report + completion action
+```
 
-### Role Isolation Rules
+**User Commands** (wake paused coordinator):
 
-#### Output Tagging（强制）
-
-所有角色的输出（SendMessage、team_msg）必须带 `[role_name]` 标识前缀。
-
-#### Planner 边界
-
-| 允许 | 禁止 |
-|------|------|
-| 需求拆解 (issue 创建) | 直接编写/修改代码 |
-| 方案设计 (issue-plan-agent) | 调用 code-developer |
-| 冲突检查 (inline files_touched) | 运行测试 |
-| 创建 EXEC-* 任务 | git commit |
-| 监控进度 (消息总线) | |
-
-#### Executor 边界
-
-| 允许 | 禁止 |
-|------|------|
-| 处理 EXEC-* 前缀的任务 | 创建 issue |
-| 调用 code-developer 实现 | 修改 solution/queue |
-| 运行测试验证 | 为 planner 创建 PLAN-* 任务 |
-| git commit 提交 | 直接与用户交互 (AskUserQuestion) |
-| SendMessage 给 planner | |
-
-### Team Configuration
-
-| Key | Value |
-|-----|-------|
-| name | planex |
-| sessionDir | `.workflow/.team/PEX-{slug}-{date}/` |
-| artifactsDir | `.workflow/.team/PEX-{slug}-{date}/artifacts/` |
-| issueDataDir | `.workflow/issues/` |
-
-### Message Bus
-
-每次 SendMessage 前，先调用 `mcp__ccw-tools__team_msg` 记录：
-
-- 参数: operation="log", team=`<session-id>`, from=`<role>`, to=`<target-role>`, type=`<type>`, summary="[`<role>`] `<summary>`", ref=`<file_path>`
-- **注意**: `team` 必须是 **session ID** (如 `PEX-project-2026-02-27`), 不是 team name. 从任务描述的 `Session:` 字段提取.
-- **CLI fallback**: 当 MCP 不可用时 -> `ccw team log --team <session-id> --from <role> --to <target> --type <type> --summary "[<role>] ..." --json`
-
-**Message types by role**:
-
-| Role | Types |
-|------|-------|
-| planner | `wave_ready`, `issue_ready`, `all_planned`, `error` |
-| executor | `impl_complete`, `impl_failed`, `wave_done`, `error` |
-
-### Task Lifecycle (Both Roles)
-
-每个 worker 启动后执行相同的任务发现流程：
-
-1. 调用 `TaskList()` 获取所有任务
-2. 筛选: subject 匹配本角色前缀 + owner 是本角色 + status 为 pending + blockedBy 为空
-3. 无任务 -> idle 等待
-4. 有任务 -> `TaskGet` 获取详情 -> `TaskUpdate` 标记 in_progress
-5. Phase 2-4: Role-specific (see roles/{role}.md)
-6. Phase 5: Report + Loop
-
-**Resume Artifact Check** (防止恢复后重复产出):
-- 检查本任务的输出产物是否已存在
-- 产物完整 -> 跳到 Phase 5 报告完成
-- 产物不完整或不存在 -> 正常执行 Phase 2-4
+| Command | Action |
+|---------|--------|
+| `check` / `status` | Output execution status graph, no advancement |
+| `resume` / `continue` | Check worker states, advance next step |
+| `add <issue-ids or --text '...' or --plan path>` | Append new tasks to planner queue |
 
 ---
 
-## Wave Pipeline (逐 Issue 节拍)
+## Command Execution Protocol
 
-```
-Issue 1:  planner 规划 solution -> 写中间产物 -> 冲突检查 -> 创建 EXEC-* -> issue_ready
-                ↓ (executor 立即开始)
-Issue 2:  planner 规划 solution -> 写中间产物 -> 冲突检查 -> 创建 EXEC-* -> issue_ready
-                ↓ (executor 并行消费)
-Issue N:  ...
-Final:    planner 发送 all_planned -> executor 完成剩余 EXEC-* -> 结束
-```
+When coordinator needs to execute a command (dispatch, monitor):
 
-**节拍规则**:
-- planner 每完成一个 issue 的 solution 后，**立即**创建 EXEC-* 任务并发送 `issue_ready` 信号
-- solution 写入中间产物文件（`artifacts/solutions/{issueId}.json`），EXEC-* 任务包含 `solution_file` 路径
-- executor 从文件加载 solution（无需再调 `ccw issue solution`），fallback 兼容旧模式
-- planner 不等待 executor，持续推进下一个 issue
-- 当 planner 发送 `all_planned` 消息后，executor 完成所有剩余任务即可结束
+1. **Read the command file**: `roles/coordinator/commands/<command-name>.md`
+2. **Follow the workflow** defined in the command file (Phase 2-4 structure)
+3. **Commands are inline execution guides** - NOT separate agents or subprocesses
+4. **Execute synchronously** - complete the command workflow before proceeding
+
+---
+
+## Input Types
+
+支持 3 种输入方式:
+
+| 输入类型 | 格式 | 示例 |
+|----------|------|------|
+| Issue IDs | 直接传入 ID | `ISS-20260215-001 ISS-20260215-002` |
+| 需求文本 | `--text '...'` | `--text '实现用户认证模块'` |
+| Plan 文件 | `--plan path` | `--plan plan/2026-02-15-auth.md` |
 
 ## Execution Method Selection
 
-在编排模式或直接调用 executor 前，**必须先确定执行方式**。支持 3 种执行后端：
+支持 3 种执行后端：
 
 | Executor | 后端 | 适用场景 |
 |----------|------|----------|
@@ -159,251 +109,145 @@ Final:    planner 发送 all_planned -> executor 完成剩余 EXEC-* -> 结束
 
 ### Selection Decision Table
 
-| Condition | Execution Method | Code Review |
-|-----------|-----------------|-------------|
-| `--exec=agent` specified | Agent | Skip |
-| `--exec=codex` specified | Codex | Skip |
-| `--exec=gemini` specified | Gemini | Skip |
-| `-y` or `--yes` flag present | Auto (default Agent) | Skip |
-| No flags (interactive) | AskUserQuestion -> user choice | AskUserQuestion -> user choice |
-| Auto + task_count <= 3 | Agent | Skip |
-| Auto + task_count > 3 | Codex | Skip |
-
-### Interactive Prompt (no flags)
-
-当无 `-y`/`--yes` 且无 `--exec` 时，通过 AskUserQuestion 交互选择：
-
-- **执行方式选项**: Agent / Codex / Gemini / Auto
-- **代码审查选项**: Skip / Gemini Review / Codex Review / Agent Review
-
-### 通过 args 指定
-
-```bash
-# 显式指定
-Skill(skill="team-planex", args="--exec=codex ISS-xxx")
-Skill(skill="team-planex", args="--exec=agent --text '简单功能'")
-
-# Auto 模式（跳过交互，-y 或 --yes）
-Skill(skill="team-planex", args="-y --text '添加日志'")
-```
+| Condition | Execution Method |
+|-----------|-----------------|
+| `--exec=agent` specified | Agent |
+| `--exec=codex` specified | Codex |
+| `--exec=gemini` specified | Gemini |
+| `-y` or `--yes` flag present | Auto (default Agent) |
+| No flags (interactive) | AskUserQuestion -> user choice |
+| Auto + task_count <= 3 | Agent |
+| Auto + task_count > 3 | Codex |
 
 ---
 
-## Orchestration Mode
+## Coordinator Spawn Template
 
-当不带 `--role` 调用时，SKILL.md 进入轻量编排模式（无独立 coordinator 角色，SKILL.md 自身承担编排）。
+### v5 Worker Spawn (all roles)
 
-**Invocation**: `Skill(skill="team-planex", args="任务描述")`
-
-**Lifecycle**:
-
-```
-用户提供任务描述
-  -> SKILL.md 解析输入（Issue IDs / 需求文本 / Plan 文件）
-  -> 初始化 sessionDir + artifacts 目录
-  -> 执行方式选择（见 Execution Method Selection）
-  -> 创建 PLAN-001 任务（owner: planner）
-  -> Spawn planner agent (后台)
-  -> Spawn executor agent (后台)
-  -> 返回（planner lead 后续推进）
-```
-
-**User Commands** (唤醒 / 检查状态):
-
-| Command | Action |
-|---------|--------|
-| `check` / `status` | 输出执行状态图，不推进 |
-| `resume` / `continue` | 检查 worker 状态，推进下一步 |
-| `add <issue-ids or --text '...' or --plan path>` | 追加新任务到 planner 队列，不影响已有任务 |
-
-**`add` 命令处理逻辑**:
-
-1. 解析输入（Issue IDs / `--text` / `--plan`）
-2. 获取当前最大 PLAN-* 序号（`TaskList` 筛选 `PLAN-*` prefix），计算下一个序号 N
-3. `TaskCreate({ subject: "PLAN-00N: ...", owner: "planner", status: "pending" })`，description 写入新 issue IDs 或需求文本
-4. 若 planner 已发送 `all_planned`（检查 team_msg 日志），额外 `SendMessage` 通知 planner 有新任务，使其重新进入 Loop Check
-5. 若 executor 已退出等待，同样发送消息唤醒 executor 继续轮询 `EXEC-*` 任务
-
-### Coordinator Spawn Template
-
-SKILL.md 编排模式 spawn workers 时使用后台模式 (Spawn-and-Go):
-
-**Planner Spawn**:
+When coordinator spawns workers, use `team-worker` agent with role-spec path:
 
 ```
 Task({
-  subagent_type: "general-purpose",
-  description: "Spawn planner worker",
+  subagent_type: "team-worker",
+  description: "Spawn <role> worker",
   team_name: <team-name>,
-  name: "planner",
+  name: "<role>",
   run_in_background: true,
-  prompt: `你是 team "<team-name>" 的 PLANNER。
+  prompt: `## Role Assignment
+role: <role>
+role_spec: .claude/skills/team-planex/role-specs/<role>.md
+session: <session-folder>
+session_id: <session-id>
+team_name: <team-name>
+requirement: <task-description>
+inner_loop: <true|false>
+execution_method: <agent|codex|gemini>
 
-## 首要指令
-你的所有工作必须通过调用 Skill 获取角色定义后执行：
-Skill(skill="team-planex", args="--role=planner")
-
-当前输入: <planner-input>
-Session: <session-dir>
-
-## 执行配置
-executor 的执行方式: <execution-method>
-创建 EXEC-* 任务时，description 中包含:
-  execution_method: <method>
-  code_review: <review-tool>
-
-## 中间产物（必须）
-每个 issue 的 solution 写入: <session-dir>/artifacts/solutions/{issueId}.json
-EXEC-* 任务 description 必须包含 solution_file 字段指向该文件
-每完成一个 issue 立即发送 issue_ready 消息并创建 EXEC-* 任务
-
-## 角色准则
-- 只处理 PLAN-* 任务，不执行其他角色工作
-- 所有输出带 [planner] 标识前缀
-- 仅与 coordinator 通信
-- 不使用 TaskCreate 为其他角色创建任务（EXEC-* 除外）
-- 每次 SendMessage 前先调用 mcp__ccw-tools__team_msg 记录
-
-## 工作流程
-1. 调用 Skill -> 获取角色定义和执行逻辑
-2. 按 role.md 5-Phase 流程执行
-3. team_msg + SendMessage 结果给 coordinator
-4. TaskUpdate completed -> 检查下一个任务`
+Read role_spec file to load Phase 2-4 domain instructions.
+Execute built-in Phase 1 (task discovery) -> role-spec Phase 2-4 -> built-in Phase 5 (report).`
 })
 ```
 
-**Executor Spawn**:
-
-```
-Task({
-  subagent_type: "general-purpose",
-  description: "Spawn executor worker",
-  team_name: <team-name>,
-  name: "executor",
-  run_in_background: true,
-  prompt: `你是 team "<team-name>" 的 EXECUTOR。
-
-## 首要指令
-你的所有工作必须通过调用 Skill 获取角色定义后执行：
-Skill(skill="team-planex", args="--role=executor")
-
-## 执行配置
-默认执行方式: <execution-method>
-代码审查: <review-tool>
-（每个 EXEC-* 任务 description 中可能包含 execution_method 覆盖）
-
-## Solution 加载
-优先从 EXEC-* 任务 description 中的 solution_file 路径读取 solution JSON 文件
-无 solution_file 时 fallback 到 ccw issue solution 命令
-
-## 角色准则
-- 只处理 EXEC-* 任务，不执行其他角色工作
-- 所有输出带 [executor] 标识前缀
-- 根据 execution_method 选择执行后端（Agent/Codex/Gemini）
-- 仅与 coordinator 通信
-- 每次 SendMessage 前先调用 mcp__ccw-tools__team_msg 记录
-
-## 工作流程
-1. 调用 Skill -> 获取角色定义和执行逻辑
-2. 按 role.md 5-Phase 流程执行
-3. team_msg + SendMessage 结果给 coordinator
-4. TaskUpdate completed -> 检查下一个任务`
-})
-```
+**Inner Loop roles** (planner, executor): Set `inner_loop: true`. The team-worker agent handles the loop internally.
 
 ---
 
-## Cadence Control
+## Pipeline Definitions
 
-**节拍模型**: Wave beat -- planner 持续推进，executor 并行消费。每个 wave = planner 完成一个 issue -> executor 开始实现。
+### Pipeline Diagram
 
 ```
-Wave Beat Cycle (逐 Issue 节拍)
-===================================================================
-  Event                    SKILL.md (编排)         Workers
--------------------------------------------------------------------
-  用户调用 ---------> ┌─ 解析输入 ─────────┐
-                      │  初始化 session      │
-                      │  选择执行方式         │
-                      ├─ 创建 PLAN-001 ─────┤
-                      │  spawn planner ──────┼──> [Planner] Phase 1-5
-                      │  spawn executor ─────┼──> [Executor] Phase 1 (idle)
-                      └─ 返回 (编排结束) ───┘         │
-                                                       │
-  Wave 1:                                    Planner: issue-1 solution
-                                             -> 写产物 -> 创建 EXEC-001
-                                             -> issue_ready ---------> Executor 开始 EXEC-001
-  Wave 2:                                    Planner: issue-2 solution
-                                             -> 写产物 -> 创建 EXEC-002
-                                             -> issue_ready ---------> Executor 并行消费
+Issue-based beat pipeline (逐 Issue 节拍)
+═══════════════════════════════════════════════════
+  PLAN-001 ──> [planner] issue-1 solution → EXEC-001
+                         issue-2 solution → EXEC-002
+                         ...
+                         issue-N solution → EXEC-00N
+                         all_planned signal
+
+  EXEC-001 ──> [executor] implement issue-1
+  EXEC-002 ──> [executor] implement issue-2
   ...
-  Wave N:                                    Planner: all_planned
-                                             Executor: 完成剩余 EXEC-*
-===================================================================
+  EXEC-00N ──> [executor] implement issue-N
+═══════════════════════════════════════════════════
 ```
 
-**Pipeline 节拍视图**:
+### Cadence Control
+
+**Beat model**: Event-driven Spawn-and-Stop. Each beat = coordinator wake -> process callback -> spawn next -> STOP.
 
 ```
-Wave pipeline (planner lead, executor follows)
-──────────────────────────────────────────────────────────
-Wave   1        2        3       ...      N       Final
-       │        │        │                │         │
-       P:iss-1  P:iss-2  P:iss-3         P:iss-N   P:all_planned
-       ↓        ↓        ↓               ↓         ↓
-       E:exec1  E:exec2  E:exec3         E:execN   E:finish
-                │        │
-         (并行消费，executor 不等 planner 全部完成)
-
-P=planner  E=executor
+Beat Cycle (Coordinator Spawn-and-Stop)
+======================================================================
+  Event                   Coordinator              Workers
+----------------------------------------------------------------------
+  用户调用 ----------> ┌─ Phase 1-3 ──────────┐
+                       │  解析输入              │
+                       │  TeamCreate            │
+                       │  创建 PLAN-001         │
+                       ├─ Phase 4 ─────────────┤
+                       │  spawn planner ────────┼──> [planner] Phase 1-5
+                       └─ STOP (idle) ──────────┘         │
+                                                          │
+  callback <─ planner issue_ready ────────────────────────┘
+         ┌─ monitor.handleCallback ─┐
+         │  检查新 EXEC-* 任务       │
+         │  spawn executor ─────────┼──> [executor] Phase 1-5
+         └─ STOP (idle) ───────────┘         │
+                                              │
+  callback <─ executor impl_complete ────────┘
+         ┌─ monitor.handleCallback ─┐
+         │  标记完成                  │
+         │  检查下一个 ready task     │
+         └─ spawn/STOP ────────────┘
+======================================================================
 ```
 
-**检查点 (Checkpoint)**:
+**Checkpoints**:
 
 | 触发条件 | 位置 | 行为 |
 |----------|------|------|
 | Planner 全部完成 | all_planned 信号 | Executor 完成剩余 EXEC-* 后结束 |
-| Pipeline 停滞 | 无 ready + 无 running | Planner 检查并 escalate to user |
-| Executor 阻塞 | Executor blocked > 2 tasks | Planner escalate to user |
+| Pipeline 停滞 | 无 ready + 无 running | Coordinator escalate to user |
+| Executor 阻塞 | blocked > 2 tasks | Coordinator escalate to user |
 
-**Stall 检测**:
-
-| 检查项 | 条件 | 处理 |
-|--------|------|------|
-| Executor 无响应 | in_progress EXEC-* 无回调 | 报告等待中的任务列表 |
-| Pipeline 死锁 | 无 ready + 无 running + 有 pending | 检查 blockedBy 依赖链 |
-| Planner 规划失败 | issue planning error | Retry once, then skip to next issue |
-
----
-
-## Task Metadata Registry
+### Task Metadata Registry
 
 | Task ID | Role | Phase | Dependencies | Description |
 |---------|------|-------|-------------|-------------|
 | PLAN-001 | planner | planning | (none) | 初始规划：需求拆解、issue 创建、方案设计 |
-| EXEC-001 | executor | execution | PLAN-001 (implicit via issue_ready) | 第一个 issue 的代码实现 |
-| EXEC-002 | executor | execution | (planner issue_ready) | 第二个 issue 的代码实现 |
-| EXEC-N | executor | execution | (planner issue_ready) | 第 N 个 issue 的代码实现 |
+| EXEC-001 | executor | execution | (created by planner at runtime) | 第一个 issue 的代码实现 |
+| EXEC-N | executor | execution | (created by planner at runtime) | 第 N 个 issue 的代码实现 |
 
 > 注: EXEC-* 任务由 planner 在运行时逐个创建（逐 Issue 节拍），不预先定义完整任务链。
 
 ---
 
-## Wisdom Accumulation (所有角色)
+## Completion Action
 
-跨任务知识积累。SKILL.md 编排模式在 session 初始化时创建 `wisdom/` 目录。
+When the pipeline completes (all tasks done, coordinator Phase 5):
 
-**目录**:
 ```
-<session-folder>/wisdom/
-├── learnings.md      # 模式和洞察
-├── decisions.md      # 架构和设计决策
-├── conventions.md    # 代码库约定
-└── issues.md         # 已知风险和问题
+AskUserQuestion({
+  questions: [{
+    question: "Team pipeline complete. What would you like to do?",
+    header: "Completion",
+    multiSelect: false,
+    options: [
+      { label: "Archive & Clean (Recommended)", description: "Archive session, clean up tasks and team resources" },
+      { label: "Keep Active", description: "Keep session active for follow-up work or inspection" },
+      { label: "Export Results", description: "Export deliverables to a specified location, then clean" }
+    ]
+  }]
+})
 ```
 
-**Worker 加载** (Phase 2): 从 task description 提取 `Session: <path>`, 读取 wisdom 目录下各文件。
-**Worker 贡献** (Phase 4/5): 将本任务发现写入对应 wisdom 文件。
+| Choice | Action |
+|--------|--------|
+| Archive & Clean | Update session status="completed" -> TeamDelete -> output final summary |
+| Keep Active | Update session status="paused" -> output resume instructions |
+| Export Results | AskUserQuestion for target path -> copy deliverables -> Archive & Clean |
 
 ---
 
@@ -426,14 +270,30 @@ P=planner  E=executor
 
 ---
 
+## Message Bus
+
+每次 SendMessage 前，先调用 `mcp__ccw-tools__team_msg` 记录：
+
+- 参数: operation="log", team=`<session-id>`, from=`<role>`, to=`<target-role>`, type=`<type>`, summary="[`<role>`] `<summary>`"
+- **注意**: `team` 必须是 **session ID** (如 `PEX-project-2026-02-27`), 不是 team name.
+
+**Message types by role**:
+
+| Role | Types |
+|------|-------|
+| planner | `issue_ready`, `all_planned`, `error` |
+| executor | `impl_complete`, `impl_failed`, `error` |
+
+---
+
 ## Error Handling
 
 | Scenario | Resolution |
 |----------|------------|
-| Unknown --role value | Error with available role list: planner, executor |
-| Missing --role arg | Enter orchestration mode (SKILL.md as lightweight coordinator) |
-| Role file not found | Error with expected path (roles/{name}.md) |
-| Planner issue planning failure | Retry once, then report error and skip to next issue |
-| Executor impl failure | Report to planner, continue with next EXEC-* task |
-| No EXEC-* tasks yet | Executor idles, polls for new tasks |
-| Pipeline stall | Planner monitors -- if executor blocked > 2 tasks, escalate to user |
+| Role spec file not found | Error with expected path (role-specs/<name>.md) |
+| Command file not found | Fallback to inline execution in coordinator role.md |
+| team-worker agent unavailable | Error: requires .claude/agents/team-worker.md |
+| Planner issue planning failure | Retry once, then skip to next issue |
+| Executor impl failure | Report to coordinator, continue with next EXEC-* task |
+| Pipeline stall | Coordinator monitors, escalate to user |
+| Completion action timeout | Default to Keep Active |

.claude/skills/team-planex/role-specs/executor.md

@@ -0,0 +1,102 @@
+---
+prefix: EXEC
+inner_loop: true
+message_types:
+  success: impl_complete
+  error: impl_failed
+---
+
+# Executor
+
+Single-issue implementation agent. Loads solution from artifact file, routes to execution backend (Agent/Codex/Gemini), verifies with tests, commits, and reports completion.
+
+## Phase 2: Task & Solution Loading
+
+| Input | Source | Required |
+|-------|--------|----------|
+| Issue ID | Task description `Issue ID:` field | Yes |
+| Solution file | Task description `Solution file:` field | Yes |
+| Session folder | Task description `Session:` field | Yes |
+| Execution method | Task description `Execution method:` field | Yes |
+| Wisdom | `<session>/wisdom/` | No |
+
+1. Extract issue ID, solution file path, session folder, execution method
+2. Load solution JSON from file (file-first)
+3. If file not found -> fallback: `ccw issue solution <issueId> --json`
+4. Load wisdom files for conventions and patterns
+5. Verify solution has required fields: title, tasks
+
+## Phase 3: Implementation
+
+### Backend Selection
+
+| Method | Backend | Agent Type |
+|--------|---------|------------|
+| `agent` | code-developer subagent | Inline delegation |
+| `codex` | `ccw cli --tool codex --mode write` | Background CLI |
+| `gemini` | `ccw cli --tool gemini --mode write` | Background CLI |
+
+### Agent Backend
+
+```
+Task({
+  subagent_type: "code-developer",
+  description: "Implement <issue-title>",
+  prompt: `Issue: <issueId>
+Title: <solution.title>
+Solution: <solution JSON>
+Implement all tasks from the solution plan.`,
+  run_in_background: false
+})
+```
+
+### CLI Backend (Codex/Gemini)
+
+```bash
+ccw cli -p "Issue: <issueId>
+Title: <solution.title>
+Solution Plan: <solution JSON>
+Implement all tasks. Follow existing patterns. Run tests." \
+  --tool <codex|gemini> --mode write
+```
+
+Wait for CLI completion before proceeding.
+
+## Phase 4: Verification + Commit
+
+### Test Verification
+
+| Check | Method | Pass Criteria |
+|-------|--------|---------------|
+| Tests | Detect and run project test command | All pass |
+| Syntax | IDE diagnostics or `tsc --noEmit` | No errors |
+
+If tests fail: retry implementation once, then report `impl_failed`.
+
+### Commit
+
+```bash
+git add -A
+git commit -m "feat(<issueId>): <solution.title>"
+```
+
+### Update Issue Status
+
+```bash
+ccw issue update <issueId> --status completed
+```
+
+### Report
+
+Send `impl_complete` message to coordinator via team_msg + SendMessage:
+- summary: `[executor] Implemented <issueId>: <title>`
+
+## Boundaries
+
+| Allowed | Prohibited |
+|---------|-----------|
+| Load solution from file | Create or modify issues |
+| Implement via Agent/Codex/Gemini | Modify solution artifacts |
+| Run tests | Spawn additional agents |
+| git commit | Direct user interaction |
+| Update issue status | Create tasks for other roles |

.claude/skills/team-planex/role-specs/planner.md

@@ -0,0 +1,111 @@
+---
+prefix: PLAN
+inner_loop: true
+subagents: [issue-plan-agent]
+message_types:
+  success: issue_ready
+  error: error
+---
+
+# Planner
+
+Requirement decomposition → issue creation → solution design → EXEC-* task creation. Processes issues one at a time, creating executor tasks as solutions are completed.
+
+## Phase 2: Context Loading
+
+| Input | Source | Required |
+|-------|--------|----------|
+| Input type + raw input | Task description | Yes |
+| Session folder | Task description `Session:` field | Yes |
+| Execution method | Task description `Execution method:` field | Yes |
+| Wisdom | `<session>/wisdom/` | No |
+
+1. Extract session path, input type, raw input, execution method from task description
+2. Load wisdom files if available
+3. Parse input to determine issue list:
+
+| Detection | Condition | Action |
+|-----------|-----------|--------|
+| Issue IDs | `ISS-\d{8}-\d{6}` pattern | Use directly |
+| `--text '...'` | Flag in input | Create issue(s) via `ccw issue create` |
+| `--plan <path>` | Flag in input | Read file, parse phases, batch create issues |
+
+## Phase 3: Issue Processing Loop
+
+For each issue, execute in sequence:
+
+### 3a. Generate Solution
+
+Delegate to `issue-plan-agent` subagent:
+
+```
+Task({
+  subagent_type: "issue-plan-agent",
+  description: "Plan issue <issueId>",
+  prompt: `issue_ids: ["<issueId>"]
+project_root: "<project-root>"
+Generate solution for this issue. Auto-bind single solution.`,
+  run_in_background: false
+})
+```
+
+### 3b. Write Solution Artifact
+
+Write solution JSON to: `<session>/artifacts/solutions/<issueId>.json`
+
+```json
+{
+  "session_id": "<session-id>",
+  "issue_id": "<issueId>",
+  "solution": <solution-from-agent>,
+  "planned_at": "<ISO timestamp>"
+}
+```
+
+### 3c. Check Conflicts
+
+Extract `files_touched` from solution. Compare against prior solutions in session.
+Overlapping files -> log warning to `wisdom/issues.md`, continue.
+
+### 3d. Create EXEC-* Task
+
+```
+TaskCreate({
+  subject: "EXEC-00N: Implement <issue-title>",
+  description: `Implement solution for issue <issueId>.
+
+Issue ID: <issueId>
+Solution file: <session>/artifacts/solutions/<issueId>.json
+Session: <session>
+Execution method: <method>
+
+InnerLoop: true`,
+  activeForm: "Implementing <issue-title>"
+})
+```
+
+### 3e. Signal issue_ready
+
+Send message via team_msg + SendMessage to coordinator:
+- type: `issue_ready`
+- summary: `[planner] Solution ready for <issueId>`
+
+### 3f. Continue Loop
+
+Process next issue. Do NOT wait for executor.
+
+## Phase 4: Completion Signal
+
+After all issues processed:
+1. Send `all_planned` message to coordinator via team_msg + SendMessage
+2. Summary: total issues planned, EXEC-* tasks created
+
+## Boundaries
+
+| Allowed | Prohibited |
+|---------|-----------|
+| Parse input, create issues | Write/modify business code |
+| Generate solutions (issue-plan-agent) | Run tests |
+| Write solution artifacts | git commit |
+| Create EXEC-* tasks | Call code-developer |
+| Conflict checking | Direct user interaction |

.claude/skills/team-planex/roles/coordinator/commands/dispatch.md

@@ -0,0 +1,87 @@
+# Command: dispatch
+
+## Purpose
+
+Create the initial task chain for team-planex pipeline. Creates PLAN-001 for planner. EXEC-* tasks are NOT pre-created — planner creates them at runtime per issue.
+
+## Phase 2: Context Loading
+
+| Input | Source | Required |
+|-------|--------|----------|
+| Input type | Phase 1 requirements | Yes |
+| Raw input | Phase 1 requirements | Yes |
+| Session folder | Phase 2 session init | Yes |
+| Execution method | Phase 1 requirements | Yes |
+
+## Phase 3: Task Chain Creation
+
+### Task Creation
+
+Create a single PLAN-001 task for the planner:
+
+```
+TaskCreate({
+  subject: "PLAN-001: Requirement decomposition and solution design",
+  description: `Decompose requirements into issues and generate solutions.
+
+Input type: <issues|text|plan>
+Input: <raw-input>
+Session: <session-folder>
+Execution method: <agent|codex|gemini>
+
+## Instructions
+1. Parse input to get issue list
+2. For each issue: call issue-plan-agent → write solution artifact
+3. After each solution: create EXEC-* task (owner: executor) with solution_file path
+4. After all issues: send all_planned signal
+
+InnerLoop: true`,
+  activeForm: "Planning requirements"
+})
+```
+
+### EXEC-* Task Template (for planner reference)
+
+Planner creates EXEC-* tasks at runtime using this template:
+
+```
+TaskCreate({
+  subject: "EXEC-00N: Implement <issue-title>",
+  description: `Implement solution for issue <issueId>.
+
+Issue ID: <issueId>
+Solution file: <session-folder>/artifacts/solutions/<issueId>.json
+Session: <session-folder>
+Execution method: <agent|codex|gemini>
+
+InnerLoop: true`,
+  activeForm: "Implementing <issue-title>"
+})
+```
+
+### Add Command Task Template
+
+When coordinator handles `add` command, create additional PLAN tasks:
+
+```
+TaskCreate({
+  subject: "PLAN-00N: Additional requirement decomposition",
+  description: `Additional requirements to decompose.
+
+Input type: <issues|text|plan>
+Input: <new-input>
+Session: <session-folder>
+Execution method: <execution-method>
+
+InnerLoop: true`,
+  activeForm: "Planning additional requirements"
+})
+```
+
+## Phase 4: Validation
+
+| Check | Criteria |
+|-------|----------|
+| PLAN-001 created | TaskList shows PLAN-001 |
+| Description complete | Contains Input, Session, Execution method |
+| No orphans | All tasks have valid owner |

.claude/skills/team-planex/roles/coordinator/commands/monitor.md

@@ -0,0 +1,163 @@
+# Command: monitor
+
+## Purpose
+
+Event-driven pipeline coordination with Spawn-and-Stop pattern. Three wake-up sources: worker callbacks, user `check`, user `resume`.
+
+## Constants
+
+| Constant | Value | Description |
+|----------|-------|-------------|
+| SPAWN_MODE | background | All workers spawned via `Task(run_in_background: true)` |
+| ONE_STEP_PER_INVOCATION | true | Coordinator does one operation then STOPS |
+| WORKER_AGENT | team-worker | All workers are team-worker agents |
+
+## Phase 2: Context Loading
+
+| Input | Source | Required |
+|-------|--------|----------|
+| Session file | `<session-folder>/team-session.json` | Yes |
+| Task list | `TaskList()` | Yes |
+| Active workers | session.active_workers[] | Yes |
+
+## Phase 3: Handler Routing
+
+### Wake-up Source Detection
+
+| Priority | Condition | Handler |
+|----------|-----------|---------|
+| 1 | Message contains `[planner]` or `[executor]` tag | handleCallback |
+| 2 | Contains "check" or "status" | handleCheck |
+| 3 | Contains "resume", "continue", or "next" | handleResume |
+| 4 | None of the above (initial spawn) | handleSpawnNext |
+
+---
+
+### Handler: handleCallback
+
+```
+Receive callback from [<role>]
+  +- Match role: planner or executor
+  +- Progress update (not final)?
+  |   +- YES -> Update session -> STOP
+  +- Task status = completed?
+  |   +- YES -> remove from active_workers -> update session
+  |   |   +- role = planner?
+  |   |   |   +- Check for new EXEC-* tasks (planner creates them)
+  |   |   |   +- -> handleSpawnNext (spawn executor for new EXEC-* tasks)
+  |   |   +- role = executor?
+  |   |       +- Mark issue done
+  |   |       +- -> handleSpawnNext (check for more EXEC-* tasks)
+  |   +- NO -> progress message -> STOP
+  +- No matching worker found
+      +- Scan all active workers for completed tasks
+      +- Found completed -> process -> handleSpawnNext
+      +- None completed -> STOP
+```
+
+---
+
+### Handler: handleCheck
+
+Read-only status report. No advancement.
+
+```
+[coordinator] PlanEx Pipeline Status
+[coordinator] Progress: <completed>/<total> (<percent>%)
+
+[coordinator] Task Graph:
+  PLAN-001: <status-icon> <summary>
+  EXEC-001: <status-icon> <issue-title>
+  EXEC-002: <status-icon> <issue-title>
+  ...
+
+  done=completed  >>>=running  o=pending
+
+[coordinator] Active Workers:
+  > <subject> (<role>) - running <elapsed>
+
+[coordinator] Ready to spawn: <subjects>
+[coordinator] Commands: 'resume' to advance | 'check' to refresh
+```
+
+Then STOP.
+
+---
+
+### Handler: handleResume
+
+```
+Load active_workers
+  +- No active workers -> handleSpawnNext
+  +- Has active workers -> check each:
+      +- completed -> mark done, log
+      +- in_progress -> still running
+      +- other -> worker failure -> reset to pending
+      After:
+        +- Some completed -> handleSpawnNext
+        +- All running -> report status -> STOP
+        +- All failed -> handleSpawnNext (retry)
+```
+
+---
+
+### Handler: handleSpawnNext
+
+```
+Collect task states from TaskList()
+  +- Filter tasks: PLAN-* and EXEC-* prefixes
+  +- readySubjects: pending + not blocked (no blockedBy or all blockedBy completed)
+  +- NONE ready + work in progress -> report waiting -> STOP
+  +- NONE ready + nothing running -> PIPELINE_COMPLETE -> Phase 5
+  +- HAS ready tasks -> for each:
+      +- Inner Loop role AND already has active_worker for that role?
+      |   +- YES -> SKIP spawn (existing worker picks up via inner loop)
+      |   +- NO -> spawn below
+      +- Determine role from task prefix:
+      |   +- PLAN-* -> planner
+      |   +- EXEC-* -> executor
+      +- Spawn team-worker:
+         Task({
+           subagent_type: "team-worker",
+           description: "Spawn <role> worker for <subject>",
+           team_name: <team-name>,
+           name: "<role>",
+           run_in_background: true,
+           prompt: `## Role Assignment
+role: <role>
+role_spec: .claude/skills/team-planex/role-specs/<role>.md
+session: <session-folder>
+session_id: <session-id>
+team_name: <team-name>
+requirement: <task-description>
+inner_loop: true
+execution_method: <method>`
+         })
+      +- Add to session.active_workers
+      Update session -> output summary -> STOP
+```
+
+---
+
+## Phase 4: Validation
+
+| Check | Criteria |
+|-------|----------|
+| Session state consistent | active_workers matches in_progress tasks |
+| No orphaned tasks | Every in_progress has active_worker |
+| Pipeline completeness | All expected EXEC-* tasks accounted for |
+
+## Worker Failure Handling
+
+1. Reset task -> pending via TaskUpdate
+2. Log via team_msg (type: error)
+3. Report to user
+
+## Error Handling
+
+| Scenario | Resolution |
+|----------|------------|
+| Session file not found | Error, suggest re-initialization |
+| Unknown role callback | Log, scan for other completions |
+| All workers running on resume | Report status, suggest check later |
+| Pipeline stall (no ready + no running + has pending) | Check blockedBy chains, report |

.claude/skills/team-planex/roles/coordinator/role.md

@@ -0,0 +1,154 @@
+# Coordinator Role
+
+Orchestrate the team-planex pipeline: parse input, create team, dispatch tasks, monitor progress via Spawn-and-Stop beats. Uses **team-worker agent** for all worker spawns.
+
+## Identity
+
+- **Name**: `coordinator` | **Tag**: `[coordinator]`
+- **Responsibility**: Parse input -> Create team -> Dispatch PLAN-001 -> Spawn planner -> Monitor callbacks -> Spawn executors -> Report
+
+## Boundaries
+
+### MUST
+- Parse user input (Issue IDs / --text / --plan) and determine execution method
+- Create team and initialize session directory
+- Dispatch tasks via `commands/dispatch.md`
+- Monitor progress via `commands/monitor.md` with Spawn-and-Stop pattern
+- Maintain session state (team-session.json)
+
+### MUST NOT
+- Execute planning or implementation work directly (delegate to workers)
+- Modify solution artifacts or code (workers own their deliverables)
+- Call implementation subagents (code-developer, etc.) directly
+- Skip dependency validation when creating task chains
+
+---
+
+## Command Execution Protocol
+
+When coordinator needs to execute a command (dispatch, monitor):
+
+1. **Read the command file**: `roles/coordinator/commands/<command-name>.md`
+2. **Follow the workflow** defined in the command file
+3. **Commands are inline execution guides** - NOT separate agents
+4. **Execute synchronously** - complete the command workflow before proceeding
+
+---
+
+## Entry Router
+
+When coordinator is invoked, detect invocation type:
+
+| Detection | Condition | Handler |
+|-----------|-----------|---------|
+| Worker callback | Message contains `[planner]` or `[executor]` tag | -> handleCallback (monitor.md) |
+| Status check | Arguments contain "check" or "status" | -> handleCheck (monitor.md) |
+| Manual resume | Arguments contain "resume" or "continue" | -> handleResume (monitor.md) |
+| Add tasks | Arguments contain "add" | -> handleAdd |
+| Interrupted session | Active/paused session exists in `.workflow/.team/PEX-*` | -> Phase 0 |
+| New session | None of above | -> Phase 1 |
+
+For callback/check/resume: load `commands/monitor.md` and execute the appropriate handler, then STOP.
+
+### handleAdd
+
+1. Parse new input (Issue IDs / `--text` / `--plan`)
+2. Get current max PLAN-* sequence from `TaskList`
+3. `TaskCreate` new PLAN-00N task (owner: planner)
+4. If planner already sent `all_planned` (check team_msg) -> `SendMessage` to planner to re-enter loop
+5. STOP
+
+---
+
+## Phase 0: Session Resume Check
+
+1. Scan `.workflow/.team/PEX-*/team-session.json` for sessions with status "active" or "paused"
+2. No sessions found -> proceed to Phase 1
+3. Single session found -> resume (Session Reconciliation)
+4. Multiple sessions -> AskUserQuestion for selection
+
+**Session Reconciliation**:
+1. Audit TaskList -> reconcile session state vs task status
+2. Reset in_progress tasks -> pending (they were interrupted)
+3. Rebuild team if needed (TeamCreate + spawn needed workers)
+4. Kick first executable task -> Phase 4
+
+---
+
+## Phase 1: Input Parsing + Execution Method
+
+1. **Parse arguments**: Extract input type (Issue IDs / --text / --plan) and optional flags (--exec, -y)
+
+2. **Determine execution method** (see SKILL.md Selection Decision Table):
+   - Explicit `--exec` flag -> use specified method
+   - `-y` / `--yes` flag -> Auto mode
+   - No flags -> AskUserQuestion for method choice
+
+3. **Store requirements**: input_type, raw_input, execution_method
+
+---
+
+## Phase 2: Create Team + Initialize Session
+
+1. Generate session ID: `PEX-<slug>-<date>`
+2. Create session folder: `.workflow/.team/<session-id>/`
+3. Create subdirectories: `artifacts/solutions/`, `wisdom/`
+4. Call `TeamCreate` with team name (default: "planex")
+5. Initialize wisdom files (learnings.md, decisions.md, conventions.md, issues.md)
+6. Write team-session.json:
+
+```
+{
+  session_id: "<session-id>",
+  input_type: "<issues|text|plan>",
+  input: "<raw-input>",
+  execution_method: "<agent|codex|gemini>",
+  status: "active",
+  active_workers: [],
+  started_at: "<ISO timestamp>"
+}
+```
+
+---
+
+## Phase 3: Create Task Chain
+
+Delegate to `commands/dispatch.md`:
+
+1. Read `roles/coordinator/commands/dispatch.md`
+2. Execute its workflow to create PLAN-001 task
+3. PLAN-001 contains input info + execution method in description
+
+---
+
+## Phase 4: Spawn-and-Stop
+
+1. Load `commands/monitor.md`
+2. Execute `handleSpawnNext` to find ready tasks and spawn planner worker
+3. Output status summary
+4. **STOP** (idle, wait for worker callback)
+
+**ONE_STEP_PER_INVOCATION**: true — coordinator does one operation per wake-up, then STOPS.
+
+---
+
+## Phase 5: Report + Completion Action
+
+When all tasks are complete (monitor.md detects PIPELINE_COMPLETE):
+
+1. Load session state -> count completed tasks, duration
+2. List deliverables with output paths
+3. Update session status -> "completed"
+4. Execute Completion Action (see SKILL.md)
+
+---
+
+## Error Handling
+
+| Error | Resolution |
+|-------|------------|
+| Session file not found | Error, suggest re-initialization |
+| Unknown worker callback | Log, scan for other completions |
+| Pipeline stall | Check missing tasks, report to user |
+| Worker crash | Reset task to pending, re-spawn on next beat |
+| All workers running on resume | Report status, suggest check later |

.claude/skills/team-planex/roles/executor.md

@@ -1,356 +0,0 @@
-# Executor Role
-
-Load solution -> Route to backend (Agent/Codex/Gemini) based on execution_method -> Test verification -> Commit. Supports multiple CLI execution backends. Execution method is determined before skill invocation (see SKILL.md Execution Method Selection).
-
-## Identity
-
-- **Name**: `executor` | **Tag**: `[executor]`
-- **Task Prefix**: `EXEC-*`
-- **Responsibility**: Code implementation (solution -> route to backend -> test -> commit)
-
-## Boundaries
-
-### MUST
-
-- Only process `EXEC-*` prefixed tasks
-- All output (SendMessage, team_msg, logs) must carry `[executor]` identifier
-- Select execution backend based on `execution_method` field in EXEC-* task
-- Notify planner after each issue completes
-- Continuously poll for new EXEC-* tasks (planner may create new waves anytime)
-
-### MUST NOT
-
-- Create issues (planner responsibility)
-- Modify solution or queue (planner responsibility)
-- Call issue-plan-agent or issue-queue-agent
-- Interact directly with user (AskUserQuestion)
-- Create PLAN-* tasks for planner
-
----
-
-## Toolbox
-
-### Execution Backends
-
-| Backend | Tool | Invocation | Mode |
-|---------|------|------------|------|
-| `agent` | code-developer subagent | `Task({ subagent_type: "code-developer" })` | Synchronous |
-| `codex` | Codex CLI | `ccw cli --tool codex --mode write` | Background |
-| `gemini` | Gemini CLI | `ccw cli --tool gemini --mode write` | Background |
-
-### Direct Capabilities
-
-| Tool | Purpose |
-|------|---------|
-| `Read` | Read solution plan and queue files |
-| `Write` | Write implementation artifacts |
-| `Edit` | Edit source code |
-| `Bash` | Run tests, git operations, CLI calls |
-
-### CLI Capabilities
-
-| CLI Command | Purpose |
-|-------------|---------|
-| `ccw issue status <id> --json` | Check issue status |
-| `ccw issue solution <id> --json` | Load single issue's bound solution (requires issue ID) |
-| `ccw issue update <id> --status executing` | Update issue status to executing |
-| `ccw issue update <id> --status completed` | Mark issue as completed |
-
----
-
-## Message Types
-
-| Type | Direction | Trigger | Description |
-|------|-----------|---------|-------------|
-| `impl_complete` | executor -> planner | Implementation and tests pass | Single issue implementation complete |
-| `impl_failed` | executor -> planner | Implementation failed after retries | Implementation failure |
-| `wave_done` | executor -> planner | All EXEC tasks in a wave completed | Entire wave complete |
-| `error` | executor -> planner | Blocking error | Execution error |
-
-## Message Bus
-
-Before every SendMessage, log via `mcp__ccw-tools__team_msg`:
-
-**NOTE**: `team` must be **session ID** (e.g., `PEX-project-2026-02-27`), NOT team name. Extract from `Session:` field in task description.
-
-```
-mcp__ccw-tools__team_msg({
-  operation: "log",
-  team: <session-id>,  // e.g., "PEX-project-2026-02-27", NOT "planex"
-  from: "executor",
-  to: "planner",
-  type: <message-type>,
-  summary: "[executor] <task-prefix> complete: <task-subject>",
-  ref: <artifact-path>
-})
-```
-
-**CLI fallback** (when MCP unavailable):
-
-```
-Bash("ccw team log --team <session-id> --from executor --to planner --type <message-type> --summary \"[executor] <task-prefix> complete\" --ref <artifact-path> --json")
-```
-
----
-
-## Execution (5-Phase)
-
-### Phase 1: Task Discovery
-
-> See SKILL.md Shared Infrastructure -> Worker Phase 1: Task Discovery
-
-Standard task discovery flow: TaskList -> filter by prefix `EXEC-*` + owner match + pending + unblocked -> TaskGet -> TaskUpdate in_progress.
-
-### Phase 2: Load Solution & Resolve Executor
-
-**Issue ID Extraction**:
-
-Extract issue ID from task description using pattern `ISS-\d{8}-\d{6}`.
-
-If no issue ID found:
-1. Log error via team_msg
-2. SendMessage error to planner
-3. TaskUpdate completed
-4. Return to idle
-
-**Solution Loading (Dual Mode)**:
-
-| Mode | Condition | Action |
-|------|-----------|--------|
-| File-first | Task description contains `solution_file: <path>` | Read JSON file, extract solution.bound |
-| CLI fallback | No solution_file field | Call `ccw issue solution <issueId> --json` |
-
-If no bound solution found:
-1. Log error via team_msg
-2. SendMessage error to planner
-3. TaskUpdate completed
-4. Return to idle
-
-**Execution Method Resolution**:
-
-| Condition | Executor |
-|-----------|----------|
-| `execution_method: Agent` in task description | agent |
-| `execution_method: Codex` in task description | codex |
-| `execution_method: Gemini` in task description | gemini |
-| `execution_method: Auto` + task_count <= 3 | agent |
-| `execution_method: Auto` + task_count > 3 | codex |
-| Unknown or missing | agent (with warning) |
-
-**Code Review Resolution**:
-
-Extract `code_review` from task description. Values: Skip | Gemini Review | Codex Review | Agent Review. Default: Skip.
-
-**Issue Status Update**:
-
-```
-Bash("ccw issue update <issueId> --status executing")
-```
-
-### Phase 3: Implementation (Multi-Backend Routing)
-
-Route to execution backend based on resolved executor.
-
-#### Option A: Agent Execution
-
-**When**: executor === 'agent' (simple tasks, task_count <= 3)
-
-**Tool call**:
-```
-Task({
-  subagent_type: "code-developer",
-  run_in_background: false,
-  description: "Implement solution for <issueId>",
-  prompt: <execution-prompt>
-})
-```
-
-Synchronous execution - wait for completion before Phase 4.
-
-#### Option B: Codex CLI Execution
-
-**When**: executor === 'codex' (complex tasks, background execution)
-
-**Tool call**:
-```
-Bash("ccw cli -p \"<execution-prompt>\" --tool codex --mode write --id planex-<issueId>", { run_in_background: true })
-```
-
-**Resume on failure**:
-```
-ccw cli -p "Continue implementation" --resume planex-<issueId> --tool codex --mode write --id planex-<issueId>-retry
-```
-
-STOP after spawn - CLI executes in background, wait for task hook callback.
-
-#### Option C: Gemini CLI Execution
-
-**When**: executor === 'gemini' (analysis-heavy tasks, background execution)
-
-**Tool call**:
-```
-Bash("ccw cli -p \"<execution-prompt>\" --tool gemini --mode write --id planex-<issueId>", { run_in_background: true })
-```
-
-STOP after spawn - CLI executes in background, wait for task hook callback.
-
-### Execution Prompt Template
-
-All backends use unified prompt structure:
-
-```
-## Issue
-ID: <issueId>
-Title: <solution-title>
-
-## Solution Plan
-<solution-bound-json>
-
-## Implementation Requirements
-
-1. Follow the solution plan tasks in order
-2. Write clean, minimal code following existing patterns
-3. Run tests after each significant change
-4. Ensure all existing tests still pass
-5. Do NOT over-engineer - implement exactly what the solution specifies
-
-## Quality Checklist
-- [ ] All solution tasks implemented
-- [ ] No TypeScript/linting errors
-- [ ] Existing tests pass
-- [ ] New tests added where appropriate
-- [ ] No security vulnerabilities introduced
-
-## Project Guidelines
-@.workflow/specs/*.md
-```
-
-### Phase 4: Verify & Commit
-
-**Test Detection**:
-
-| Detection | Method |
-|-----------|--------|
-| package.json scripts.test | Use `npm test` |
-| package.json scripts.test:unit | Use `npm run test:unit` |
-| No test script found | Skip verification, proceed to commit |
-
-**Test Verification**:
-
-```
-Bash("<testCmd> 2>&1 || echo TEST_FAILED")
-```
-
-Check output for `TEST_FAILED` or `FAIL` strings.
-
-**Test Failure Handling**:
-
-| Condition | Action |
-|-----------|--------|
-| Tests failing | Report impl_failed to planner with test output + resume command |
-| Tests passing | Proceed to code review (if configured) |
-
-**Code Review (Optional)**:
-
-| Review Tool | Execution |
-|-------------|-----------|
-| Gemini Review | `ccw cli -p "<review-prompt>" --tool gemini --mode analysis --id planex-review-<issueId>` (background) |
-| Codex Review | `ccw cli --tool codex --mode review --uncommitted` (background, no prompt with target flags) |
-| Agent Review | Current agent performs inline review against solution convergence criteria |
-
-**Code Review Prompt**:
-```
-PURPOSE: Code review for <issueId> implementation against solution plan
-TASK: Verify solution convergence criteria | Check test coverage | Analyze code quality | Identify issues
-MODE: analysis
-CONTEXT: @**/* | Memory: Review planex execution for <issueId>
-EXPECTED: Quality assessment with issue identification and recommendations
-CONSTRAINTS: Focus on solution adherence and code quality | analysis=READ-ONLY
-```
-
-**Issue Completion**:
-
-```
-Bash("ccw issue update <issueId> --status completed")
-```
-
-### Phase 5: Report + Loop
-
-> See SKILL.md Shared Infrastructure -> Worker Phase 5: Report
-
-**Success Report**:
-
-```
-mcp__ccw-tools__team_msg({
-  operation: "log",
-  team: <session-id>,  // e.g., "PEX-project-2026-02-27", NOT "planex"
-  from: "executor",
-  to: "planner",
-  type: "impl_complete",
-  summary: "[executor] Implementation complete for <issueId> via <executor>, tests passing"
-})
-
-SendMessage({
-  type: "message",
-  recipient: "planner",
-  content: `## [executor] Implementation Complete
-
-**Issue**: <issueId>
-**Executor**: <executor>
-**Solution**: <solution-id>
-**Code Review**: <codeReview>
-**Status**: All tests passing
-**Issue Status**: Updated to resolved`,
-  summary: "[executor] EXEC complete: <issueId> (<executor>)"
-})
-
-TaskUpdate({ taskId: <task-id>, status: "completed" })
-```
-
-**Loop Check**:
-
-Query for next `EXEC-*` task with owner=executor, status=pending, blockedBy empty.
-
-| Condition | Action |
-|-----------|--------|
-| Tasks available | Return to Phase 1 for next task |
-| No tasks + planner sent all_planned | Send wave_done and idle |
-| No tasks + planner still planning | Idle for more tasks |
-
-**Wave Done Signal**:
-
-```
-mcp__ccw-tools__team_msg({
-  operation: "log",
-  team: <session-id>,  // e.g., "PEX-project-2026-02-27", NOT "planex"
-  from: "executor",
-  to: "planner",
-  type: "wave_done",
-  summary: "[executor] All EXEC tasks completed"
-})
-
-SendMessage({
-  type: "message",
-  recipient: "planner",
-  content: "## [executor] All Tasks Done\n\nAll EXEC-* tasks have been completed. Pipeline finished.",
-  summary: "[executor] wave_done: all complete"
-})
-```
-
----
-
-## Error Handling
-
-| Scenario | Resolution |
-|----------|------------|
-| No EXEC-* tasks available | Idle, wait for planner to create tasks |
-| Solution plan not found | Report error to planner |
-| Unknown execution_method | Fallback to `agent` with warning |
-| Agent (code-developer) failure | Retry once, then report impl_failed |
-| CLI (Codex/Gemini) failure | Provide resume command with fixed ID, report impl_failed |
-| CLI timeout | Use fixed ID `planex-{issueId}` for resume |
-| Tests failing after implementation | Report impl_failed with test output + resume info |
-| Issue status update failure | Log warning, continue with report |
-| Dependency not yet complete | Wait - task is blocked by blockedBy |
-| All tasks done but planner still planning | Send wave_done, then idle for more |
-| Critical issue beyond scope | SendMessage error to planner |

.claude/skills/team-planex/roles/planner.md

@@ -1,315 +0,0 @@
-# Planner Role
-
-Demand decomposition -> Issue creation -> Solution design -> Conflict check -> EXEC task dispatch. Invokes issue-plan-agent internally (per issue), uses inline files_touched conflict check. Dispatches EXEC-* task immediately after each issue's solution is ready. Planner also serves as lead role (no separate coordinator).
-
-## Identity
-
-- **Name**: `planner` | **Tag**: `[planner]`
-- **Task Prefix**: `PLAN-*`
-- **Responsibility**: Planning lead (requirement -> issues -> solutions -> queue -> dispatch)
-
-## Boundaries
-
-### MUST
-
-- Only process `PLAN-*` prefixed tasks
-- All output (SendMessage, team_msg, logs) must carry `[planner]` identifier
-- Immediately create `EXEC-*` task after completing each issue's solution and send `issue_ready` signal
-- Continue pushing forward without waiting for executor
-- Write solution artifacts to `<sessionDir>/artifacts/solutions/{issueId}.json`
-
-### MUST NOT
-
-- Directly write/modify business code (executor responsibility)
-- Call code-developer agent
-- Run project tests
-- git commit code changes
-
----
-
-## Toolbox
-
-### Subagent Capabilities
-
-| Agent Type | Purpose |
-|------------|---------|
-| `issue-plan-agent` | Closed-loop planning: ACE exploration + solution generation + binding (single issue granularity) |
-
-### CLI Capabilities
-
-| CLI Command | Purpose |
-|-------------|---------|
-| `ccw issue create --data '{"title":"..."}' --json` | Create issue from text |
-| `ccw issue status <id> --json` | Check issue status |
-| `ccw issue solution <id> --json` | Get single issue's solutions (requires issue ID) |
-| `ccw issue solutions --status planned --brief` | Batch list all bound solutions (cross-issue) |
-| `ccw issue bind <id> <sol-id>` | Bind solution to issue |
-
-### Skill Capabilities
-
-| Skill | Purpose |
-|-------|---------|
-| `Skill(skill="issue:new", args="--text '...'")` | Create issue from text |
-
----
-
-## Message Types
-
-| Type | Direction | Trigger | Description |
-|------|-----------|---------|-------------|
-| `issue_ready` | planner -> executor | Single issue solution + EXEC task created | Per-issue beat signal |
-| `wave_ready` | planner -> executor | All issues in a wave dispatched | Wave summary signal |
-| `all_planned` | planner -> executor | All waves planning complete | Final signal |
-| `error` | planner -> executor | Blocking error | Planning failure |
-
-## Message Bus
-
-Before every SendMessage, log via `mcp__ccw-tools__team_msg`:
-
-**NOTE**: `team` must be **session ID** (e.g., `PEX-project-2026-02-27`), NOT team name. Extract from `Session:` field in task description.
-
-```
-mcp__ccw-tools__team_msg({
-  operation: "log",
-  team: <session-id>,  // e.g., "PEX-project-2026-02-27", NOT "planex"
-  from: "planner",
-  to: "executor",
-  type: <message-type>,
-  summary: "[planner] <task-prefix> complete: <task-subject>",
-  ref: <artifact-path>
-})
-```
-
-**CLI fallback** (when MCP unavailable):
-
-```
-Bash("ccw team log --team <session-id> --from planner --to executor --type <message-type> --summary \"[planner] <task-prefix> complete\" --ref <artifact-path> --json")
-```
-
----
-
-## Execution (5-Phase)
-
-### Phase 1: Task Discovery
-
-> See SKILL.md Shared Infrastructure -> Worker Phase 1: Task Discovery
-
-Standard task discovery flow: TaskList -> filter by prefix `PLAN-*` + owner match + pending + unblocked -> TaskGet -> TaskUpdate in_progress.
-
-### Phase 2: Input Parsing
-
-Parse task description and arguments to determine input type.
-
-**Input Type Detection**:
-
-| Detection | Condition | Handler |
-|-----------|-----------|---------|
-| Issue IDs | Task description contains `ISS-\d{8}-\d{6}` pattern | Path C: Direct to planning |
-| Text input | Arguments contain `--text '...'` | Path A: Create issue first |
-| Plan file | Arguments contain `--plan <path>` | Path B: Parse and batch create |
-| Execution plan JSON | Plan file is `execution-plan.json` from req-plan | Path D: Wave-aware processing |
-| Description text | None of above | Treat task description as requirement text |
-
-**Execution Config Extraction**:
-
-From arguments, extract:
-- `execution_method`: Agent | Codex | Gemini | Auto (default: Auto)
-- `code_review`: Skip | Gemini Review | Codex Review | Agent Review (default: Skip)
-
-### Phase 3: Issue Processing Pipeline
-
-Execute different processing paths based on input type.
-
-#### Path A: Text Input -> Create Issue
-
-**Workflow**:
-1. Use `issue:new` skill to create issue from text
-2. Capture created issue ID
-3. Add to issue list for planning
-
-**Tool calls**:
-```
-Skill(skill="issue:new", args="--text '<requirement-text>'")
-```
-
-#### Path B: Plan File -> Batch Create Issues
-
-**Workflow**:
-1. Read plan file content
-2. Parse phases/steps from markdown structure
-3. For each phase/step, create an issue
-4. Add all issue IDs to list for planning
-
-**Plan Parsing Rules**:
-- Match `## Phase N: Title` or `## Step N: Title` or `### N. Title`
-- Each match creates one issue with title and description
-- Fallback: If no phase structure, entire content becomes single issue
-
-#### Path C: Issue IDs -> Direct Planning
-
-Issue IDs are ready, proceed directly to solution planning.
-
-#### Path D: execution-plan.json -> Wave-Aware Processing
-
-**Workflow**:
-1. Parse execution-plan.json with waves array
-2. For each wave, process issues sequentially
-3. For each issue in wave:
-   - Call issue-plan-agent to generate solution
-   - Write solution artifact to `<sessionDir>/artifacts/solutions/{issueId}.json`
-   - Perform inline conflict check
-   - Create EXEC-* task with solution_file path
-   - Send issue_ready signal
-4. After each wave completes, send wave_ready signal
-5. After all waves, send all_planned signal
-
-**Issue Planning (per issue)**:
-
-```
-Task({
-  subagent_type: "issue-plan-agent",
-  run_in_background: false,
-  description: "Plan solution for <issueId>",
-  prompt: `issue_ids: ["<issueId>"]
-project_root: "<projectRoot>"
-
-## Requirements
-- Generate solution for this issue
-- Auto-bind single solution
-- Issues come from req-plan decomposition (tags: req-plan)
-- Respect dependencies: <issue_dependencies>`
-})
-```
-
-**Solution Artifact**:
-
-```
-Write({
-  file_path: "<sessionDir>/artifacts/solutions/<issueId>.json",
-  content: JSON.stringify({
-    session_id: <sessionId>,
-    issue_id: <issueId>,
-    ...solution,
-    execution_config: { execution_method, code_review },
-    timestamp: <ISO-timestamp>
-  }, null, 2)
-})
-```
-
-**EXEC Task Creation**:
-
-```
-TaskCreate({
-  subject: "EXEC-W<waveNum>-<issueId>: Implement <solution-title>",
-  description: `## Execution Task
-**Wave**: <waveNum>
-**Issue**: <issueId>
-**solution_file**: <solutionFile>
-**execution_method**: <method>
-**code_review**: <review>`,
-  activeForm: "Implementing <issueId>",
-  owner: "executor"
-})
-```
-
-#### Wave Processing (Path A/B/C Convergence)
-
-For non-execution-plan inputs, process all issues as a single logical wave:
-
-**Workflow**:
-1. For each issue in list:
-   - Call issue-plan-agent
-   - Write solution artifact
-   - Perform inline conflict check
-   - Create EXEC-* task
-   - Send issue_ready signal
-2. After all issues complete, send wave_ready signal
-
-### Phase 4: Inline Conflict Check + Dispatch
-
-Perform conflict detection using files_touched overlap analysis.
-
-**Conflict Detection Rules**:
-
-| Condition | Action |
-|-----------|--------|
-| File overlap detected | Add blockedBy dependency to previous task |
-| Explicit dependency in solution.bound.dependencies.on_issues | Add blockedBy to referenced task |
-| No conflict | No blockedBy, task is immediately executable |
-
-**Inline Conflict Check Algorithm**:
-
-1. Get current solution's files_touched (or affected_files)
-2. For each previously dispatched solution:
-   - Check if any files overlap
-   - If overlap, add previous execTaskId to blockedBy
-3. Check explicit dependencies from solution.bound.dependencies.on_issues
-4. Return blockedBy array for TaskUpdate
-
-**Wave Summary Signal** (after all issues in wave):
-
-```
-mcp__ccw-tools__team_msg({
-  operation: "log", team: <session-id>, from: "planner", to: "executor",  // team = session ID
-  type: "wave_ready",
-  summary: "[planner] Wave <waveNum> fully dispatched: <issueCount> issues"
-})
-
-SendMessage({
-  type: "message", recipient: "executor",
-  content: "## [planner] Wave <waveNum> Complete\nAll issues dispatched, <count> EXEC tasks created.",
-  summary: "[planner] wave_ready: wave <waveNum>"
-})
-```
-
-### Phase 5: Report + Finalize
-
-> See SKILL.md Shared Infrastructure -> Worker Phase 5: Report
-
-**Final Signal** (all waves complete):
-
-```
-mcp__ccw-tools__team_msg({
-  operation: "log",
-  team: <session-id>,  // e.g., "PEX-project-2026-02-27", NOT "planex"
-  from: "planner",
-  to: "executor",
-  type: "all_planned",
-  summary: "[planner] All <waveCount> waves planned, <issueCount> issues total"
-})
-
-SendMessage({
-  type: "message",
-  recipient: "executor",
-  content: `## [planner] All Waves Planned
-
-**Total Waves**: <waveCount>
-**Total Issues**: <issueCount>
-**Status**: All planning complete, waiting for executor to finish remaining EXEC-* tasks
-
-Pipeline complete when executor sends wave_done confirmation.`,
-  summary: "[planner] all_planned: <waveCount> waves, <issueCount> issues"
-})
-
-TaskUpdate({ taskId: <task-id>, status: "completed" })
-```
-
-**Loop Check**: Query for next `PLAN-*` task with owner=planner, status=pending, blockedBy empty. If found, return to Phase 1.
-
----
-
-## Error Handling
-
-| Scenario | Resolution |
-|----------|------------|
-| No PLAN-* tasks available | Idle, wait for orchestrator |
-| Issue creation failure | Retry once with simplified text, then report error |
-| issue-plan-agent failure | Retry once, then report error and skip to next issue |
-| Inline conflict check failure | Skip conflict detection, create EXEC task without blockedBy |
-| Plan file not found | Report error with expected path |
-| execution-plan.json parse failure | Fallback to plan_file parsing (Path B) |
-| execution-plan.json missing waves | Report error, suggest re-running req-plan |
-| Empty input (no issues, no text, no plan) | AskUserQuestion for clarification |
-| Solution artifact write failure | Log warning, create EXEC task without solution_file (executor fallback) |
-| Wave partially failed | Report partial success, continue with successful issues |
-| Critical issue beyond scope | SendMessage error to executor |

.claude/skills/workflow-wave-plan/SKILL.md

@@ -0,0 +1,896 @@
+---
+name: workflow-wave-plan
+description: CSV Wave planning and execution - explore via wave, resolve conflicts, execute from CSV with linked exploration context. Triggers on "workflow:wave-plan".
+argument-hint: "<task description> [--yes|-y] [--concurrency|-c N]"
+allowed-tools: Task, AskUserQuestion, Read, Write, Edit, Bash, Glob, Grep
+---
+
+# Workflow Wave Plan
+
+CSV Wave-based planning and execution. Uses structured CSV state for both exploration and execution, with cross-phase context propagation via `context_from` linking.
+
+## Architecture
+
+```
+Requirement
+    ↓
+┌─ Phase 1: Decompose ─────────────────────┐
+│  Analyze requirement → explore.csv         │
+│  (1 row per exploration angle)             │
+└────────────────────┬──────────────────────┘
+                     ↓
+┌─ Phase 2: Wave Explore ──────────────────┐
+│  Wave loop: spawn Explore agents          │
+│  → findings/key_files → explore.csv       │
+└────────────────────┬──────────────────────┘
+                     ↓
+┌─ Phase 3: Synthesize & Plan ─────────────┐
+│  Read explore findings → cross-reference  │
+│  → resolve conflicts → tasks.csv          │
+│  (context_from links to E* explore rows)  │
+└────────────────────┬──────────────────────┘
+                     ↓
+┌─ Phase 4: Wave Execute ──────────────────┐
+│  Wave loop: build prev_context from CSV   │
+│  → spawn code-developer agents per wave   │
+│  → results → tasks.csv                    │
+└────────────────────┬──────────────────────┘
+                     ↓
+┌─ Phase 5: Aggregate ─────────────────────┐
+│  results.csv + context.md + summary       │
+└───────────────────────────────────────────┘
+```
+
+## Context Flow
+
+```
+explore.csv             tasks.csv
+┌──────────┐           ┌──────────┐
+│ E1: arch │──────────→│ T1: setup│ context_from: E1;E2
+│ findings │           │ prev_ctx │← E1+E2 findings
+├──────────┤           ├──────────┤
+│ E2: deps │──────────→│ T2: impl │ context_from: E1;T1
+│ findings │           │ prev_ctx │← E1+T1 findings
+├──────────┤           ├──────────┤
+│ E3: test │──┐   ┌───→│ T3: test │ context_from: E3;T2
+│ findings │  └───┘    │ prev_ctx │← E3+T2 findings
+└──────────┘           └──────────┘
+
+Two context channels:
+1. Directed: context_from → prev_context (from CSV findings)
+2. Broadcast: discoveries.ndjson (append-only shared board)
+```
+
+---
+
+## CSV Schemas
+
+### explore.csv
+
+| Column | Type | Set By | Description |
+|--------|------|--------|-------------|
+| `id` | string | Decomposer | E1, E2, ... |
+| `angle` | string | Decomposer | Exploration angle name |
+| `description` | string | Decomposer | What to explore from this angle |
+| `focus` | string | Decomposer | Keywords and focus areas |
+| `deps` | string | Decomposer | Semicolon-separated dep IDs (usually empty) |
+| `wave` | integer | Wave Engine | Wave number (usually 1) |
+| `status` | enum | Agent | pending / completed / failed |
+| `findings` | string | Agent | Discoveries (max 800 chars) |
+| `key_files` | string | Agent | Relevant files (semicolon-separated) |
+| `error` | string | Agent | Error message if failed |
+
+### tasks.csv
+
+| Column | Type | Set By | Description |
+|--------|------|--------|-------------|
+| `id` | string | Planner | T1, T2, ... |
+| `title` | string | Planner | Task title |
+| `description` | string | Planner | Self-contained task description — what to implement |
+| `test` | string | Planner | Test cases: what tests to write and how to verify (unit/integration/edge) |
+| `acceptance_criteria` | string | Planner | Measurable conditions that define "done" |
+| `scope` | string | Planner | Target file/directory glob — constrains agent write area, prevents cross-task file conflicts |
+| `hints` | string | Planner | Implementation tips + reference files. Format: `tips text \|\| file1;file2`. Either part is optional |
+| `execution_directives` | string | Planner | Execution constraints: commands to run for verification, tool restrictions |
+| `deps` | string | Planner | Dependency task IDs: T1;T2 |
+| `context_from` | string | Planner | Context source IDs: **E1;E2;T1** |
+| `wave` | integer | Wave Engine | Wave number (computed from deps) |
+| `status` | enum | Agent | pending / completed / failed / skipped |
+| `findings` | string | Agent | Execution findings (max 500 chars) |
+| `files_modified` | string | Agent | Files modified (semicolon-separated) |
+| `tests_passed` | boolean | Agent | Whether all defined test cases passed (true/false) |
+| `acceptance_met` | string | Agent | Summary of which acceptance criteria were met/unmet |
+| `error` | string | Agent | Error if failed |
+
+**context_from prefix convention**: `E*` → explore.csv lookup, `T*` → tasks.csv lookup.
+
+---
+
+## Session Structure
+
+```
+.workflow/.wave-plan/{session-id}/
+├── explore.csv              # Exploration state
+├── tasks.csv                # Execution state
+├── discoveries.ndjson       # Shared discovery board
+├── explore-results/         # Detailed per-angle results
+│   ├── E1.json
+│   └── E2.json
+├── task-results/            # Detailed per-task results
+│   ├── T1.json
+│   └── T2.json
+├── results.csv              # Final results export
+└── context.md               # Full context summary
+```
+
+---
+
+## Session Initialization
+
+```javascript
+const getUtc8ISOString = () => new Date(Date.now() + 8 * 60 * 60 * 1000).toISOString()
+
+// Parse flags
+const AUTO_YES = $ARGUMENTS.includes('--yes') || $ARGUMENTS.includes('-y')
+const concurrencyMatch = $ARGUMENTS.match(/(?:--concurrency|-c)\s+(\d+)/)
+const maxConcurrency = concurrencyMatch ? parseInt(concurrencyMatch[1]) : 4
+
+const requirement = $ARGUMENTS
+  .replace(/--yes|-y|--concurrency\s+\d+|-c\s+\d+/g, '')
+  .trim()
+
+const slug = requirement.toLowerCase()
+  .replace(/[^a-z0-9\u4e00-\u9fa5]+/g, '-')
+  .substring(0, 40)
+const dateStr = getUtc8ISOString().substring(0, 10).replace(/-/g, '')
+const sessionId = `wp-${slug}-${dateStr}`
+const sessionFolder = `.workflow/.wave-plan/${sessionId}`
+
+Bash(`mkdir -p ${sessionFolder}/explore-results ${sessionFolder}/task-results`)
+```
+
+---
+
+## Phase 1: Decompose → explore.csv
+
+### 1.1 Analyze Requirement
+
+```javascript
+const complexity = analyzeComplexity(requirement)
+// Low: 1 angle | Medium: 2-3 angles | High: 3-4 angles
+
+const ANGLE_PRESETS = {
+  architecture: ['architecture', 'dependencies', 'integration-points', 'modularity'],
+  security:     ['security', 'auth-patterns', 'dataflow', 'validation'],
+  performance:  ['performance', 'bottlenecks', 'caching', 'data-access'],
+  bugfix:       ['error-handling', 'dataflow', 'state-management', 'edge-cases'],
+  feature:      ['patterns', 'integration-points', 'testing', 'dependencies']
+}
+
+function selectAngles(text, count) {
+  let preset = 'feature'
+  if (/refactor|architect|restructure|modular/.test(text)) preset = 'architecture'
+  else if (/security|auth|permission|access/.test(text)) preset = 'security'
+  else if (/performance|slow|optimi|cache/.test(text)) preset = 'performance'
+  else if (/fix|bug|error|broken/.test(text)) preset = 'bugfix'
+  return ANGLE_PRESETS[preset].slice(0, count)
+}
+
+const angleCount = complexity === 'High' ? 4 : complexity === 'Medium' ? 3 : 1
+const angles = selectAngles(requirement, angleCount)
+```
+
+### 1.2 Generate explore.csv
+
+```javascript
+const header = 'id,angle,description,focus,deps,wave,status,findings,key_files,error'
+const rows = angles.map((angle, i) => {
+  const id = `E${i + 1}`
+  const desc = `Explore codebase from ${angle} perspective for: ${requirement}`
+  return `"${id}","${angle}","${escCSV(desc)}","${angle}","",1,"pending","","",""`
+})
+
+Write(`${sessionFolder}/explore.csv`, [header, ...rows].join('\n'))
+```
+
+All exploration rows default to wave 1 (independent parallel). If angle dependencies exist, compute waves.
+
+---
+
+## Phase 2: Wave Explore
+
+Execute exploration waves using `Task(Explore)` agents.
+
+### 2.1 Wave Loop
+
+```javascript
+const exploreCSV = parseCSV(Read(`${sessionFolder}/explore.csv`))
+const maxExploreWave = Math.max(...exploreCSV.map(r => parseInt(r.wave)))
+
+for (let wave = 1; wave <= maxExploreWave; wave++) {
+  const waveRows = exploreCSV.filter(r =>
+    parseInt(r.wave) === wave && r.status === 'pending'
+  )
+  if (waveRows.length === 0) continue
+
+  // Skip rows with failed dependencies
+  const validRows = waveRows.filter(r => {
+    if (!r.deps) return true
+    return r.deps.split(';').filter(Boolean).every(depId => {
+      const dep = exploreCSV.find(d => d.id === depId)
+      return dep && dep.status === 'completed'
+    })
+  })
+
+  waveRows.filter(r => !validRows.includes(r)).forEach(r => {
+    r.status = 'skipped'
+    r.error = 'Dependency failed/skipped'
+  })
+
+  // ★ Spawn ALL explore agents in SINGLE message → parallel execution
+  const results = validRows.map(row =>
+    Task({
+      subagent_type: "Explore",
+      run_in_background: false,
+      description: `Explore: ${row.angle}`,
+      prompt: buildExplorePrompt(row, requirement, sessionFolder)
+    })
+  )
+
+  // Collect results from JSON files → update explore.csv
+  validRows.forEach((row, i) => {
+    const resultPath = `${sessionFolder}/explore-results/${row.id}.json`
+    if (fileExists(resultPath)) {
+      const result = JSON.parse(Read(resultPath))
+      row.status = result.status || 'completed'
+      row.findings = truncate(result.findings, 800)
+      row.key_files = Array.isArray(result.key_files)
+        ? result.key_files.join(';')
+        : (result.key_files || '')
+      row.error = result.error || ''
+    } else {
+      // Fallback: parse from agent output text
+      row.status = 'completed'
+      row.findings = truncate(results[i], 800)
+    }
+  })
+
+  writeCSV(`${sessionFolder}/explore.csv`, exploreCSV)
+}
+```
+
+### 2.2 Explore Agent Prompt
+
+```javascript
+function buildExplorePrompt(row, requirement, sessionFolder) {
+  return `## Exploration: ${row.angle}
+
+**Requirement**: ${requirement}
+**Focus**: ${row.focus}
+
+### MANDATORY FIRST STEPS
+1. Read shared discoveries: ${sessionFolder}/discoveries.ndjson (if exists, skip if not)
+2. Read project context: .workflow/project-tech.json (if exists)
+
+---
+
+## Instructions
+Explore the codebase from the **${row.angle}** perspective:
+1. Discover relevant files, modules, and patterns
+2. Identify integration points and dependencies
+3. Note constraints, risks, and conventions
+4. Find existing patterns to follow
+5. Share discoveries: append findings to ${sessionFolder}/discoveries.ndjson
+
+## Output
+Write findings to: ${sessionFolder}/explore-results/${row.id}.json
+
+JSON format:
+{
+  "status": "completed",
+  "findings": "Concise summary of ${row.angle} discoveries (max 800 chars)",
+  "key_files": ["relevant/file1.ts", "relevant/file2.ts"],
+  "details": {
+    "patterns": ["pattern descriptions"],
+    "integration_points": [{"file": "path", "description": "..."}],
+    "constraints": ["constraint descriptions"],
+    "recommendations": ["recommendation descriptions"]
+  }
+}
+
+Also provide a 2-3 sentence summary.`
+}
+```
+
+---
+
+## Phase 3: Synthesize & Plan → tasks.csv
+
+Read exploration findings, cross-reference, resolve conflicts, generate execution tasks.
+
+### 3.1 Load Explore Results
+
+```javascript
+const exploreCSV = parseCSV(Read(`${sessionFolder}/explore.csv`))
+const completed = exploreCSV.filter(r => r.status === 'completed')
+
+// Load detailed result JSONs where available
+const detailedResults = {}
+completed.forEach(r => {
+  const path = `${sessionFolder}/explore-results/${r.id}.json`
+  if (fileExists(path)) detailedResults[r.id] = JSON.parse(Read(path))
+})
+```
+
+### 3.2 Conflict Resolution Protocol
+
+Cross-reference findings across all exploration angles:
+
+```javascript
+// 1. Identify common files referenced by multiple angles
+const fileRefs = {}
+completed.forEach(r => {
+  r.key_files.split(';').filter(Boolean).forEach(f => {
+    if (!fileRefs[f]) fileRefs[f] = []
+    fileRefs[f].push({ angle: r.angle, id: r.id })
+  })
+})
+const sharedFiles = Object.entries(fileRefs).filter(([_, refs]) => refs.length > 1)
+
+// 2. Detect conflicting recommendations
+//    Compare recommendations from different angles for same file/module
+//    Flag contradictions (angle A says "refactor X" vs angle B says "extend X")
+
+// 3. Resolution rules:
+//    a. Safety first — when approaches conflict, choose safer option
+//    b. Consistency — prefer approaches aligned with existing patterns
+//    c. Scope — prefer minimal-change approaches
+//    d. Document — note all resolved conflicts for transparency
+
+const synthesis = {
+  sharedFiles,
+  conflicts: detectConflicts(completed, detailedResults),
+  resolutions: [],
+  allKeyFiles: [...new Set(completed.flatMap(r => r.key_files.split(';').filter(Boolean)))]
+}
+```
+
+### 3.3 Generate tasks.csv
+
+Decompose into execution tasks based on synthesized exploration:
+
+```javascript
+// Task decomposition rules:
+// 1. Group by feature/module (not per-file)
+// 2. Each description is self-contained (agent sees only its row + prev_context)
+// 3. deps only when task B requires task A's output
+// 4. context_from links relevant explore rows (E*) and predecessor tasks (T*)
+// 5. Prefer parallel (minimize deps)
+// 6. Use exploration findings: key_files → target files, patterns → references,
+//    integration_points → dependency relationships, constraints → included in description
+// 7. Each task MUST include: test (how to verify), acceptance_criteria (what defines done)
+// 8. scope must not overlap between tasks in the same wave
+// 9. hints = implementation tips + reference files (format: tips || file1;file2)
+// 10. execution_directives = commands to run for verification, tool restrictions
+
+const tasks = []
+// Claude decomposes requirement using exploration synthesis
+// Example:
+// tasks.push({ id: 'T1', title: 'Setup types', description: '...', test: 'Verify types compile', acceptance_criteria: 'All interfaces exported', scope: 'src/types/**', hints: 'Follow existing type patterns || src/types/index.ts', execution_directives: 'tsc --noEmit', deps: '', context_from: 'E1;E2' })
+// tasks.push({ id: 'T2', title: 'Implement core', description: '...', test: 'Unit test: core logic', acceptance_criteria: 'All functions pass tests', scope: 'src/core/**', hints: 'Reuse BaseService || src/services/Base.ts', execution_directives: 'npm test -- --grep core', deps: 'T1', context_from: 'E1;E2;T1' })
+// tasks.push({ id: 'T3', title: 'Add tests', description: '...', test: 'Integration test suite', acceptance_criteria: '>80% coverage', scope: 'tests/**', hints: 'Follow existing test patterns || tests/auth.test.ts', execution_directives: 'npm test', deps: 'T2', context_from: 'E3;T2' })
+
+// Compute waves
+const waves = computeWaves(tasks)
+tasks.forEach(t => { t.wave = waves[t.id] })
+
+// Write tasks.csv
+const header = 'id,title,description,test,acceptance_criteria,scope,hints,execution_directives,deps,context_from,wave,status,findings,files_modified,tests_passed,acceptance_met,error'
+const rows = tasks.map(t =>
+  [t.id, escCSV(t.title), escCSV(t.description), escCSV(t.test), escCSV(t.acceptance_criteria), escCSV(t.scope), escCSV(t.hints), escCSV(t.execution_directives), t.deps, t.context_from, t.wave, 'pending', '', '', '', '', '']
+    .map(v => `"${v}"`).join(',')
+)
+
+Write(`${sessionFolder}/tasks.csv`, [header, ...rows].join('\n'))
+```
+
+### 3.4 User Confirmation
+
+```javascript
+if (!AUTO_YES) {
+  const maxWave = Math.max(...tasks.map(t => t.wave))
+
+  console.log(`
+## Execution Plan
+
+Explore: ${completed.length} angles completed
+Conflicts resolved: ${synthesis.conflicts.length}
+Tasks: ${tasks.length} across ${maxWave} waves
+
+${Array.from({length: maxWave}, (_, i) => i + 1).map(w => {
+  const wt = tasks.filter(t => t.wave === w)
+  return `### Wave ${w} (${wt.length} tasks, concurrent)
+${wt.map(t => `  - [${t.id}] ${t.title} (from: ${t.context_from})`).join('\n')}`
+}).join('\n')}
+  `)
+
+  AskUserQuestion({
+    questions: [{
+      question: `Proceed with ${tasks.length} tasks across ${maxWave} waves?`,
+      header: "Confirm",
+      multiSelect: false,
+      options: [
+        { label: "Execute", description: "Proceed with wave execution" },
+        { label: "Modify", description: `Edit ${sessionFolder}/tasks.csv then re-run` },
+        { label: "Cancel", description: "Abort" }
+      ]
+    }]
+  })
+}
+```
+
+---
+
+## Phase 4: Wave Execute
+
+Execute tasks from tasks.csv in wave order, with prev_context built from both explore.csv and tasks.csv.
+
+### 4.1 Wave Loop
+
+```javascript
+const exploreCSV = parseCSV(Read(`${sessionFolder}/explore.csv`))
+const failedIds = new Set()
+const skippedIds = new Set()
+
+let tasksCSV = parseCSV(Read(`${sessionFolder}/tasks.csv`))
+const maxWave = Math.max(...tasksCSV.map(r => parseInt(r.wave)))
+
+for (let wave = 1; wave <= maxWave; wave++) {
+  // Re-read master CSV (updated by previous wave)
+  tasksCSV = parseCSV(Read(`${sessionFolder}/tasks.csv`))
+
+  const waveRows = tasksCSV.filter(r =>
+    parseInt(r.wave) === wave && r.status === 'pending'
+  )
+  if (waveRows.length === 0) continue
+
+  // Skip on failed dependencies (cascade)
+  const validRows = []
+  for (const row of waveRows) {
+    const deps = (row.deps || '').split(';').filter(Boolean)
+    if (deps.some(d => failedIds.has(d) || skippedIds.has(d))) {
+      skippedIds.add(row.id)
+      row.status = 'skipped'
+      row.error = 'Dependency failed/skipped'
+      continue
+    }
+    validRows.push(row)
+  }
+
+  if (validRows.length === 0) {
+    writeCSV(`${sessionFolder}/tasks.csv`, tasksCSV)
+    continue
+  }
+
+  // Build prev_context for each row from explore.csv + tasks.csv
+  validRows.forEach(row => {
+    row._prev_context = buildPrevContext(row.context_from, exploreCSV, tasksCSV)
+  })
+
+  // ★ Spawn ALL task agents in SINGLE message → parallel execution
+  const results = validRows.map(row =>
+    Task({
+      subagent_type: "code-developer",
+      run_in_background: false,
+      description: row.title,
+      prompt: buildExecutePrompt(row, requirement, sessionFolder)
+    })
+  )
+
+  // Collect results → update tasks.csv
+  validRows.forEach((row, i) => {
+    const resultPath = `${sessionFolder}/task-results/${row.id}.json`
+    if (fileExists(resultPath)) {
+      const result = JSON.parse(Read(resultPath))
+      row.status = result.status || 'completed'
+      row.findings = truncate(result.findings, 500)
+      row.files_modified = Array.isArray(result.files_modified)
+        ? result.files_modified.join(';')
+        : (result.files_modified || '')
+      row.tests_passed = String(result.tests_passed ?? '')
+      row.acceptance_met = result.acceptance_met || ''
+      row.error = result.error || ''
+    } else {
+      row.status = 'completed'
+      row.findings = truncate(results[i], 500)
+    }
+
+    if (row.status === 'failed') failedIds.add(row.id)
+    delete row._prev_context  // runtime-only, don't persist
+  })
+
+  writeCSV(`${sessionFolder}/tasks.csv`, tasksCSV)
+}
+```
+
+### 4.2 prev_context Builder
+
+The key function linking exploration context to execution:
+
+```javascript
+function buildPrevContext(contextFrom, exploreCSV, tasksCSV) {
+  if (!contextFrom) return 'No previous context available'
+
+  const ids = contextFrom.split(';').filter(Boolean)
+  const entries = []
+
+  ids.forEach(id => {
+    if (id.startsWith('E')) {
+      // ← Look up in explore.csv (cross-phase link)
+      const row = exploreCSV.find(r => r.id === id)
+      if (row && row.status === 'completed' && row.findings) {
+        entries.push(`[Explore ${row.angle}] ${row.findings}`)
+        if (row.key_files) entries.push(`  Key files: ${row.key_files}`)
+      }
+    } else if (id.startsWith('T')) {
+      // ← Look up in tasks.csv (same-phase link)
+      const row = tasksCSV.find(r => r.id === id)
+      if (row && row.status === 'completed' && row.findings) {
+        entries.push(`[Task ${row.id}: ${row.title}] ${row.findings}`)
+        if (row.files_modified) entries.push(`  Modified: ${row.files_modified}`)
+      }
+    }
+  })
+
+  return entries.length > 0 ? entries.join('\n') : 'No previous context available'
+}
+```
+
+### 4.3 Execute Agent Prompt
+
+```javascript
+function buildExecutePrompt(row, requirement, sessionFolder) {
+  return `## Task: ${row.title}
+
+**ID**: ${row.id}
+**Goal**: ${requirement}
+**Scope**: ${row.scope || 'Not specified'}
+
+## Description
+${row.description}
+
+### Implementation Hints & Reference Files
+${row.hints || 'None'}
+
+> Format: \`tips text || file1;file2\`. Read ALL reference files (after ||) before starting. Apply tips (before ||) as guidance.
+
+### Execution Directives
+${row.execution_directives || 'None'}
+
+> Commands to run for verification, tool restrictions, or environment requirements.
+
+### Test Cases
+${row.test || 'None specified'}
+
+### Acceptance Criteria
+${row.acceptance_criteria || 'None specified'}
+
+## Previous Context (from exploration and predecessor tasks)
+${row._prev_context}
+
+### MANDATORY FIRST STEPS
+1. Read shared discoveries: ${sessionFolder}/discoveries.ndjson (if exists, skip if not)
+2. Read project context: .workflow/project-tech.json (if exists)
+
+---
+
+## Execution Protocol
+
+1. **Read references**: Parse hints — read all files listed after \`||\` to understand existing patterns
+2. **Read discoveries**: Load ${sessionFolder}/discoveries.ndjson for shared exploration findings
+3. **Use context**: Apply previous tasks' findings from prev_context above
+4. **Stay in scope**: ONLY create/modify files within ${row.scope || 'project'} — do NOT touch files outside this boundary
+5. **Apply hints**: Follow implementation tips from hints (before \`||\`)
+6. **Implement**: Execute changes described in the task description
+7. **Write tests**: Implement the test cases defined above
+8. **Run directives**: Execute commands from execution_directives to verify your work
+9. **Verify acceptance**: Ensure all acceptance criteria are met before reporting completion
+10. **Share discoveries**: Append exploration findings to shared board:
+   \\\`\\\`\\\`bash
+   echo '{"ts":"<ISO>","worker":"${row.id}","type":"<type>","data":{...}}' >> ${sessionFolder}/discoveries.ndjson
+   \\\`\\\`\\\`
+11. **Report result**: Write JSON to output file
+
+## Output
+Write results to: ${sessionFolder}/task-results/${row.id}.json
+
+{
+  "status": "completed" | "failed",
+  "findings": "What was done (max 500 chars)",
+  "files_modified": ["file1.ts", "file2.ts"],
+  "tests_passed": true | false,
+  "acceptance_met": "Summary of which acceptance criteria were met/unmet",
+  "error": ""
+}
+
+**IMPORTANT**: Set status to "completed" ONLY if:
+- All test cases pass
+- All acceptance criteria are met
+Otherwise set status to "failed" with details in error field.`
+}
+```
+
+---
+
+## Phase 5: Aggregate
+
+### 5.1 Generate Results
+
+```javascript
+const finalTasks = parseCSV(Read(`${sessionFolder}/tasks.csv`))
+const exploreCSV = parseCSV(Read(`${sessionFolder}/explore.csv`))
+
+Bash(`cp "${sessionFolder}/tasks.csv" "${sessionFolder}/results.csv"`)
+
+const completed = finalTasks.filter(r => r.status === 'completed')
+const failed = finalTasks.filter(r => r.status === 'failed')
+const skipped = finalTasks.filter(r => r.status === 'skipped')
+const maxWave = Math.max(...finalTasks.map(r => parseInt(r.wave)))
+```
+
+### 5.2 Generate context.md
+
+```javascript
+const contextMd = `# Wave Plan Results
+
+**Requirement**: ${requirement}
+**Session**: ${sessionId}
+**Timestamp**: ${getUtc8ISOString()}
+
+## Summary
+
+| Metric | Count |
+|--------|-------|
+| Explore Angles | ${exploreCSV.length} |
+| Total Tasks | ${finalTasks.length} |
+| Completed | ${completed.length} |
+| Failed | ${failed.length} |
+| Skipped | ${skipped.length} |
+| Waves | ${maxWave} |
+
+## Exploration Results
+
+${exploreCSV.map(e => `### ${e.id}: ${e.angle} (${e.status})
+${e.findings || 'N/A'}
+Key files: ${e.key_files || 'none'}`).join('\n\n')}
+
+## Task Results
+
+${finalTasks.map(t => `### ${t.id}: ${t.title} (${t.status})
+
+| Field | Value |
+|-------|-------|
+| Wave | ${t.wave} |
+| Scope | ${t.scope || 'none'} |
+| Dependencies | ${t.deps || 'none'} |
+| Context From | ${t.context_from || 'none'} |
+| Tests Passed | ${t.tests_passed || 'N/A'} |
+| Acceptance Met | ${t.acceptance_met || 'N/A'} |
+| Error | ${t.error || 'none'} |
+
+**Description**: ${t.description}
+
+**Test Cases**: ${t.test || 'N/A'}
+
+**Acceptance Criteria**: ${t.acceptance_criteria || 'N/A'}
+
+**Hints**: ${t.hints || 'N/A'}
+
+**Execution Directives**: ${t.execution_directives || 'N/A'}
+
+**Findings**: ${t.findings || 'N/A'}
+
+**Files Modified**: ${t.files_modified || 'none'}`).join('\n\n---\n\n')}
+
+## All Modified Files
+
+${[...new Set(finalTasks.flatMap(t =>
+  (t.files_modified || '').split(';')).filter(Boolean)
+)].map(f => '- ' + f).join('\n') || 'None'}
+`
+
+Write(`${sessionFolder}/context.md`, contextMd)
+```
+
+### 5.3 Summary & Next Steps
+
+```javascript
+console.log(`
+## Wave Plan Complete
+
+Session: ${sessionFolder}
+Explore: ${exploreCSV.filter(r => r.status === 'completed').length}/${exploreCSV.length} angles
+Tasks: ${completed.length}/${finalTasks.length} completed, ${failed.length} failed, ${skipped.length} skipped
+Waves: ${maxWave}
+
+Files:
+- explore.csv — exploration state
+- tasks.csv — execution state
+- results.csv — final results
+- context.md — full report
+- discoveries.ndjson — shared discoveries
+`)
+
+if (!AUTO_YES && failed.length > 0) {
+  AskUserQuestion({
+    questions: [{
+      question: `${failed.length} tasks failed. Next action?`,
+      header: "Next Step",
+      multiSelect: false,
+      options: [
+        { label: "Retry Failed", description: "Reset failed + skipped, re-execute Phase 4" },
+        { label: "View Report", description: "Display context.md" },
+        { label: "Done", description: "Complete session" }
+      ]
+    }]
+  })
+  // If Retry: reset failed/skipped status to pending, re-run Phase 4
+}
+```
+
+---
+
+## Utilities
+
+### Wave Computation (Kahn's BFS)
+
+```javascript
+function computeWaves(tasks) {
+  const inDegree = {}, adj = {}, depth = {}
+  tasks.forEach(t => { inDegree[t.id] = 0; adj[t.id] = []; depth[t.id] = 1 })
+
+  tasks.forEach(t => {
+    const deps = (t.deps || '').split(';').filter(Boolean)
+    deps.forEach(dep => {
+      if (adj[dep]) { adj[dep].push(t.id); inDegree[t.id]++ }
+    })
+  })
+
+  const queue = Object.keys(inDegree).filter(id => inDegree[id] === 0)
+  queue.forEach(id => { depth[id] = 1 })
+
+  while (queue.length > 0) {
+    const current = queue.shift()
+    adj[current].forEach(next => {
+      depth[next] = Math.max(depth[next], depth[current] + 1)
+      inDegree[next]--
+      if (inDegree[next] === 0) queue.push(next)
+    })
+  }
+
+  if (Object.values(inDegree).some(d => d > 0)) {
+    throw new Error('Circular dependency detected')
+  }
+
+  return depth // { taskId: waveNumber }
+}
+```
+
+### CSV Helpers
+
+```javascript
+function escCSV(s) { return String(s || '').replace(/"/g, '""') }
+
+function parseCSV(content) {
+  const lines = content.trim().split('\n')
+  const header = lines[0].split(',').map(h => h.replace(/"/g, '').trim())
+  return lines.slice(1).filter(l => l.trim()).map(line => {
+    const values = parseCSVLine(line)
+    const row = {}
+    header.forEach((col, i) => { row[col] = (values[i] || '').replace(/^"|"$/g, '') })
+    return row
+  })
+}
+
+function writeCSV(path, rows) {
+  if (rows.length === 0) return
+  // Exclude runtime-only columns (prefixed with _)
+  const cols = Object.keys(rows[0]).filter(k => !k.startsWith('_'))
+  const header = cols.join(',')
+  const lines = rows.map(r =>
+    cols.map(c => `"${escCSV(r[c])}"`).join(',')
+  )
+  Write(path, [header, ...lines].join('\n'))
+}
+
+function truncate(s, max) {
+  s = String(s || '')
+  return s.length > max ? s.substring(0, max - 3) + '...' : s
+}
+```
+
+---
+
+## Discovery Board Protocol
+
+Shared `discoveries.ndjson` — append-only NDJSON accessible to all agents across all phases.
+
+**Lifecycle**:
+- Created by the first agent to write a discovery
+- Carries over across all phases and waves — never cleared
+- Agents append via `echo '...' >> discoveries.ndjson`
+
+**Format**: NDJSON, each line is a self-contained JSON:
+
+```jsonl
+{"ts":"...","worker":"E1","type":"code_pattern","data":{"name":"repo-pattern","file":"src/repos/Base.ts"}}
+{"ts":"...","worker":"T2","type":"integration_point","data":{"file":"src/auth/index.ts","exports":["auth"]}}
+```
+
+**Discovery Types**:
+
+| type | Dedup Key | Description |
+|------|-----------|-------------|
+| `code_pattern` | `data.name` | Reusable code pattern found |
+| `integration_point` | `data.file` | Module connection point |
+| `convention` | singleton | Code style conventions |
+| `blocker` | `data.issue` | Blocking issue encountered |
+| `tech_stack` | singleton | Project technology stack |
+| `test_command` | singleton | Test commands discovered |
+
+**Protocol Rules**:
+1. Read board before own exploration → skip covered areas
+2. Write discoveries immediately via `echo >>` → don't batch
+3. Deduplicate — check existing entries; skip if same type + dedup key exists
+4. Append-only — never modify or delete existing lines
+
+---
+
+## Error Handling
+
+| Error | Resolution |
+|-------|------------|
+| Explore agent failure | Mark as failed in explore.csv, exclude from planning |
+| All explores failed | Fallback: plan directly from requirement without exploration |
+| Execute agent failure | Mark as failed, skip dependents (cascade) |
+| Agent timeout | Mark as failed in results, continue with wave |
+| Circular dependency | Abort wave computation, report cycle |
+| CSV parse error | Validate CSV format before execution, show line number |
+| discoveries.ndjson corrupt | Ignore malformed lines, continue with valid entries |
+
+---
+
+## Core Rules
+
+1. **Wave Order is Sacred**: Never execute wave N before wave N-1 completes
+2. **CSV is Source of Truth**: Read master CSV before each wave, write after
+3. **Context via CSV**: prev_context built from CSV findings, not from memory
+4. **E* ↔ T* Linking**: tasks.csv `context_from` references explore.csv rows for cross-phase context
+5. **Skip on Failure**: Failed dep → skip dependent (cascade)
+6. **Discovery Board Append-Only**: Never clear or modify discoveries.ndjson
+7. **Explore Before Execute**: Phase 2 completes before Phase 4 starts
+8. **DO NOT STOP**: Continuous execution until all waves complete or remaining skipped
+
+---
+
+## Best Practices
+
+1. **Exploration Angles**: 1 for simple, 3-4 for complex; avoid redundant angles
+2. **Context Linking**: Link every task to at least one explore row (E*) — exploration was done for a reason
+3. **Task Granularity**: 3-10 tasks optimal; too many = overhead, too few = no parallelism
+4. **Minimize Cross-Wave Deps**: More tasks in wave 1 = more parallelism
+5. **Specific Descriptions**: Agent sees only its CSV row + prev_context — make description self-contained
+6. **Non-Overlapping Scopes**: Same-wave tasks must not write to the same files
+7. **Context From ≠ Deps**: `deps` = execution order constraint; `context_from` = information flow
+
+---
+
+## Usage Recommendations
+
+| Scenario | Recommended Approach |
+|----------|---------------------|
+| Complex feature (unclear architecture) | `workflow:wave-plan` — explore first, then plan |
+| Simple known-pattern task | `$csv-wave-pipeline` — skip exploration, direct execution |
+| Independent parallel tasks | `$csv-wave-pipeline -c 8` — single wave, max parallelism |
+| Diamond dependency (A→B,C→D) | `workflow:wave-plan` — 3 waves with context propagation |
+| Unknown codebase | `workflow:wave-plan` — exploration phase is essential |

.codex/skills/csv-wave-pipeline/SKILL.md

@@ -0,0 +1,906 @@
+---
+name: csv-wave-pipeline
+description: Requirement planning to wave-based CSV execution pipeline. Decomposes requirement into dependency-sorted CSV tasks, computes execution waves, runs wave-by-wave via spawn_agents_on_csv with cross-wave context propagation.
+argument-hint: "[-y|--yes] [-c|--concurrency N] [--continue] \"requirement description\""
+allowed-tools: spawn_agents_on_csv, Read, Write, Edit, Bash, Glob, Grep, AskUserQuestion
+---
+
+## Auto Mode
+
+When `--yes` or `-y`: Auto-confirm task decomposition, skip interactive validation, use defaults.
+
+# CSV Wave Pipeline
+
+## Usage
+
+```bash
+$csv-wave-pipeline "Implement user authentication with OAuth, JWT, and 2FA"
+$csv-wave-pipeline -c 4 "Refactor payment module with Stripe and PayPal"
+$csv-wave-pipeline -y "Build notification system with email and SMS"
+$csv-wave-pipeline --continue "auth-20260228"
+```
+
+**Flags**:
+- `-y, --yes`: Skip all confirmations (auto mode)
+- `-c, --concurrency N`: Max concurrent agents within each wave (default: 4)
+- `--continue`: Resume existing session
+
+**Output Directory**: `.workflow/.csv-wave/{session-id}/`
+**Core Output**: `tasks.csv` (master state) + `results.csv` (final) + `discoveries.ndjson` (shared exploration) + `context.md` (human-readable report)
+
+---
+
+## Overview
+
+Wave-based batch execution using `spawn_agents_on_csv` with **cross-wave context propagation**. Tasks are grouped into dependency waves; each wave executes concurrently, and its results feed into the next wave.
+
+**Core workflow**: Decompose → Compute Waves → Execute Wave-by-Wave → Aggregate
+
+```
+┌─────────────────────────────────────────────────────────────────────────┐
+│                    CSV BATCH EXECUTION WORKFLOW                          │
+├─────────────────────────────────────────────────────────────────────────┤
+│                                                                          │
+│  Phase 1: Requirement → CSV                                              │
+│     ├─ Parse requirement into subtasks (3-10 tasks)                      │
+│     ├─ Identify dependencies (deps column)                               │
+│     ├─ Compute dependency waves (topological sort → depth grouping)      │
+│     ├─ Generate tasks.csv with wave column                               │
+│     └─ User validates task breakdown (skip if -y)                        │
+│                                                                          │
+│  Phase 2: Wave Execution Engine                                          │
+│     ├─ For each wave (1..N):                                             │
+│     │   ├─ Build wave CSV (filter rows for this wave)                    │
+│     │   ├─ Inject previous wave findings into prev_context column        │
+│     │   ├─ spawn_agents_on_csv(wave CSV)                                 │
+│     │   ├─ Collect results, merge into master tasks.csv                  │
+│     │   └─ Check: any failed? → skip dependents or retry                 │
+│     └─ discoveries.ndjson shared across all waves (append-only)          │
+│                                                                          │
+│  Phase 3: Results Aggregation                                            │
+│     ├─ Export final results.csv                                          │
+│     ├─ Generate context.md with all findings                             │
+│     ├─ Display summary: completed/failed/skipped per wave                │
+│     └─ Offer: view results | retry failed | done                         │
+│                                                                          │
+└─────────────────────────────────────────────────────────────────────────┘
+```
+
+---
+
+## CSV Schema
+
+### tasks.csv (Master State)
+
+```csv
+id,title,description,test,acceptance_criteria,scope,hints,execution_directives,deps,context_from,wave,status,findings,files_modified,tests_passed,acceptance_met,error
+"1","Setup auth module","Create auth directory structure and base files","Verify directory exists and base files export expected interfaces","auth/ dir created; index.ts and types.ts export AuthProvider interface","src/auth/**","Follow monorepo module pattern || package.json;src/shared/types.ts","","","","1","","","","","",""
+"2","Implement OAuth","Add OAuth provider integration with Google and GitHub","Unit test: mock OAuth callback returns valid token; Integration test: verify redirect URL generation","OAuth login redirects to provider; callback returns JWT; supports Google and GitHub","src/auth/oauth/**","Use passport.js strategy pattern || src/auth/index.ts;docs/oauth-flow.md","Run npm test -- --grep oauth before completion","1","1","2","","","","","",""
+"3","Add JWT tokens","Implement JWT generation and validation","Unit test: sign/verify round-trip; Edge test: expired token returns 401","generateToken() returns valid JWT; verifyToken() rejects expired/tampered tokens","src/auth/jwt/**","Use jsonwebtoken library; Set default expiry 1h || src/config/auth.ts","Ensure tsc --noEmit passes","1","1","2","","","","","",""
+"4","Setup 2FA","Add TOTP-based 2FA with QR code generation","Unit test: TOTP verify with correct code; Test: QR data URL is valid","QR code generates scannable image; TOTP verification succeeds within time window","src/auth/2fa/**","Use speakeasy + qrcode libraries || src/auth/oauth/strategy.ts;src/auth/jwt/token.ts","Run full test suite: npm test","2;3","1;2;3","3","","","","","",""
+```
+
+**Columns**:
+
+| Column | Phase | Description |
+|--------|-------|-------------|
+| `id` | Input | Unique task identifier (string) |
+| `title` | Input | Short task title |
+| `description` | Input | Detailed task description — what to implement |
+| `test` | Input | Test cases: what tests to write and how to verify (unit/integration/edge) |
+| `acceptance_criteria` | Input | Acceptance criteria: measurable conditions that define "done" |
+| `scope` | Input | Target file/directory glob — constrains agent work area, prevents cross-task file conflicts |
+| `hints` | Input | Implementation tips + reference files. Format: `tips text \|\| file1;file2`. Before `\|\|` = how to implement; after `\|\|` = existing files to read before starting. Either part is optional |
+| `execution_directives` | Input | Execution constraints: commands to run for verification, tool restrictions, environment requirements |
+| `deps` | Input | Semicolon-separated dependency task IDs (empty = no deps) |
+| `context_from` | Input | Semicolon-separated task IDs whose findings this task needs |
+| `wave` | Computed | Wave number (computed by topological sort, 1-based) |
+| `status` | Output | `pending` → `completed` / `failed` / `skipped` |
+| `findings` | Output | Key discoveries or implementation notes (max 500 chars) |
+| `files_modified` | Output | Semicolon-separated file paths |
+| `tests_passed` | Output | Whether all defined test cases passed (true/false) |
+| `acceptance_met` | Output | Summary of which acceptance criteria were met/unmet |
+| `error` | Output | Error message if failed (empty if success) |
+
+### Per-Wave CSV (Temporary)
+
+Each wave generates a temporary `wave-{N}.csv` with an extra `prev_context` column:
+
+```csv
+id,title,description,test,acceptance_criteria,scope,hints,execution_directives,deps,context_from,wave,prev_context
+"2","Implement OAuth","Add OAuth integration","Unit test: mock OAuth callback returns valid token","OAuth login redirects to provider; callback returns JWT","src/auth/oauth/**","Use passport.js strategy pattern || src/auth/index.ts;docs/oauth-flow.md","Run npm test -- --grep oauth","1","1","2","[Task 1] Created auth/ with index.ts and types.ts"
+"3","Add JWT tokens","Implement JWT","Unit test: sign/verify round-trip; Edge test: expired token returns 401","generateToken() returns valid JWT; verifyToken() rejects expired/tampered tokens","src/auth/jwt/**","Use jsonwebtoken library; Set default expiry 1h || src/config/auth.ts","Ensure tsc --noEmit passes","1","1","2","[Task 1] Created auth/ with index.ts and types.ts"
+```
+
+The `prev_context` column is built from `context_from` by looking up completed tasks' `findings` in the master CSV.
+
+---
+
+## Output Artifacts
+
+| File | Purpose | Lifecycle |
+|------|---------|-----------|
+| `tasks.csv` | Master state — all tasks with status/findings | Updated after each wave |
+| `wave-{N}.csv` | Per-wave input (temporary) | Created before wave, deleted after |
+| `results.csv` | Final export of all task results | Created in Phase 3 |
+| `discoveries.ndjson` | Shared exploration board across all agents | Append-only, carries across waves |
+| `context.md` | Human-readable execution report | Created in Phase 3 |
+
+---
+
+## Session Structure
+
+```
+.workflow/.csv-wave/{session-id}/
+├── tasks.csv                  # Master state (updated per wave)
+├── results.csv                # Final results export
+├── discoveries.ndjson         # Shared discovery board (all agents)
+├── context.md                 # Human-readable report
+└── wave-{N}.csv               # Temporary per-wave input (cleaned up)
+```
+
+---
+
+## Implementation
+
+### Session Initialization
+
+```javascript
+const getUtc8ISOString = () => new Date(Date.now() + 8 * 60 * 60 * 1000).toISOString()
+
+// Parse flags
+const AUTO_YES = $ARGUMENTS.includes('--yes') || $ARGUMENTS.includes('-y')
+const continueMode = $ARGUMENTS.includes('--continue')
+const concurrencyMatch = $ARGUMENTS.match(/(?:--concurrency|-c)\s+(\d+)/)
+const maxConcurrency = concurrencyMatch ? parseInt(concurrencyMatch[1]) : 4
+
+// Clean requirement text (remove flags)
+const requirement = $ARGUMENTS
+  .replace(/--yes|-y|--continue|--concurrency\s+\d+|-c\s+\d+/g, '')
+  .trim()
+
+const slug = requirement.toLowerCase()
+  .replace(/[^a-z0-9\u4e00-\u9fa5]+/g, '-')
+  .substring(0, 40)
+const dateStr = getUtc8ISOString().substring(0, 10).replace(/-/g, '')
+const sessionId = `cwp-${slug}-${dateStr}`
+const sessionFolder = `.workflow/.csv-wave/${sessionId}`
+
+// Continue mode: find existing session
+if (continueMode) {
+  const existing = Bash(`ls -t .workflow/.csv-wave/ 2>/dev/null | head -1`).trim()
+  if (existing) {
+    sessionId = existing
+    sessionFolder = `.workflow/.csv-wave/${sessionId}`
+    // Read existing tasks.csv, find incomplete waves, resume from there
+    const existingCsv = Read(`${sessionFolder}/tasks.csv`)
+    // → jump to Phase 2 with remaining waves
+  }
+}
+
+Bash(`mkdir -p ${sessionFolder}`)
+```
+
+---
+
+### Phase 1: Requirement → CSV
+
+**Objective**: Decompose requirement into tasks, compute dependency waves, generate tasks.csv.
+
+**Steps**:
+
+1. **Decompose Requirement**
+
+   ```javascript
+   // Use ccw cli to decompose requirement into subtasks
+   Bash({
+     command: `ccw cli -p "PURPOSE: Decompose requirement into 3-10 atomic tasks for batch agent execution. Each task must include implementation description, test cases, and acceptance criteria.
+TASK:
+  • Parse requirement into independent subtasks
+  • Identify dependencies between tasks (which must complete before others)
+  • Identify context flow (which tasks need previous tasks' findings)
+  • For each task, define concrete test cases (unit/integration/edge)
+  • For each task, define measurable acceptance criteria (what defines 'done')
+  • Each task must be executable by a single agent with file read/write access
+MODE: analysis
+CONTEXT: @**/*
+EXPECTED: JSON object with tasks array. Each task: {id: string, title: string, description: string, test: string, acceptance_criteria: string, scope: string, hints: string, execution_directives: string, deps: string[], context_from: string[]}.
+  - description: what to implement (specific enough for an agent to execute independently)
+  - test: what tests to write and how to verify (e.g. 'Unit test: X returns Y; Edge test: handles Z')
+  - acceptance_criteria: measurable conditions that define done (e.g. 'API returns 200; token expires after 1h')
+  - scope: target file/directory glob (e.g. 'src/auth/**') — tasks in same wave MUST have non-overlapping scopes
+  - hints: implementation tips + reference files, format '<tips> || <ref_file1>;<ref_file2>' (e.g. 'Use strategy pattern || src/base/Strategy.ts;docs/design.md')
+  - execution_directives: commands to run for verification or tool constraints (e.g. 'Run npm test --bail; Ensure tsc passes')
+  - deps: task IDs that must complete first
+  - context_from: task IDs whose findings are needed
+CONSTRAINTS: 3-10 tasks | Each task is atomic | No circular deps | test and acceptance_criteria must be concrete and verifiable | Same-wave tasks must have non-overlapping scopes
+
+REQUIREMENT: ${requirement}" --tool gemini --mode analysis --rule planning-breakdown-task-steps`,
+     run_in_background: true
+   })
+   // Wait for CLI completion via hook callback
+   // Parse JSON from CLI output → decomposedTasks[]
+   ```
+
+2. **Compute Waves** (Topological Sort → Depth Grouping)
+
+   ```javascript
+   function computeWaves(tasks) {
+     // Build adjacency: task.deps → predecessors
+     const taskMap = new Map(tasks.map(t => [t.id, t]))
+     const inDegree = new Map(tasks.map(t => [t.id, 0]))
+     const adjList = new Map(tasks.map(t => [t.id, []]))
+
+     for (const task of tasks) {
+       for (const dep of task.deps) {
+         if (taskMap.has(dep)) {
+           adjList.get(dep).push(task.id)
+           inDegree.set(task.id, inDegree.get(task.id) + 1)
+         }
+       }
+     }
+
+     // BFS-based topological sort with depth tracking
+     const queue = []  // [taskId, depth]
+     const waveAssignment = new Map()
+
+     for (const [id, deg] of inDegree) {
+       if (deg === 0) {
+         queue.push([id, 1])
+         waveAssignment.set(id, 1)
+       }
+     }
+
+     let maxWave = 1
+     let idx = 0
+     while (idx < queue.length) {
+       const [current, depth] = queue[idx++]
+       for (const next of adjList.get(current)) {
+         const newDeg = inDegree.get(next) - 1
+         inDegree.set(next, newDeg)
+         const nextDepth = Math.max(waveAssignment.get(next) || 0, depth + 1)
+         waveAssignment.set(next, nextDepth)
+         if (newDeg === 0) {
+           queue.push([next, nextDepth])
+           maxWave = Math.max(maxWave, nextDepth)
+         }
+       }
+     }
+
+     // Detect cycles: any task without wave assignment
+     for (const task of tasks) {
+       if (!waveAssignment.has(task.id)) {
+         throw new Error(`Circular dependency detected involving task ${task.id}`)
+       }
+     }
+
+     return { waveAssignment, maxWave }
+   }
+
+   const { waveAssignment, maxWave } = computeWaves(decomposedTasks)
+   ```
+
+3. **Generate tasks.csv**
+
+   ```javascript
+   const header = 'id,title,description,test,acceptance_criteria,scope,hints,execution_directives,deps,context_from,wave,status,findings,files_modified,tests_passed,acceptance_met,error'
+   const rows = decomposedTasks.map(task => {
+     const wave = waveAssignment.get(task.id)
+     return [
+       task.id,
+       csvEscape(task.title),
+       csvEscape(task.description),
+       csvEscape(task.test),
+       csvEscape(task.acceptance_criteria),
+       csvEscape(task.scope),
+       csvEscape(task.hints),
+       csvEscape(task.execution_directives),
+       task.deps.join(';'),
+       task.context_from.join(';'),
+       wave,
+       'pending',  // status
+       '',         // findings
+       '',         // files_modified
+       '',         // tests_passed
+       '',         // acceptance_met
+       ''          // error
+     ].map(cell => `"${String(cell).replace(/"/g, '""')}"`).join(',')
+   })
+
+   Write(`${sessionFolder}/tasks.csv`, [header, ...rows].join('\n'))
+   ```
+
+4. **User Validation** (skip if AUTO_YES)
+
+   ```javascript
+   if (!AUTO_YES) {
+     // Display task breakdown with wave assignment
+     console.log(`\n## Task Breakdown (${decomposedTasks.length} tasks, ${maxWave} waves)\n`)
+     for (let w = 1; w <= maxWave; w++) {
+       const waveTasks = decomposedTasks.filter(t => waveAssignment.get(t.id) === w)
+       console.log(`### Wave ${w} (${waveTasks.length} tasks, concurrent)`)
+       waveTasks.forEach(t => console.log(`  - [${t.id}] ${t.title}`))
+     }
+
+     const answer = AskUserQuestion({
+       questions: [{
+         question: "Approve task breakdown?",
+         header: "Validation",
+         multiSelect: false,
+         options: [
+           { label: "Approve", description: "Proceed with wave execution" },
+           { label: "Modify", description: `Edit ${sessionFolder}/tasks.csv manually, then --continue` },
+           { label: "Cancel", description: "Abort" }
+         ]
+       }]
+     })  // BLOCKS
+
+     if (answer.Validation === "Modify") {
+       console.log(`Edit: ${sessionFolder}/tasks.csv\nResume: $csv-wave-pipeline --continue`)
+       return
+     } else if (answer.Validation === "Cancel") {
+       return
+     }
+   }
+   ```
+
+**Success Criteria**:
+- tasks.csv created with valid schema and wave assignments
+- No circular dependencies
+- User approved (or AUTO_YES)
+
+---
+
+### Phase 2: Wave Execution Engine
+
+**Objective**: Execute tasks wave-by-wave via `spawn_agents_on_csv`. Each wave sees previous waves' results.
+
+**Steps**:
+
+1. **Wave Loop**
+
+   ```javascript
+   const failedIds = new Set()
+   const skippedIds = new Set()
+
+   for (let wave = 1; wave <= maxWave; wave++) {
+     console.log(`\n## Wave ${wave}/${maxWave}\n`)
+
+     // 1. Read current master CSV
+     const masterCsv = parseCsv(Read(`${sessionFolder}/tasks.csv`))
+
+     // 2. Filter tasks for this wave
+     const waveTasks = masterCsv.filter(row => parseInt(row.wave) === wave)
+
+     // 3. Skip tasks whose deps failed
+     const executableTasks = []
+     for (const task of waveTasks) {
+       const deps = task.deps.split(';').filter(Boolean)
+       if (deps.some(d => failedIds.has(d) || skippedIds.has(d))) {
+         skippedIds.add(task.id)
+         // Update master CSV: mark as skipped
+         updateMasterCsvRow(sessionFolder, task.id, {
+           status: 'skipped',
+           error: 'Dependency failed or skipped'
+         })
+         console.log(`  [${task.id}] ${task.title} → SKIPPED (dependency failed)`)
+         continue
+       }
+       executableTasks.push(task)
+     }
+
+     if (executableTasks.length === 0) {
+       console.log(`  No executable tasks in wave ${wave}`)
+       continue
+     }
+
+     // 4. Build prev_context for each task
+     for (const task of executableTasks) {
+       const contextIds = task.context_from.split(';').filter(Boolean)
+       const prevFindings = contextIds
+         .map(id => {
+           const prevRow = masterCsv.find(r => r.id === id)
+           if (prevRow && prevRow.status === 'completed' && prevRow.findings) {
+             return `[Task ${id}: ${prevRow.title}] ${prevRow.findings}`
+           }
+           return null
+         })
+         .filter(Boolean)
+         .join('\n')
+       task.prev_context = prevFindings || 'No previous context available'
+     }
+
+     // 5. Write wave CSV
+     const waveHeader = 'id,title,description,test,acceptance_criteria,scope,hints,execution_directives,deps,context_from,wave,prev_context'
+     const waveRows = executableTasks.map(t =>
+       [t.id, t.title, t.description, t.test, t.acceptance_criteria, t.scope, t.hints, t.execution_directives, t.deps, t.context_from, t.wave, t.prev_context]
+         .map(cell => `"${String(cell).replace(/"/g, '""')}"`)
+         .join(',')
+     )
+     Write(`${sessionFolder}/wave-${wave}.csv`, [waveHeader, ...waveRows].join('\n'))
+
+     // 6. Execute wave
+     console.log(`  Executing ${executableTasks.length} tasks (concurrency: ${maxConcurrency})...`)
+
+     const waveResult = spawn_agents_on_csv({
+       csv_path: `${sessionFolder}/wave-${wave}.csv`,
+       id_column: "id",
+       instruction: buildInstructionTemplate(sessionFolder, wave),
+       max_concurrency: maxConcurrency,
+       max_runtime_seconds: 600,
+       output_csv_path: `${sessionFolder}/wave-${wave}-results.csv`,
+       output_schema: {
+         type: "object",
+         properties: {
+           id: { type: "string" },
+           status: { type: "string", enum: ["completed", "failed"] },
+           findings: { type: "string" },
+           files_modified: { type: "array", items: { type: "string" } },
+           tests_passed: { type: "boolean" },
+           acceptance_met: { type: "string" },
+           error: { type: "string" }
+         },
+         required: ["id", "status", "findings", "tests_passed"]
+       }
+     })
+     // ↑ Blocks until all agents in this wave complete
+
+     // 7. Merge results into master CSV
+     const waveResults = parseCsv(Read(`${sessionFolder}/wave-${wave}-results.csv`))
+     for (const result of waveResults) {
+       updateMasterCsvRow(sessionFolder, result.id, {
+         status: result.status,
+         findings: result.findings || '',
+         files_modified: (result.files_modified || []).join(';'),
+         tests_passed: String(result.tests_passed ?? ''),
+         acceptance_met: result.acceptance_met || '',
+         error: result.error || ''
+       })
+
+       if (result.status === 'failed') {
+         failedIds.add(result.id)
+         console.log(`  [${result.id}] ${result.title} → FAILED: ${result.error}`)
+       } else {
+         console.log(`  [${result.id}] ${result.title} → COMPLETED`)
+       }
+     }
+
+     // 8. Cleanup temporary wave CSV
+     Bash(`rm -f "${sessionFolder}/wave-${wave}.csv"`)
+
+     console.log(`  Wave ${wave} done: ${waveResults.filter(r => r.status === 'completed').length} completed, ${waveResults.filter(r => r.status === 'failed').length} failed`)
+   }
+   ```
+
+2. **Instruction Template Builder**
+
+   ```javascript
+   function buildInstructionTemplate(sessionFolder, wave) {
+     return `
+## TASK ASSIGNMENT
+
+### MANDATORY FIRST STEPS
+1. Read shared discoveries: ${sessionFolder}/discoveries.ndjson (if exists, skip if not)
+2. Read project context: .workflow/project-tech.json (if exists)
+
+---
+
+## Your Task
+
+**Task ID**: {id}
+**Title**: {title}
+**Description**: {description}
+**Scope**: {scope}
+
+### Implementation Hints & Reference Files
+{hints}
+
+> Format: \`<tips> || <ref_file1>;<ref_file2>\`. Read ALL reference files (after ||) before starting implementation. Apply tips (before ||) as implementation guidance.
+
+### Execution Directives
+{execution_directives}
+
+> Commands to run for verification, tool restrictions, or environment requirements. Follow these constraints during and after implementation.
+
+### Test Cases
+{test}
+
+### Acceptance Criteria
+{acceptance_criteria}
+
+### Previous Tasks' Findings (Context)
+{prev_context}
+
+---
+
+## Execution Protocol
+
+1. **Read references**: Parse {hints} — read all files listed after \`||\` to understand existing patterns
+2. **Read discoveries**: Load ${sessionFolder}/discoveries.ndjson for shared exploration findings
+3. **Use context**: Apply previous tasks' findings from prev_context above
+4. **Stay in scope**: ONLY create/modify files within {scope} — do NOT touch files outside this boundary
+5. **Apply hints**: Follow implementation tips from {hints} (before \`||\`)
+6. **Execute**: Implement the task as described
+7. **Write tests**: Implement the test cases defined above
+8. **Run directives**: Execute commands from {execution_directives} to verify your work
+9. **Verify acceptance**: Ensure all acceptance criteria are met before reporting completion
+10. **Share discoveries**: Append exploration findings to shared board:
+   \`\`\`bash
+   echo '{"ts":"<ISO8601>","worker":"{id}","type":"<type>","data":{...}}' >> ${sessionFolder}/discoveries.ndjson
+   \`\`\`
+11. **Report result**: Return JSON via report_agent_job_result
+
+### Discovery Types to Share
+- \`code_pattern\`: {name, file, description} — reusable patterns found
+- \`integration_point\`: {file, description, exports[]} — module connection points
+- \`convention\`: {naming, imports, formatting} — code style conventions
+- \`blocker\`: {issue, severity, impact} — blocking issues encountered
+
+---
+
+## Output (report_agent_job_result)
+
+Return JSON:
+{
+  "id": "{id}",
+  "status": "completed" | "failed",
+  "findings": "Key discoveries and implementation notes (max 500 chars)",
+  "files_modified": ["path1", "path2"],
+  "tests_passed": true | false,
+  "acceptance_met": "Summary of which acceptance criteria were met/unmet",
+  "error": ""
+}
+
+**IMPORTANT**: Set status to "completed" ONLY if:
+- All test cases pass
+- All acceptance criteria are met
+Otherwise set status to "failed" with details in error field.
+`
+   }
+   ```
+
+3. **Master CSV Update Helper**
+
+   ```javascript
+   function updateMasterCsvRow(sessionFolder, taskId, updates) {
+     const csvPath = `${sessionFolder}/tasks.csv`
+     const content = Read(csvPath)
+     const lines = content.split('\n')
+     const header = lines[0].split(',')
+
+     for (let i = 1; i < lines.length; i++) {
+       const cells = parseCsvLine(lines[i])
+       if (cells[0] === taskId || cells[0] === `"${taskId}"`) {
+         // Update specified columns
+         for (const [col, val] of Object.entries(updates)) {
+           const colIdx = header.indexOf(col)
+           if (colIdx >= 0) {
+             cells[colIdx] = `"${String(val).replace(/"/g, '""')}"`
+           }
+         }
+         lines[i] = cells.join(',')
+         break
+       }
+     }
+
+     Write(csvPath, lines.join('\n'))
+   }
+   ```
+
+**Success Criteria**:
+- All waves executed in order
+- Each wave's results merged into master CSV before next wave starts
+- Dependent tasks skipped when predecessor failed
+- discoveries.ndjson accumulated across all waves
+
+---
+
+### Phase 3: Results Aggregation
+
+**Objective**: Generate final results and human-readable report.
+
+**Steps**:
+
+1. **Export results.csv**
+
+   ```javascript
+   const masterCsv = Read(`${sessionFolder}/tasks.csv`)
+   // results.csv = master CSV (already has all results populated)
+   Write(`${sessionFolder}/results.csv`, masterCsv)
+   ```
+
+2. **Generate context.md**
+
+   ```javascript
+   const tasks = parseCsv(masterCsv)
+   const completed = tasks.filter(t => t.status === 'completed')
+   const failed = tasks.filter(t => t.status === 'failed')
+   const skipped = tasks.filter(t => t.status === 'skipped')
+
+   const contextContent = `# CSV Batch Execution Report
+
+**Session**: ${sessionId}
+**Requirement**: ${requirement}
+**Completed**: ${getUtc8ISOString()}
+**Waves**: ${maxWave} | **Concurrency**: ${maxConcurrency}
+
+---
+
+## Summary
+
+| Metric | Count |
+|--------|-------|
+| Total Tasks | ${tasks.length} |
+| Completed | ${completed.length} |
+| Failed | ${failed.length} |
+| Skipped | ${skipped.length} |
+| Waves | ${maxWave} |
+
+---
+
+## Wave Execution
+
+${Array.from({ length: maxWave }, (_, i) => i + 1).map(w => {
+  const waveTasks = tasks.filter(t => parseInt(t.wave) === w)
+  return `### Wave ${w}
+${waveTasks.map(t => `- **[${t.id}] ${t.title}**: ${t.status}${t.tests_passed ? ' ✓tests' : ''}${t.error ? ' — ' + t.error : ''}
+  ${t.findings ? 'Findings: ' + t.findings : ''}`).join('\n')}`
+}).join('\n\n')}
+
+---
+
+## Task Details
+
+${tasks.map(t => `### ${t.id}: ${t.title}
+
+| Field | Value |
+|-------|-------|
+| Status | ${t.status} |
+| Wave | ${t.wave} |
+| Scope | ${t.scope || 'none'} |
+| Dependencies | ${t.deps || 'none'} |
+| Context From | ${t.context_from || 'none'} |
+| Tests Passed | ${t.tests_passed || 'N/A'} |
+| Acceptance Met | ${t.acceptance_met || 'N/A'} |
+| Error | ${t.error || 'none'} |
+
+**Description**: ${t.description}
+
+**Test Cases**: ${t.test || 'N/A'}
+
+**Acceptance Criteria**: ${t.acceptance_criteria || 'N/A'}
+
+**Hints**: ${t.hints || 'N/A'}
+
+**Execution Directives**: ${t.execution_directives || 'N/A'}
+
+**Findings**: ${t.findings || 'N/A'}
+
+**Files Modified**: ${t.files_modified || 'none'}
+`).join('\n---\n')}
+
+---
+
+## All Modified Files
+
+${[...new Set(tasks.flatMap(t => (t.files_modified || '').split(';')).filter(Boolean))].map(f => '- ' + f).join('\n') || 'None'}
+`
+
+   Write(`${sessionFolder}/context.md`, contextContent)
+   ```
+
+3. **Display Summary**
+
+   ```javascript
+   console.log(`
+## Execution Complete
+
+- **Session**: ${sessionId}
+- **Waves**: ${maxWave}
+- **Completed**: ${completed.length}/${tasks.length}
+- **Failed**: ${failed.length}
+- **Skipped**: ${skipped.length}
+
+**Results**: ${sessionFolder}/results.csv
+**Report**: ${sessionFolder}/context.md
+**Discoveries**: ${sessionFolder}/discoveries.ndjson
+`)
+   ```
+
+4. **Offer Next Steps** (skip if AUTO_YES)
+
+   ```javascript
+   if (!AUTO_YES && failed.length > 0) {
+     const answer = AskUserQuestion({
+       questions: [{
+         question: `${failed.length} tasks failed. Next action?`,
+         header: "Next Step",
+         multiSelect: false,
+         options: [
+           { label: "Retry Failed", description: `Re-execute ${failed.length} failed tasks with updated context` },
+           { label: "View Report", description: "Display context.md" },
+           { label: "Done", description: "Complete session" }
+         ]
+       }]
+     })  // BLOCKS
+
+     if (answer['Next Step'] === "Retry Failed") {
+       // Reset failed tasks to pending, re-run Phase 2 for their waves
+       for (const task of failed) {
+         updateMasterCsvRow(sessionFolder, task.id, { status: 'pending', error: '' })
+       }
+       // Also reset skipped tasks whose deps are now retrying
+       for (const task of skipped) {
+         updateMasterCsvRow(sessionFolder, task.id, { status: 'pending', error: '' })
+       }
+       // Re-execute Phase 2 (loop will skip already-completed tasks)
+       // → goto Phase 2
+     } else if (answer['Next Step'] === "View Report") {
+       console.log(Read(`${sessionFolder}/context.md`))
+     }
+   }
+   ```
+
+**Success Criteria**:
+- results.csv exported
+- context.md generated
+- Summary displayed to user
+
+---
+
+## Shared Discovery Board Protocol
+
+All agents across all waves share `discoveries.ndjson`. This eliminates redundant codebase exploration.
+
+**Lifecycle**:
+- Created by the first agent to write a discovery
+- Carries over across waves — never cleared
+- Agents append via `echo '...' >> discoveries.ndjson`
+
+**Format**: NDJSON, each line is a self-contained JSON:
+
+```jsonl
+{"ts":"2026-02-28T10:00:00+08:00","worker":"1","type":"code_pattern","data":{"name":"repository-pattern","file":"src/repos/Base.ts","description":"Abstract CRUD repository"}}
+{"ts":"2026-02-28T10:01:00+08:00","worker":"2","type":"integration_point","data":{"file":"src/auth/index.ts","description":"Auth module entry","exports":["authenticate","authorize"]}}
+```
+
+**Discovery Types**:
+
+| type | Dedup Key | Description |
+|------|-----------|-------------|
+| `code_pattern` | `data.name` | Reusable code pattern found |
+| `integration_point` | `data.file` | Module connection point |
+| `convention` | singleton | Code style conventions |
+| `blocker` | `data.issue` | Blocking issue encountered |
+| `tech_stack` | singleton | Project technology stack |
+| `test_command` | singleton | Test commands discovered |
+
+**Protocol Rules**:
+1. Read board before own exploration → skip covered areas
+2. Write discoveries immediately via `echo >>` → don't batch
+3. Deduplicate — check existing entries; skip if same type + dedup key exists
+4. Append-only — never modify or delete existing lines
+
+---
+
+## Wave Computation Details
+
+### Algorithm
+
+Kahn's BFS topological sort with depth tracking:
+
+```
+Input:  tasks[] with deps[]
+Output: waveAssignment (taskId → wave number)
+
+1. Build in-degree map and adjacency list from deps
+2. Enqueue all tasks with in-degree 0 at wave 1
+3. BFS: for each dequeued task at wave W:
+   - For each dependent task D:
+     - Decrement D's in-degree
+     - D.wave = max(D.wave, W + 1)
+     - If D's in-degree reaches 0, enqueue D
+4. Any task without wave assignment → circular dependency error
+```
+
+### Wave Properties
+
+- **Wave 1**: No dependencies — all tasks in wave 1 are fully independent
+- **Wave N**: All dependencies are in waves 1..(N-1) — guaranteed completed before wave N starts
+- **Within a wave**: Tasks are independent of each other → safe for concurrent execution
+
+### Example
+
+```
+Task A (no deps)      → Wave 1
+Task B (no deps)      → Wave 1
+Task C (deps: A)      → Wave 2
+Task D (deps: A, B)   → Wave 2
+Task E (deps: C, D)   → Wave 3
+
+Execution:
+  Wave 1: [A, B]  ← concurrent
+  Wave 2: [C, D]  ← concurrent, sees A+B findings
+  Wave 3: [E]     ← sees A+B+C+D findings
+```
+
+---
+
+## Context Propagation Flow
+
+```
+Wave 1 agents:
+  ├─ Execute tasks (no prev_context)
+  ├─ Write findings to report_agent_job_result
+  └─ Append discoveries to discoveries.ndjson
+
+        ↓ merge results into master CSV
+
+Wave 2 agents:
+  ├─ Read discoveries.ndjson (exploration sharing)
+  ├─ Read prev_context column (wave 1 findings from context_from)
+  ├─ Execute tasks with full upstream context
+  ├─ Write findings to report_agent_job_result
+  └─ Append new discoveries to discoveries.ndjson
+
+        ↓ merge results into master CSV
+
+Wave 3 agents:
+  ├─ Read discoveries.ndjson (accumulated from waves 1+2)
+  ├─ Read prev_context column (wave 1+2 findings from context_from)
+  ├─ Execute tasks
+  └─ ...
+```
+
+**Two context channels**:
+1. **CSV findings** (structured): `context_from` column → `prev_context` injection — task-specific directed context
+2. **NDJSON discoveries** (broadcast): `discoveries.ndjson` — general exploration findings available to all
+
+---
+
+## Error Handling
+
+| Error | Resolution |
+|-------|------------|
+| Circular dependency | Detect in wave computation, abort with error message |
+| Agent timeout | Mark as failed in results, continue with wave |
+| Agent failed | Mark as failed, skip dependent tasks in later waves |
+| All agents in wave failed | Log error, offer retry or abort |
+| CSV parse error | Validate CSV format before execution, show line number |
+| discoveries.ndjson corrupt | Ignore malformed lines, continue with valid entries |
+| Continue mode: no session found | List available sessions, prompt user to select |
+
+---
+
+## Core Rules
+
+1. **Start Immediately**: First action is session initialization, then Phase 1
+2. **Wave Order is Sacred**: Never execute wave N before wave N-1 completes and results are merged
+3. **CSV is Source of Truth**: Master tasks.csv holds all state — always read before wave, always write after
+4. **Context Propagation**: prev_context built from master CSV, not from memory
+5. **Discovery Board is Append-Only**: Never clear, modify, or recreate discoveries.ndjson
+6. **Skip on Failure**: If a dependency failed, skip the dependent task (don't attempt)
+7. **Cleanup Temp Files**: Remove wave-{N}.csv after results are merged
+8. **DO NOT STOP**: Continuous execution until all waves complete or all remaining tasks are skipped
+
+---
+
+## Best Practices
+
+1. **Task Granularity**: 3-10 tasks optimal; too many = overhead, too few = no parallelism benefit
+2. **Minimize Cross-Wave Deps**: More tasks in wave 1 = more parallelism
+3. **Specific Descriptions**: Agent sees only its CSV row + prev_context — make description self-contained
+4. **Context From ≠ Deps**: `deps` = execution order constraint; `context_from` = information flow. A task can have `context_from` without `deps` (it just reads previous findings but doesn't require them to be done first in its wave)
+5. **Concurrency Tuning**: `-c 1` for serial execution (maximum context sharing); `-c 8` for I/O-bound tasks
+
+---
+
+## Usage Recommendations
+
+| Scenario | Recommended Approach |
+|----------|---------------------|
+| Independent parallel tasks (no deps) | `$csv-wave-pipeline -c 8` — single wave, max parallelism |
+| Linear pipeline (A→B→C) | `$csv-wave-pipeline -c 1` — 3 waves, serial, full context |
+| Diamond dependency (A→B,C→D) | `$csv-wave-pipeline` — 3 waves, B+C concurrent in wave 2 |
+| Complex requirement, unclear tasks | Use `$roadmap-with-file` first for planning, then feed issues here |
+| Single complex task | Use `$lite-execute` instead |

.codex/skills/review-cycle/SKILL.md

@@ -1,7 +1,6 @@
 ---
 name: review-cycle
 description: Unified multi-dimensional code review with automated fix orchestration. Supports session-based (git changes) and module-based (path patterns) review modes with 7-dimension parallel analysis, iterative deep-dive, and automated fix pipeline. Triggers on "workflow:review-cycle", "workflow:review-session-cycle", "workflow:review-module-cycle", "workflow:review-cycle-fix".
-allowed-tools: spawn_agent, wait, send_input, close_agent, AskUserQuestion, Read, Write, Edit, Bash, Glob, Grep
 ---
 
 # Review Cycle

.codex/skills/roadmap-with-file/SKILL.md

@@ -2,7 +2,6 @@
 name: roadmap-with-file
 description: Strategic requirement roadmap with iterative decomposition and issue creation. Outputs roadmap.md (human-readable, single source) + issues.jsonl (machine-executable). Handoff to team-planex.
 argument-hint: "[-y|--yes] [-c|--continue] [-m progressive|direct|auto] \"requirement description\""
-allowed-tools: spawn_agent, wait, send_input, close_agent, AskUserQuestion, Read, Write, Edit, Bash, Glob, Grep
 ---
 
 ## Auto Mode

.codex/skills/team-lifecycle/SKILL.md

@@ -1,8 +1,6 @@
 ---
 name: team-lifecycle
 description: Full lifecycle orchestrator - spec/impl/test. Spawn-wait-close pipeline with inline discuss subagent, shared explore cache, fast-advance, and consensus severity routing.
-agents: analyst, writer, planner, executor, tester, reviewer, architect, fe-developer, fe-qa
-phases: 5
 ---
 
 # Team Lifecycle Orchestrator

.codex/skills/team-planex/SKILL.md

@@ -0,0 +1,261 @@
+---
+name: team-planex
+description: |
+  Inline planning + delegated execution pipeline. Main flow does planning directly,
+  spawns Codex executor per issue immediately. All execution via Codex CLI only.
+---
+
+# Team PlanEx (Codex)
+
+主流程内联规划 + 委托执行。SKILL.md 自身完成规划（不再 spawn planner agent），每完成一个 issue 的 solution 后立即 spawn executor agent 并行实现，无需等待所有规划完成。
+
+## Architecture
+
+```
+┌────────────────────────────────────────┐
+│  SKILL.md (主流程 = 规划 + 节拍控制)     │
+│                                         │
+│  Phase 1: 解析输入 + 初始化 session      │
+│  Phase 2: 逐 issue 规划循环 (内联)       │
+│    ├── issue-plan → 写 solution artifact │
+│    ├── spawn executor agent ────────────┼──> [executor] 实现
+│    └── continue (不等 executor)          │
+│  Phase 3: 等待所有 executors             │
+│  Phase 4: 汇总报告                       │
+└────────────────────────────────────────┘
+```
+
+## Agent Registry
+
+| Agent | Role File | Responsibility |
+|-------|-----------|----------------|
+| `executor` | `~/.codex/agents/planex-executor.md` | Codex CLI implementation per issue |
+
+> Executor agent must be deployed to `~/.codex/agents/` before use.
+> Source: `.codex/skills/team-planex/agents/`
+
+---
+
+## Input Parsing
+
+Supported input types (parse from `$ARGUMENTS`):
+
+| Type | Detection | Handler |
+|------|-----------|---------|
+| Issue IDs | `ISS-\d{8}-\d{6}` regex | Use directly for planning |
+| Text | `--text '...'` flag | Create issue(s) first via CLI |
+| Plan file | `--plan <path>` flag | Read file, parse phases, batch create issues |
+
+### Issue Creation (when needed)
+
+For `--text` input:
+```bash
+ccw issue create --data '{"title":"<title>","description":"<description>"}' --json
+```
+
+For `--plan` input:
+- Match `## Phase N: Title`, `## Step N: Title`, or `### N. Title`
+- Each match → one issue (title + description from section content)
+- Fallback: no structure found → entire file as single issue
+
+---
+
+## Session Setup
+
+Before processing issues, initialize session directory:
+
+```javascript
+const slug = toSlug(inputDescription).slice(0, 20)
+const date = new Date().toISOString().slice(0, 10).replace(/-/g, '')
+const sessionDir = `.workflow/.team/PEX-${slug}-${date}`
+const artifactsDir = `${sessionDir}/artifacts/solutions`
+
+Bash(`mkdir -p "${artifactsDir}"`)
+
+Write({
+  file_path: `${sessionDir}/team-session.json`,
+  content: JSON.stringify({
+    session_id: `PEX-${slug}-${date}`,
+    input_type: inputType,
+    input: rawInput,
+    status: "running",
+    started_at: new Date().toISOString(),
+    executors: []
+  }, null, 2)
+})
+```
+
+---
+
+## Phase 1: Parse Input + Initialize
+
+1. Parse `$ARGUMENTS` to determine input type
+2. Create issues if needed (--text / --plan)
+3. Collect all issue IDs
+4. Initialize session directory
+
+---
+
+## Phase 2: Inline Planning Loop
+
+For each issue, execute planning inline (no planner agent):
+
+### 2a. Generate Solution via issue-plan-agent
+
+```javascript
+const planAgent = spawn_agent({
+  message: `
+## TASK ASSIGNMENT
+
+### MANDATORY FIRST STEPS (Agent Execute)
+1. **Read role definition**: ~/.codex/agents/issue-plan-agent.md (MUST read first)
+
+---
+
+issue_ids: ["${issueId}"]
+project_root: "${projectRoot}"
+
+## Requirements
+- Generate solution for this issue
+- Auto-bind single solution
+- Output solution JSON when complete
+`
+})
+
+const result = wait({ ids: [planAgent], timeout_ms: 600000 })
+close_agent({ id: planAgent })
+```
+
+### 2b. Write Solution Artifact
+
+```javascript
+const solution = parseSolution(result)
+
+Write({
+  file_path: `${artifactsDir}/${issueId}.json`,
+  content: JSON.stringify({
+    session_id: sessionId,
+    issue_id: issueId,
+    solution: solution,
+    planned_at: new Date().toISOString()
+  }, null, 2)
+})
+```
+
+### 2c. Spawn Executor Immediately
+
+```javascript
+const executorId = spawn_agent({
+  message: `
+## TASK ASSIGNMENT
+
+### MANDATORY FIRST STEPS (Agent Execute)
+1. **Read role definition**: ~/.codex/agents/planex-executor.md (MUST read first)
+
+---
+
+## Issue
+Issue ID: ${issueId}
+Solution file: ${artifactsDir}/${issueId}.json
+Session: ${sessionDir}
+
+## Execution
+Load solution from file → implement via Codex CLI → verify tests → commit → report.
+`
+})
+
+executorIds.push(executorId)
+executorIssueMap[executorId] = issueId
+```
+
+### 2d. Continue to Next Issue
+
+Do NOT wait for executor. Proceed to next issue immediately.
+
+---
+
+## Phase 3: Wait All Executors
+
+```javascript
+if (executorIds.length > 0) {
+  const execResults = wait({ ids: executorIds, timeout_ms: 1800000 })
+
+  if (execResults.timed_out) {
+    const pending = executorIds.filter(id => !execResults.status[id]?.completed)
+    if (pending.length > 0) {
+      const pendingIssues = pending.map(id => executorIssueMap[id])
+      Write({
+        file_path: `${sessionDir}/pending-executors.json`,
+        content: JSON.stringify({ pending_issues: pendingIssues, executor_ids: pending }, null, 2)
+      })
+    }
+  }
+
+  // Collect summaries
+  const summaries = executorIds.map(id => ({
+    issue_id: executorIssueMap[id],
+    status: execResults.status[id]?.completed ? 'completed' : 'timeout',
+    output: execResults.status[id]?.completed ?? null
+  }))
+
+  // Cleanup
+  executorIds.forEach(id => {
+    try { close_agent({ id }) } catch { /* already closed */ }
+  })
+}
+```
+
+---
+
+## Phase 4: Report
+
+```javascript
+const completed = summaries.filter(s => s.status === 'completed').length
+const failed = summaries.filter(s => s.status === 'timeout').length
+
+// Update session
+Write({
+  file_path: `${sessionDir}/team-session.json`,
+  content: JSON.stringify({
+    ...session,
+    status: "completed",
+    completed_at: new Date().toISOString(),
+    results: { total: executorIds.length, completed, failed }
+  }, null, 2)
+})
+
+return `
+## Pipeline Complete
+
+**Total issues**: ${executorIds.length}
+**Completed**: ${completed}
+**Timed out**: ${failed}
+
+${summaries.map(s => `- ${s.issue_id}: ${s.status}`).join('\n')}
+
+Session: ${sessionDir}
+`
+```
+
+---
+
+## User Commands
+
+During execution, the user may issue:
+
+| Command | Action |
+|---------|--------|
+| `check` / `status` | Show executor progress summary |
+| `resume` / `continue` | Urge stalled executor |
+
+---
+
+## Error Handling
+
+| Scenario | Resolution |
+|----------|------------|
+| issue-plan-agent timeout (>10 min) | Skip issue, log error, continue to next |
+| issue-plan-agent failure | Retry once, then skip with error log |
+| Solution file not written | Executor reports error, logs to `${sessionDir}/errors.json` |
+| Executor (Codex CLI) failure | Executor handles resume; logs CLI resume command |
+| No issues to process | Report error: no issues found |

.codex/skills/team-planex/agents/executor.md

@@ -23,13 +23,13 @@ completion report.
 
 | Action | Allowed |
 |--------|---------|
-| Read solution artifact from disk | ✅ |
-| Implement via Codex CLI | ✅ |
-| Run tests for verification | ✅ |
-| git commit completed work | ✅ |
-| Create or modify issues | ❌ |
-| Spawn subagents | ❌ |
-| Interact with user (AskUserQuestion) | ❌ |
+| Read solution artifact from disk | Yes |
+| Implement via Codex CLI | Yes |
+| Run tests for verification | Yes |
+| git commit completed work | Yes |
+| Create or modify issues | No |
+| Spawn subagents | No |
+| Interact with user (AskUserQuestion) | No |
 
 ---
 
@@ -38,7 +38,6 @@ completion report.
 ### Step 1: Load Context
 
 After reading role definition:
-- Run: `ccw spec load --category execution`
 - Extract issue ID, solution file path, session dir from task message
 
 ### Step 2: Load Solution
@@ -52,7 +51,7 @@ const solution = solutionData.solution
 
 If file not found or invalid:
 - Log error: `[executor] ERROR: Solution file not found: ${solutionFile}`
-- Output: `EXEC_FAILED:{issueId}:solution_file_missing`
+- Output: `EXEC_FAILED:${issueId}:solution_file_missing`
 - Stop execution
 
 Verify solution has required fields:
@@ -81,10 +80,9 @@ ${JSON.stringify(solution.bound, null, 2)}
 ## Implementation Requirements
 1. Follow the solution plan tasks in order
 2. Write clean, minimal code following existing patterns
-3. Read .workflow/specs/*.md for project conventions
-4. Run tests after each significant change
-5. Ensure all existing tests still pass
-6. Do NOT over-engineer - implement exactly what the solution specifies
+3. Run tests after each significant change
+4. Ensure all existing tests still pass
+5. Do NOT over-engineer - implement exactly what the solution specifies
 
 ## Quality Checklist
 - [ ] All solution tasks implemented
@@ -92,9 +90,6 @@ ${JSON.stringify(solution.bound, null, 2)}
 - [ ] Existing tests pass
 - [ ] New tests added where specified in solution
 - [ ] No security vulnerabilities introduced
-
-## Project Guidelines
-@.workflow/specs/*.md
 PROMPT_EOF
 )" --tool codex --mode write --id planex-${issueId}
 ```
@@ -120,7 +115,6 @@ if (testCmd) {
   const testResult = Bash(`${testCmd} 2>&1 || echo TEST_FAILED`)
 
   if (testResult.includes('TEST_FAILED') || testResult.includes('FAIL')) {
-    // Report failure with resume command
     const resumeCmd = `ccw cli -p "Fix failing tests" --resume planex-${issueId} --tool codex --mode write`
 
     Write({
@@ -179,7 +173,6 @@ EXEC_DONE:${issueId}
 If Codex CLI execution fails or times out:
 
 ```bash
-# Resume with same session ID
 ccw cli -p "Continue implementation from where stopped" \
   --resume planex-${issueId} \
   --tool codex --mode write \
@@ -194,19 +187,19 @@ Resume command is always logged to `${sessionDir}/errors.json` on any failure.
 
 | Scenario | Resolution |
 |----------|------------|
-| Solution file missing | Output `EXEC_FAILED:{id}:solution_file_missing`, stop |
-| Solution JSON malformed | Output `EXEC_FAILED:{id}:solution_invalid`, stop |
+| Solution file missing | Output `EXEC_FAILED:<id>:solution_file_missing`, stop |
+| Solution JSON malformed | Output `EXEC_FAILED:<id>:solution_invalid`, stop |
 | Issue status update fails | Log warning, continue |
-| Codex CLI failure | Log resume command to errors.json, output `EXEC_FAILED:{id}:codex_failed` |
-| Tests failing | Log test output + resume command, output `EXEC_FAILED:{id}:tests_failing` |
-| Commit fails | Log warning, still output `EXEC_DONE:{id}` (implementation complete) |
+| Codex CLI failure | Log resume command to errors.json, output `EXEC_FAILED:<id>:codex_failed` |
+| Tests failing | Log test output + resume command, output `EXEC_FAILED:<id>:tests_failing` |
+| Commit fails | Log warning, still output `EXEC_DONE:<id>` (implementation complete) |
 | No test command found | Skip test step, proceed to commit |
 
 ## Key Reminders
 
 **ALWAYS**:
-- Output `EXEC_DONE:{issueId}` on its own line when implementation succeeds
-- Output `EXEC_FAILED:{issueId}:{reason}` on its own line when implementation fails
+- Output `EXEC_DONE:<issueId>` on its own line when implementation succeeds
+- Output `EXEC_FAILED:<issueId>:<reason>` on its own line when implementation fails
 - Log resume command to errors.json on any failure
 - Use `[executor]` prefix in all status messages
 

.codex/skills/team-planex/agents/planner.md

@@ -1,183 +0,0 @@
----
-name: planex-planner
-description: |
-  PlanEx planner agent. Issue decomposition + solution design with beat protocol.
-  Outputs ISSUE_READY:{id} after each solution, waits for "Continue" signal.
-  Deploy to: ~/.codex/agents/planex-planner.md
-color: blue
----
-
-# PlanEx Planner
-
-Requirement decomposition → issue creation → solution design, one issue at a time.
-Outputs `ISSUE_READY:{issueId}` after each solution and waits for orchestrator to signal
-"Continue". Only outputs `ALL_PLANNED:{count}` when all issues are processed.
-
-## Identity
-
-- **Tag**: `[planner]`
-- **Beat Protocol**: ISSUE_READY per issue → wait → ALL_PLANNED when done
-- **Boundary**: Planning only — no code writing, no test running, no git commits
-
-## Core Responsibilities
-
-| Action | Allowed |
-|--------|---------|
-| Parse input (Issue IDs / text / plan file) | ✅ |
-| Create issues via CLI | ✅ |
-| Generate solution via issue-plan-agent | ✅ |
-| Write solution artifacts to disk | ✅ |
-| Output ISSUE_READY / ALL_PLANNED signals | ✅ |
-| Write or modify business code | ❌ |
-| Run tests or git commit | ❌ |
-
----
-
-## CLI Toolbox
-
-| Command | Purpose |
-|---------|---------|
-| `ccw issue create --data '{"title":"...","description":"..."}' --json` | Create issue |
-| `ccw issue status <id> --json` | Check issue status |
-| `ccw issue plan <id>` | Plan single issue (generates solution) |
-
----
-
-## Execution Flow
-
-### Step 1: Load Context
-
-After reading role definition, load project context:
-- Run: `ccw spec load --category planning`
-- Extract session directory and artifacts directory from task message
-
-### Step 2: Parse Input
-
-Determine input type from task message:
-
-| Detection | Condition | Action |
-|-----------|-----------|--------|
-| Issue IDs | `ISS-\d{8}-\d{6}` pattern | Use directly for planning |
-| `--text '...'` | Flag in message | Create issue(s) first via CLI |
-| `--plan <path>` | Flag in message | Read file, parse phases, batch create issues |
-
-**Plan file parsing rules** (when `--plan` is used):
-- Match `## Phase N: Title`, `## Step N: Title`, or `### N. Title`
-- Each match → one issue (title + description from section content)
-- Fallback: no structure found → entire file as single issue
-
-### Step 3: Issue Processing Loop (Beat Protocol)
-
-For each issue, execute in sequence:
-
-#### 3a. Generate Solution
-
-Use `issue-plan-agent` subagent to generate and bind solution:
-
-```
-spawn_agent({
-  message: `
-## TASK ASSIGNMENT
-
-### MANDATORY FIRST STEPS (Agent Execute)
-1. **Read role definition**: ~/.codex/agents/issue-plan-agent.md (MUST read first)
-2. Run: `ccw spec load --category planning`
-
----
-
-issue_ids: ["${issueId}"]
-project_root: "${projectRoot}"
-
-## Requirements
-- Generate solution for this issue
-- Auto-bind single solution
-- Output solution JSON when complete
-`
-})
-
-const result = wait({ ids: [agent], timeout_ms: 600000 })
-close_agent({ id: agent })
-```
-
-#### 3b. Write Solution Artifact
-
-```javascript
-// Extract solution from issue-plan-agent result
-const solution = parseSolution(result)
-
-Write({
-  file_path: `${artifactsDir}/${issueId}.json`,
-  content: JSON.stringify({
-    session_id: sessionId,
-    issue_id: issueId,
-    solution: solution,
-    planned_at: new Date().toISOString()
-  }, null, 2)
-})
-```
-
-#### 3c. Output Beat Signal
-
-Output EXACTLY (no surrounding text on this line):
-```
-ISSUE_READY:{issueId}
-```
-
-Then STOP. Do not process next issue. Wait for "Continue" message from orchestrator.
-
-### Step 4: After All Issues
-
-When every issue has been processed and confirmed with "Continue":
-
-Output EXACTLY:
-```
-ALL_PLANNED:{totalCount}
-```
-
-Where `{totalCount}` is the integer count of issues planned.
-
----
-
-## Issue Creation (when needed)
-
-For `--text` input:
-
-```bash
-ccw issue create --data '{"title":"<title>","description":"<description>"}' --json
-```
-
-Parse returned JSON for `id` field → use as issue ID.
-
-For `--plan` input, create issues one at a time:
-```bash
-# For each parsed phase/step:
-ccw issue create --data '{"title":"<phase-title>","description":"<phase-content>"}' --json
-```
-
-Collect all created issue IDs before proceeding to Step 3.
-
----
-
-## Error Handling
-
-| Scenario | Resolution |
-|----------|------------|
-| Issue creation failure | Retry once with simplified text, then report error |
-| `issue-plan-agent` failure | Retry once, then skip issue with `ISSUE_SKIP:{issueId}:reason` signal |
-| Plan file not found | Output error immediately, do not proceed |
-| Artifact write failure | Log warning inline, still output ISSUE_READY (executor will handle missing file) |
-| "Continue" not received after 5 min | Re-output `ISSUE_READY:{issueId}` once as reminder |
-
-## Key Reminders
-
-**ALWAYS**:
-- Output `ISSUE_READY:{issueId}` on its own line with no surrounding text
-- Wait after each ISSUE_READY — do NOT auto-continue
-- Write solution file before outputting ISSUE_READY
-- Use `[planner]` prefix in all status messages
-
-**NEVER**:
-- Output multiple ISSUE_READY signals before waiting for "Continue"
-- Proceed to next issue without receiving "Continue"
-- Write or modify any business logic files
-- Run tests or execute git commands

.codex/skills/team-planex/orchestrator.md

@@ -1,284 +0,0 @@
----
-name: team-planex
-description: |
-  Beat pipeline: planner decomposes requirements issue-by-issue, orchestrator spawns
-  Codex executor per issue immediately. All execution via Codex CLI only.
-agents: 2
-phases: 3
----
-
-# Team PlanEx (Codex)
-
-逐 Issue 节拍流水线。Planner 每完成一个 issue 的 solution 立即输出 `ISSUE_READY` 信号，Orchestrator 即刻 spawn 独立 Codex executor 并行实现，无需等待 planner 完成全部规划。
-
-## Architecture
-
-```
-Input (Issue IDs / --text / --plan)
-  → Orchestrator: parse input → init session → spawn planner
-  → Beat loop:
-      wait(planner) → ISSUE_READY:{issueId} → spawn_agent(executor)
-                   → send_input(planner, "Continue")
-                   → ALL_PLANNED:{count}   → close_agent(planner)
-  → wait(all executors) → report
-```
-
-## Agent Registry
-
-| Agent | Role File | Responsibility |
-|-------|-----------|----------------|
-| `planner` | `~/.codex/agents/planex-planner.md` | Issue decomp → solution design → ISSUE_READY signals |
-| `executor` | `~/.codex/agents/planex-executor.md` | Codex CLI implementation per issue |
-
-> Both agents must be deployed to `~/.codex/agents/` before use.
-> Source: `.codex/skills/team-planex/agents/`
-
----
-
-## Input Parsing
-
-Supported input types (parse from `$ARGUMENTS`):
-
-| Type | Detection | Handler |
-|------|-----------|---------|
-| Issue IDs | `ISS-\d{8}-\d{6}` regex | Pass directly to planner |
-| Text | `--text '...'` flag | Planner creates issue(s) first |
-| Plan file | `--plan <path>` flag | Planner reads file, batch creates issues |
-
----
-
-## Session Setup
-
-Before spawning agents, initialize session directory:
-
-```javascript
-// Generate session slug from input description (max 20 chars, kebab-case)
-const slug = toSlug(inputDescription).slice(0, 20)
-const date = new Date().toISOString().slice(0, 10).replace(/-/g, '')
-const sessionDir = `.workflow/.team/PEX-${slug}-${date}`
-const artifactsDir = `${sessionDir}/artifacts/solutions`
-
-Bash(`mkdir -p "${artifactsDir}"`)
-
-// Write initial session state
-Write({
-  file_path: `${sessionDir}/team-session.json`,
-  content: JSON.stringify({
-    session_id: `PEX-${slug}-${date}`,
-    input_type: inputType,
-    input: rawInput,
-    status: "running",
-    started_at: new Date().toISOString(),
-    executors: []
-  }, null, 2)
-})
-```
-
----
-
-## Phase 1: Spawn Planner
-
-```javascript
-const plannerAgent = spawn_agent({
-  message: `
-## TASK ASSIGNMENT
-
-### MANDATORY FIRST STEPS (Agent Execute)
-1. **Read role definition**: ~/.codex/agents/planex-planner.md (MUST read first)
-2. Run: `ccw spec load --category "planning execution"`
-
----
-
-## Session
-Session directory: ${sessionDir}
-Artifacts directory: ${artifactsDir}
-
-## Input
-${inputType === 'issues' ? `Issue IDs: ${issueIds.join(' ')}` : ''}
-${inputType === 'text' ? `Requirement: ${requirementText}` : ''}
-${inputType === 'plan' ? `Plan file: ${planPath}` : ''}
-
-## Beat Protocol (CRITICAL)
-Process issues one at a time. After completing each issue's solution:
-1. Write solution JSON to: ${artifactsDir}/{issueId}.json
-2. Output EXACTLY this line: ISSUE_READY:{issueId}
-3. STOP and wait — do NOT continue until you receive "Continue"
-
-When ALL issues are processed:
-1. Output EXACTLY: ALL_PLANNED:{totalCount}
-`
-})
-```
-
----
-
-## Phase 2: Beat Loop
-
-Orchestrator coordinates the planner-executor pipeline:
-
-```javascript
-const executorIds = []
-const executorIssueMap = {}
-
-while (true) {
-  // Wait for planner beat signal (up to 10 min per issue)
-  const plannerOut = wait({ ids: [plannerAgent], timeout_ms: 600000 })
-
-  // Handle timeout: urge convergence and retry
-  if (plannerOut.timed_out) {
-    send_input({
-      id: plannerAgent,
-      message: "Please output ISSUE_READY:{issueId} for current issue or ALL_PLANNED if done."
-    })
-    continue
-  }
-
-  const output = plannerOut.status[plannerAgent].completed
-
-  // Detect ALL_PLANNED — pipeline complete
-  if (output.includes('ALL_PLANNED')) {
-    const match = output.match(/ALL_PLANNED:(\d+)/)
-    const total = match ? parseInt(match[1]) : executorIds.length
-    close_agent({ id: plannerAgent })
-    break
-  }
-
-  // Detect ISSUE_READY — spawn executor immediately
-  const issueMatch = output.match(/ISSUE_READY:(ISS-\d{8}-\d{6}|[A-Z0-9-]+)/)
-  if (issueMatch) {
-    const issueId = issueMatch[1]
-    const solutionFile = `${artifactsDir}/${issueId}.json`
-
-    const executorId = spawn_agent({
-      message: `
-## TASK ASSIGNMENT
-
-### MANDATORY FIRST STEPS (Agent Execute)
-1. **Read role definition**: ~/.codex/agents/planex-executor.md (MUST read first)
-2. Run: `ccw spec load --category "planning execution"`
-
----
-
-## Issue
-Issue ID: ${issueId}
-Solution file: ${solutionFile}
-Session: ${sessionDir}
-
-## Execution
-Load solution from file → implement via Codex CLI → verify tests → commit → report.
-`
-    })
-
-    executorIds.push(executorId)
-    executorIssueMap[executorId] = issueId
-
-    // Signal planner to continue to next issue
-    send_input({ id: plannerAgent, message: "Continue with next issue." })
-    continue
-  }
-
-  // Unexpected output: urge convergence
-  send_input({
-    id: plannerAgent,
-    message: "Output ISSUE_READY:{issueId} when solution is ready, or ALL_PLANNED when all done."
-  })
-}
-```
-
----
-
-## Phase 3: Wait All Executors
-
-```javascript
-if (executorIds.length > 0) {
-  // Extended timeout: Codex CLI execution per issue (~10-20 min each)
-  const execResults = wait({ ids: executorIds, timeout_ms: 1800000 })
-
-  if (execResults.timed_out) {
-    const completed = executorIds.filter(id => execResults.status[id]?.completed)
-    const pending   = executorIds.filter(id => !execResults.status[id]?.completed)
-    // Log pending issues for manual follow-up
-    if (pending.length > 0) {
-      const pendingIssues = pending.map(id => executorIssueMap[id])
-      Write({
-        file_path: `${sessionDir}/pending-executors.json`,
-        content: JSON.stringify({ pending_issues: pendingIssues, executor_ids: pending }, null, 2)
-      })
-    }
-  }
-
-  // Collect summaries
-  const summaries = executorIds.map(id => ({
-    issue_id: executorIssueMap[id],
-    status: execResults.status[id]?.completed ? 'completed' : 'timeout',
-    output: execResults.status[id]?.completed ?? null
-  }))
-
-  // Cleanup
-  executorIds.forEach(id => {
-    try { close_agent({ id }) } catch { /* already closed */ }
-  })
-
-  // Final report
-  const completed = summaries.filter(s => s.status === 'completed').length
-  const failed    = summaries.filter(s => s.status === 'timeout').length
-
-  return `
-## Pipeline Complete
-
-**Total issues**: ${executorIds.length}
-**Completed**: ${completed}
-**Timed out**: ${failed}
-
-${summaries.map(s => `- ${s.issue_id}: ${s.status}`).join('\n')}
-
-Session: ${sessionDir}
-`
-}
-```
-
----
-
-## User Commands
-
-During execution, the user may issue:
-
-| Command | Action |
-|---------|--------|
-| `check` / `status` | Show executor progress summary |
-| `resume` / `continue` | Urge stalled planner or executor |
-| `add <issue-ids>` | `send_input` to planner with new issue IDs |
-| `add --text '...'` | `send_input` to planner to create and plan new issue |
-| `add --plan <path>` | `send_input` to planner to parse and batch create from plan file |
-
-**`add` handler** (inject mid-execution):
-
-```javascript
-// Get current planner agent ID from session state
-const session = JSON.parse(Read(`${sessionDir}/team-session.json`))
-const plannerAgentId = session.planner_agent_id  // saved during Phase 1
-
-send_input({
-  id: plannerAgentId,
-  message: `
-## NEW ISSUES INJECTED
-${newInput}
-
-Process these after current issue (or immediately if idle).
-Follow beat protocol: ISSUE_READY → wait for Continue → next issue.
-`
-})
-```
-
----
-
-## Error Handling
-
-| Scenario | Resolution |
-|----------|------------|
-| Planner timeout (>10 min per issue) | `send_input` urge convergence, re-enter loop |
-| Planner never outputs ISSUE_READY | After 3 retries, `close_agent` + report stall |
-| Solution file not written | Executor reports error, logs to `${sessionDir}/errors.json` |
-| Executor (Codex CLI) failure | Executor handles resume; logs CLI resume command |
-| ALL_PLANNED never received | After 60 min total, close planner, wait remaining executors |
-| No issues to process | AskUserQuestion for clarification |

.github/workflows/docs.yml

@@ -0,0 +1,105 @@
+name: Docs Build & Deploy
+
+on:
+  push:
+    branches:
+      - main
+      - develop
+    paths:
+      - 'docs/**'
+  pull_request:
+    branches:
+      - main
+      - develop
+    paths:
+      - 'docs/**'
+  workflow_dispatch:
+
+permissions:
+  contents: read
+  pages: write
+  id-token: write
+
+concurrency:
+  group: "pages"
+  cancel-in-progress: false
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    defaults:
+      run:
+        working-directory: docs
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: '20'
+          cache: 'npm'
+          cache-dependency-path: docs/package-lock.json
+
+      - name: Install dependencies
+        run: npm ci
+
+      - name: Build
+        run: npm run docs:build
+
+      - name: Upload artifact
+        uses: actions/upload-pages-artifact@v3
+        with:
+          path: docs/.vitepress/dist
+
+  deploy:
+    if: (github.event_name == 'push' || github.event_name == 'workflow_dispatch') && github.ref == 'refs/heads/main'
+    environment:
+      name: github-pages
+      url: ${{ steps.deployment.outputs.page_url }}
+    runs-on: ubuntu-latest
+    needs: build
+    steps:
+      - name: Deploy to GitHub Pages
+        id: deployment
+        uses: actions/deploy-pages@v4
+
+  # Lighthouse CI for PRs
+  lighthouse:
+    if: github.event_name == 'pull_request'
+    runs-on: ubuntu-latest
+    needs: build
+    defaults:
+      run:
+        working-directory: docs
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: '20'
+          cache: 'npm'
+          cache-dependency-path: docs/package-lock.json
+
+      - name: Install dependencies
+        run: npm ci
+
+      - name: Build
+        run: npm run docs:build
+
+      - name: Start preview server
+        run: |
+          npm run docs:preview -- --host 127.0.0.1 --port 4173 &
+          npx --yes wait-on http://127.0.0.1:4173/
+
+      - name: Run Lighthouse CI
+        uses: treosh/lighthouse-ci-action@v10
+        with:
+          urls: |
+            http://127.0.0.1:4173/
+          uploadArtifacts: true
+          temporaryPublicStorage: true
+          commentPR: true
+          budgetPath: docs/lighthouse-budget.json

.gitignore

@@ -138,3 +138,8 @@ ccw/.tmp-ccw-auth-home/
 
 # Skills library (local only)
 .claude/skills_lib/
+
+# Docs site
+docs/node_modules/
+docs/.vitepress/dist/
+docs/.vitepress/cache/

ccw/docs/CODEX_MCP_IMPLEMENTATION_SUMMARY.md

@@ -1,18 +1,21 @@
 # Codex MCP 功能实现总结
 
+> **注意**: 此文档描述的是旧的 vanilla JS 前端架构。当前版本 (v7.0+) 使用 React SPA 前端。
+> 请参考 `ccw/frontend/src/` 目录中的 React 组件。
+
 ## 📝 已完成的修复
 
 ### 1. CCW Tools MCP 卡片样式修复
 
-**文件**: `ccw/src/templates/dashboard-js/views/mcp-manager.js`
+**文件**: `ccw/frontend/src/components/McpManager.tsx` (React)
 
 **修改内容**:
-- ✅ 卡片边框: `border-primary` → `border-orange-500` (第345行)
-- ✅ 图标背景: `bg-primary` → `bg-orange-500` (第348行)
-- ✅ 图标颜色: `text-primary-foreground` → `text-white` (第349行)
-- ✅ "Available"徽章: `bg-primary/20 text-primary` → `bg-orange-500/20 text-orange-600` (第360行)
-- ✅ 选择按钮颜色: `text-primary` → `text-orange-500` (第378-379行)
-- ✅ 安装按钮: `bg-primary` → `bg-orange-500` (第386行、第399行)
+- ✅ 卡片边框: `border-primary` → `border-orange-500`
+- ✅ 图标背景: `bg-primary` → `bg-orange-500`
+- ✅ 图标颜色: `text-primary-foreground` → `text-white`
+- ✅ "Available"徽章: `bg-primary/20 text-primary` → `bg-orange-500/20 text-orange-600`
+- ✅ 选择按钮颜色: `text-primary` → `text-orange-500`
+- ✅ 安装按钮: `bg-primary` → `bg-orange-500`
 
 **影响范围**: Claude 模式下的 CCW Tools MCP 卡片
 
@@ -20,10 +23,10 @@
 
 ### 2. Toast 消息显示时间增强
 
-**文件**: `ccw/src/templates/dashboard-js/components/navigation.js`
+**文件**: `ccw/frontend/src/hooks/useToast.ts` (React)
 
 **修改内容**:
-- ✅ 显示时间: 2000ms → 3500ms (第300行)
+- ✅ 显示时间: 2000ms → 3500ms
 
 **影响范围**: 所有 Toast 消息（MCP 安装、删除、切换等操作反馈）
 
@@ -55,38 +58,33 @@ API 请求: POST /api/codex-mcp-add
     ↓
 前端更新:
   1. loadMcpConfig() - 重新加载配置
-  2. renderMcpManager() - 重新渲染 UI
-  3. showRefreshToast(...) - 显示成功/失败消息 (3.5秒)
+  2. 状态更新触发 UI 重新渲染
+  3. Toast 显示成功/失败消息 (3.5秒)
 ```
 
 ---
 
 ## 📍 关键代码位置
 
-### 前端
+### 前端 (React SPA)
 
-| 功能 | 文件 | 行号 | 说明 |
-|------|------|------|------|
-| 复制到 Codex | `components/mcp-manager.js` | 175-177 | `copyClaudeServerToCodex()` 函数 |
-| 添加到 Codex | `components/mcp-manager.js` | 87-114 | `addCodexMcpServer()` 函数 |
-| Toast 消息 | `components/navigation.js` | 286-301 | `showRefreshToast()` 函数 |
-| CCW Tools 样式 | `views/mcp-manager.js` | 342-415 | Claude 模式卡片渲染 |
-| 其他项目按钮 | `views/mcp-manager.js` | 1015-1020 | "Install to Codex" 按钮 |
+| 功能 | 文件 | 说明 |
+|------|------|------|
+| MCP 管理 | `ccw/frontend/src/components/McpManager.tsx` | MCP 管理组件 |
+| Toast 消息 | `ccw/frontend/src/hooks/useToast.ts` | Toast hook |
+| 复制到 Codex | `ccw/frontend/src/api/mcp.ts` | MCP API 调用 |
 
 ### 后端
 
-| 功能 | 文件 | 行号 | 说明 |
-|------|------|------|------|
-| API 端点 | `core/routes/mcp-routes.ts` | 1001-1010 | `/api/codex-mcp-add` 路由 |
-| 添加服务器 | `core/routes/mcp-routes.ts` | 251-330 | `addCodexMcpServer()` 函数 |
-| TOML 序列化 | `core/routes/mcp-routes.ts` | 166-188 | `serializeToml()` 函数 |
+| 功能 | 文件 | 说明 |
+|------|------|------|
+| API 端点 | `ccw/src/core/routes/mcp-routes.ts` | `/api/codex-mcp-add` 路由 |
+| 添加服务器 | `ccw/src/core/routes/mcp-routes.ts` | `addCodexMcpServer()` 函数 |
+| TOML 序列化 | `ccw/src/core/routes/mcp-routes.ts` | `serializeToml()` 函数 |
 
-### CSS
+### CSS (Tailwind)
 
-| 功能 | 文件 | 行号 | 说明 |
-|------|------|------|------|
-| Toast 样式 | `dashboard-css/06-cards.css` | 1501-1538 | Toast 容器和类型样式 |
-| Toast 动画 | `dashboard-css/06-cards.css` | 1540-1551 | 滑入/淡出动画 |
+Toast 样式使用 Tailwind CSS 内联样式，定义在 React 组件中。
 
 ---
 
@@ -220,33 +218,31 @@ API 请求: POST /api/codex-mcp-add
 
 ## 📦 相关文件清单
 
-### 已修改文件
+### 前端文件 (React SPA)
 
-1. `ccw/src/templates/dashboard-js/views/mcp-manager.js`
-   - 修改: CCW Tools 卡片样式（第342-415行）
-   
-2. `ccw/src/templates/dashboard-js/components/navigation.js`
-   - 修改: Toast 显示时间（第300行）
+1. `ccw/frontend/src/components/McpManager.tsx`
+   - MCP 管理组件（包含 CCW Tools 卡片样式）
 
-### 核心功能文件（未修改但相关）
+2. `ccw/frontend/src/hooks/useToast.ts`
+   - Toast 消息 hook（显示时间 3.5秒）
 
-3. `ccw/src/templates/dashboard-js/components/mcp-manager.js`
-   - 包含: `addCodexMcpServer()`, `copyClaudeServerToCodex()` 函数
+3. `ccw/frontend/src/api/mcp.ts`
+   - MCP API 调用函数
+
+### 后端文件
 
 4. `ccw/src/core/routes/mcp-routes.ts`
-   - 包含: Codex MCP API 端点和后端逻辑
+   - Codex MCP API 端点和后端逻辑
 
-5. `ccw/src/templates/dashboard-css/06-cards.css`
-   - 包含: Toast 样式定义
+### 文档
 
-### 新增文档
-
-6. `ccw/docs/CODEX_MCP_TESTING_GUIDE.md`
+5. `ccw/docs/CODEX_MCP_TESTING_GUIDE.md`
    - 详细测试指南
 
-7. `ccw/docs/QUICK_TEST_CODEX_MCP.md`
+6. `ccw/docs/QUICK_TEST_CODEX_MCP.md`
    - 快速测试步骤
 
+
 8. `ccw/docs/CODEX_MCP_IMPLEMENTATION_SUMMARY.md`
    - 本文档
 

ccw/docs/CODEX_MCP_TESTING_GUIDE.md

@@ -277,7 +277,7 @@ _____
 ### Toast 消息机制
 
 **实现位置**:
-- `ccw/src/templates/dashboard-js/components/navigation.js:286-301`
+- `ccw/frontend/src/hooks/useToast.ts` (React)
 - 显示时间：3500ms (3.5秒)
 - 淡出动画：300ms
 

ccw/docs/GRAPH_EXPLORER_FIX.md

@@ -218,8 +218,8 @@ To verify the fix works:
 ## Related Files
 
 - **Implementation**: `ccw/src/core/routes/graph-routes.ts`
-- **Frontend**: `ccw/src/templates/dashboard-js/views/graph-explorer.js`
-- **Styles**: `ccw/src/templates/dashboard-css/14-graph-explorer.css`
+- **Frontend**: `ccw/frontend/src/components/GraphExplorer.tsx` (React SPA)
+- **Styles**: Embedded in React components
 - **API Docs**: `ccw/src/core/routes/graph-routes.md`
 - **Migration**: `codex-lens/src/codexlens/storage/migrations/migration_005_cleanup_unused_fields.py`
 

ccw/docs/GRAPH_EXPLORER_TROUBLESHOOTING.md

@@ -238,23 +238,22 @@ ccw view
    rg "handleGraphRoutes" src/
    ```
 
-2. **检查前端是否包含 graph-explorer 视图**：
+2. **检查前端是否包含 Graph Explorer 组件**：
    ```bash
-   ls src/templates/dashboard-js/views/graph-explorer.js
+   ls ccw/frontend/src/components/GraphExplorer.tsx
    ```
 
-3. **检查 dashboard-generator.ts 是否包含 graph explorer**：
+3. **检查 React 前端是否正确构建**：
    ```bash
-   rg "graph-explorer" src/core/dashboard-generator.ts
+   ls ccw/frontend/dist/index.html
    ```
 
 ### 解决方案
 
 确保以下文件存在且正确：
-- `src/core/routes/graph-routes.ts` - API 路由处理
-- `src/templates/dashboard-js/views/graph-explorer.js` - 前端视图
-- `src/templates/dashboard-css/14-graph-explorer.css` - 样式
-- `src/templates/dashboard.html` - 包含 Graph 导航项（line 334）
+- `ccw/src/core/routes/graph-routes.ts` - API 路由处理
+- `ccw/frontend/src/components/GraphExplorer.tsx` - React 前端组件
+- `ccw/frontend/dist/index.html` - 构建后的前端入口
 
 ---
 

ccw/frontend/src/App.tsx

@@ -53,13 +53,18 @@ function QueryInvalidator() {
   const registerQueryInvalidator = useWorkflowStore((state) => state.registerQueryInvalidator);
 
   useEffect(() => {
-    // Register callback to invalidate all 'workspace' prefixed queries
+    // Register callback to invalidate all workspace-related queries on workspace switch
     const callback = () => {
       queryClient.invalidateQueries({
         predicate: (query) => {
           const queryKey = query.queryKey;
-          // Check if the first element of the query key is 'workspace'
-          return Array.isArray(queryKey) && queryKey[0] === 'workspace';
+          if (!Array.isArray(queryKey)) return false;
+          const prefix = queryKey[0];
+          // Invalidate all query families that depend on workspace data
+          return prefix === 'workspace'
+            || prefix === 'projectOverview'
+            || prefix === 'workflowStatusCounts'
+            || prefix === 'dashboardStats';
         },
       });
     };

ccw/frontend/src/components/dashboard/widgets/RecentSessionsWidget.tsx

@@ -69,6 +69,7 @@ const statusIcons: Record<string, React.ElementType> = {
   cancelled: XCircle,
   idle: Clock,
   initializing: Loader2,
+  ready: CheckCircle2,
 };
 
 // Status color mapping
@@ -83,6 +84,7 @@ const statusColors: Record<string, string> = {
   cancelled: 'bg-destructive/20 text-destructive border-destructive/30',
   idle: 'bg-muted text-muted-foreground border-border',
   initializing: 'bg-info/20 text-info border-info/30',
+  ready: 'bg-success/20 text-success border-success/30',
 };
 
 // Status to i18n key mapping
@@ -97,6 +99,7 @@ const statusI18nKeys: Record<string, string> = {
   cancelled: 'cancelled',
   idle: 'idle',
   initializing: 'initializing',
+  ready: 'ready',
 };
 
 // Lite task sub-type icons

ccw/frontend/src/components/dashboard/widgets/WorkflowTaskWidget.tsx

@@ -119,6 +119,10 @@ const sessionStatusColors: Record<string, { bg: string; text: string }> = {
   in_progress: { bg: 'bg-warning/20', text: 'text-warning' },
   completed: { bg: 'bg-success/20', text: 'text-success' },
   paused: { bg: 'bg-slate-400/20', text: 'text-slate-500' },
+  ready: { bg: 'bg-success/20', text: 'text-success' },
+  initialized: { bg: 'bg-info/20', text: 'text-info' },
+  archived: { bg: 'bg-slate-300/20', text: 'text-slate-400' },
+  failed: { bg: 'bg-destructive/20', text: 'text-destructive' },
 };
 
 // ---- Mini Stat Card with Sparkline ----

ccw/frontend/src/components/issue/discovery/DiscoveryCard.tsx

@@ -21,17 +21,17 @@ const statusConfig = {
   running: {
     icon: Clock,
     variant: 'warning' as const,
-    label: 'issues.discovery.status.running',
+    label: 'issues.discovery.session.status.running',
   },
   completed: {
     icon: CheckCircle,
     variant: 'success' as const,
-    label: 'issues.discovery.status.completed',
+    label: 'issues.discovery.session.status.completed',
   },
   failed: {
     icon: XCircle,
     variant: 'destructive' as const,
-    label: 'issues.discovery.status.failed',
+    label: 'issues.discovery.session.status.failed',
   },
 };
 
@@ -79,7 +79,7 @@ export function DiscoveryCard({ session, isActive, onClick }: DiscoveryCardProps
       <div className="flex items-center justify-between text-sm">
         <div className="flex items-center gap-4">
           <div className="flex items-center gap-1">
-            <span className="text-muted-foreground">{formatMessage({ id: 'issues.discovery.findings' })}:</span>
+            <span className="text-muted-foreground">{formatMessage({ id: 'issues.discovery.session.findings' }, { count: session.findings_count })}:</span>
             <span className="font-medium text-foreground">{session.findings_count}</span>
           </div>
         </div>

ccw/frontend/src/components/issue/discovery/DiscoveryDetail.tsx

@@ -12,6 +12,7 @@ import { Tabs, TabsList, TabsTrigger, TabsContent } from '@/components/ui/Tabs';
 import { Badge } from '@/components/ui/Badge';
 import { Progress } from '@/components/ui/Progress';
 import { IssueDrawer } from '@/components/issue/hub/IssueDrawer';
+import { FindingDrawer } from './FindingDrawer';
 import type { DiscoverySession, Finding } from '@/lib/api';
 import type { Issue } from '@/lib/api';
 import type { FindingFilters } from '@/hooks/useIssues';
@@ -43,6 +44,7 @@ export function DiscoveryDetail({
   const { formatMessage } = useIntl();
   const [activeTab, setActiveTab] = useState('findings');
   const [selectedIssue, setSelectedIssue] = useState<Issue | null>(null);
+  const [selectedFinding, setSelectedFinding] = useState<Finding | null>(null);
   const [selectedIds, setSelectedIds] = useState<string[]>([]);
 
   const handleFindingClick = (finding: Finding) => {
@@ -51,14 +53,21 @@ export function DiscoveryDetail({
       const relatedIssue = issues.find(i => i.id === finding.issue_id);
       if (relatedIssue) {
         setSelectedIssue(relatedIssue);
+        return;
       }
     }
+    // Otherwise, show the finding details in FindingDrawer
+    setSelectedFinding(finding);
   };
 
-  const handleCloseDrawer = () => {
+  const handleCloseIssueDrawer = () => {
     setSelectedIssue(null);
   };
 
+  const handleCloseFindingDrawer = () => {
+    setSelectedFinding(null);
+  };
+
   const handleExportSelected = async () => {
     if (onExportSelected && selectedIds.length > 0) {
       await onExportSelected(selectedIds);
@@ -130,7 +139,7 @@ export function DiscoveryDetail({
         <Badge
           variant={session.status === 'completed' ? 'success' : session.status === 'failed' ? 'destructive' : 'warning'}
         >
-          {formatMessage({ id: `issues.discovery.status.${session.status}` })}
+          {formatMessage({ id: `issues.discovery.session.status.${session.status}` })}
         </Badge>
         <span className="text-sm text-muted-foreground">
           {formatMessage({ id: 'issues.discovery.createdAt' })}: {formatDate(session.created_at)}
@@ -192,7 +201,7 @@ export function DiscoveryDetail({
                   <Badge
                     variant={severity === 'critical' || severity === 'high' ? 'destructive' : severity === 'medium' ? 'warning' : 'secondary'}
                   >
-                    {formatMessage({ id: `issues.discovery.severity.${severity}` })}
+                    {formatMessage({ id: `issues.discovery.findings.severity.${severity}` })}
                   </Badge>
                   <span className="font-medium">{count}</span>
                 </div>
@@ -240,7 +249,7 @@ export function DiscoveryDetail({
               <Badge
                 variant={session.status === 'completed' ? 'success' : session.status === 'failed' ? 'destructive' : 'warning'}
               >
-                {formatMessage({ id: `issues.discovery.status.${session.status}` })}
+                {formatMessage({ id: `issues.discovery.session.status.${session.status}` })}
               </Badge>
             </div>
             <div>
@@ -277,7 +286,14 @@ export function DiscoveryDetail({
       <IssueDrawer
         issue={selectedIssue}
         isOpen={selectedIssue !== null}
-        onClose={handleCloseDrawer}
+        onClose={handleCloseIssueDrawer}
+      />
+
+      {/* Finding Detail Drawer */}
+      <FindingDrawer
+        finding={selectedFinding}
+        isOpen={selectedFinding !== null}
+        onClose={handleCloseFindingDrawer}
       />
     </div>
   );

ccw/frontend/src/components/issue/discovery/FindingDrawer.tsx

@@ -0,0 +1,213 @@
+// ========================================
+// FindingDrawer Component
+// ========================================
+// Right-side finding detail drawer for displaying discovery finding details
+
+import { useEffect } from 'react';
+import { useIntl } from 'react-intl';
+import { X, FileText, AlertTriangle, ExternalLink, MapPin, Code, Lightbulb, Target } from 'lucide-react';
+import { Badge } from '@/components/ui/Badge';
+import { Button } from '@/components/ui/Button';
+import { cn } from '@/lib/utils';
+import type { Finding } from '@/lib/api';
+
+// ========== Types ==========
+export interface FindingDrawerProps {
+  finding: Finding | null;
+  isOpen: boolean;
+  onClose: () => void;
+}
+
+// ========== Severity Configuration ==========
+const severityConfig: Record<string, { label: string; variant: 'default' | 'secondary' | 'destructive' | 'outline' | 'success' | 'warning' | 'info' }> = {
+  critical: { label: 'issues.discovery.findings.severity.critical', variant: 'destructive' },
+  high: { label: 'issues.discovery.findings.severity.high', variant: 'destructive' },
+  medium: { label: 'issues.discovery.findings.severity.medium', variant: 'warning' },
+  low: { label: 'issues.discovery.findings.severity.low', variant: 'secondary' },
+};
+
+function getSeverityConfig(severity: string) {
+  return severityConfig[severity] || { label: 'issues.discovery.findings.severity.unknown', variant: 'outline' };
+}
+
+// ========== Component ==========
+
+export function FindingDrawer({ finding, isOpen, onClose }: FindingDrawerProps) {
+  const { formatMessage } = useIntl();
+
+  // ESC key to close
+  useEffect(() => {
+    if (!isOpen) return;
+    const handleEsc = (e: KeyboardEvent) => {
+      if (e.key === 'Escape') onClose();
+    };
+    window.addEventListener('keydown', handleEsc);
+    return () => window.removeEventListener('keydown', handleEsc);
+  }, [isOpen, onClose]);
+
+  if (!finding || !isOpen) {
+    return null;
+  }
+
+  const severity = getSeverityConfig(finding.severity);
+
+  return (
+    <>
+      {/* Overlay */}
+      <div
+        className={cn(
+          'fixed inset-0 bg-black/40 transition-opacity z-40',
+          isOpen ? 'opacity-100' : 'opacity-0 pointer-events-none'
+        )}
+        onClick={onClose}
+        aria-hidden="true"
+      />
+
+      {/* Drawer */}
+      <div
+        className={cn(
+          'fixed top-0 right-0 h-full w-1/2 bg-background border-l border-border shadow-2xl z-50 flex flex-col transition-transform duration-300 ease-in-out',
+          isOpen ? 'translate-x-0' : 'translate-x-full'
+        )}
+        role="dialog"
+        aria-modal="true"
+        style={{ minWidth: '400px', maxWidth: '800px' }}
+      >
+        {/* Header */}
+        <div className="flex items-start justify-between p-6 border-b border-border bg-card">
+          <div className="flex-1 min-w-0 mr-4">
+            <div className="flex items-center gap-2 mb-2 flex-wrap">
+              <span className="text-xs font-mono text-muted-foreground">{finding.id}</span>
+              <Badge variant={severity.variant}>
+                {formatMessage({ id: severity.label })}
+              </Badge>
+              {finding.type && (
+                <Badge variant="outline">{finding.type}</Badge>
+              )}
+              {finding.category && (
+                <Badge variant="info">{finding.category}</Badge>
+              )}
+            </div>
+            <h2 className="text-lg font-semibold text-foreground">
+              {finding.title}
+            </h2>
+          </div>
+          <Button variant="ghost" size="icon" onClick={onClose} className="flex-shrink-0 hover:bg-secondary">
+            <X className="h-5 w-5" />
+          </Button>
+        </div>
+
+        {/* Content */}
+        <div className="flex-1 overflow-y-auto p-6 space-y-6">
+          {/* Description */}
+          <div>
+            <h3 className="text-sm font-semibold text-foreground mb-2 flex items-center gap-2">
+              <AlertTriangle className="h-4 w-4" />
+              {formatMessage({ id: 'issues.discovery.findings.description' })}
+            </h3>
+            <p className="text-sm text-muted-foreground whitespace-pre-wrap">
+              {finding.description}
+            </p>
+          </div>
+
+          {/* File Location */}
+          {finding.file && (
+            <div>
+              <h3 className="text-sm font-semibold text-foreground mb-2 flex items-center gap-2">
+                <MapPin className="h-4 w-4" />
+                {formatMessage({ id: 'issues.discovery.findings.location' })}
+              </h3>
+              <div className="flex items-center gap-2 text-sm">
+                <FileText className="h-4 w-4 text-muted-foreground" />
+                <code className="px-2 py-1 bg-muted rounded text-xs">
+                  {finding.file}
+                  {finding.line && `:${finding.line}`}
+                </code>
+              </div>
+            </div>
+          )}
+
+          {/* Code Snippet */}
+          {finding.code_snippet && (
+            <div>
+              <h3 className="text-sm font-semibold text-foreground mb-2 flex items-center gap-2">
+                <Code className="h-4 w-4" />
+                {formatMessage({ id: 'issues.discovery.findings.codeSnippet' })}
+              </h3>
+              <pre className="p-3 bg-muted rounded-md overflow-x-auto text-xs border border-border">
+                <code>{finding.code_snippet}</code>
+              </pre>
+            </div>
+          )}
+
+          {/* Suggested Fix */}
+          {finding.suggested_issue && (
+            <div>
+              <h3 className="text-sm font-semibold text-foreground mb-2 flex items-center gap-2">
+                <Lightbulb className="h-4 w-4" />
+                {formatMessage({ id: 'issues.discovery.findings.suggestedFix' })}
+              </h3>
+              <p className="text-sm text-muted-foreground whitespace-pre-wrap">
+                {finding.suggested_issue}
+              </p>
+            </div>
+          )}
+
+          {/* Confidence */}
+          {finding.confidence !== undefined && (
+            <div>
+              <h3 className="text-sm font-semibold text-foreground mb-2 flex items-center gap-2">
+                <Target className="h-4 w-4" />
+                {formatMessage({ id: 'issues.discovery.findings.confidence' })}
+              </h3>
+              <div className="flex items-center gap-2">
+                <div className="flex-1 h-2 bg-muted rounded-full overflow-hidden">
+                  <div
+                    className={cn(
+                      "h-full transition-all",
+                      finding.confidence >= 0.9 ? "bg-green-500" :
+                      finding.confidence >= 0.7 ? "bg-yellow-500" : "bg-red-500"
+                    )}
+                    style={{ width: `${finding.confidence * 100}%` }}
+                  />
+                </div>
+                <span className="text-sm font-medium">
+                  {Math.round(finding.confidence * 100)}%
+                </span>
+              </div>
+            </div>
+          )}
+
+          {/* Reference */}
+          {finding.reference && (
+            <div>
+              <h3 className="text-sm font-semibold text-foreground mb-2 flex items-center gap-2">
+                <ExternalLink className="h-4 w-4" />
+                {formatMessage({ id: 'issues.discovery.findings.reference' })}
+              </h3>
+              <a
+                href={finding.reference}
+                target="_blank"
+                rel="noopener noreferrer"
+                className="text-sm text-primary hover:underline break-all"
+              >
+                {finding.reference}
+              </a>
+            </div>
+          )}
+
+          {/* Perspective */}
+          {finding.perspective && (
+            <div className="pt-4 border-t border-border">
+              <Badge variant="secondary" className="text-xs">
+                {formatMessage({ id: 'issues.discovery.findings.perspective' })}: {finding.perspective}
+              </Badge>
+            </div>
+          )}
+        </div>
+      </div>
+    </>
+  );
+}
+
+export default FindingDrawer;

ccw/frontend/src/components/issue/discovery/FindingList.tsx

@@ -24,14 +24,14 @@ interface FindingListProps {
 }
 
 const severityConfig: Record<string, { variant: 'destructive' | 'warning' | 'secondary' | 'outline' | 'success' | 'info' | 'default'; label: string }> = {
-  critical: { variant: 'destructive', label: 'issues.discovery.severity.critical' },
-  high: { variant: 'destructive', label: 'issues.discovery.severity.high' },
-  medium: { variant: 'warning', label: 'issues.discovery.severity.medium' },
-  low: { variant: 'secondary', label: 'issues.discovery.severity.low' },
+  critical: { variant: 'destructive', label: 'issues.discovery.findings.severity.critical' },
+  high: { variant: 'destructive', label: 'issues.discovery.findings.severity.high' },
+  medium: { variant: 'warning', label: 'issues.discovery.findings.severity.medium' },
+  low: { variant: 'secondary', label: 'issues.discovery.findings.severity.low' },
 };
 
 function getSeverityConfig(severity: string) {
-  return severityConfig[severity] || { variant: 'outline', label: 'issues.discovery.severity.unknown' };
+  return severityConfig[severity] || { variant: 'outline', label: 'issues.discovery.findings.severity.unknown' };
 }
 
 export function FindingList({
@@ -116,10 +116,10 @@ export function FindingList({
           </SelectTrigger>
           <SelectContent>
             <SelectItem value="all">{formatMessage({ id: 'issues.discovery.findings.severity.all' })}</SelectItem>
-            <SelectItem value="critical">{formatMessage({ id: 'issues.discovery.severity.critical' })}</SelectItem>
-            <SelectItem value="high">{formatMessage({ id: 'issues.discovery.severity.high' })}</SelectItem>
-            <SelectItem value="medium">{formatMessage({ id: 'issues.discovery.severity.medium' })}</SelectItem>
-            <SelectItem value="low">{formatMessage({ id: 'issues.discovery.severity.low' })}</SelectItem>
+            <SelectItem value="critical">{formatMessage({ id: 'issues.discovery.findings.severity.critical' })}</SelectItem>
+            <SelectItem value="high">{formatMessage({ id: 'issues.discovery.findings.severity.high' })}</SelectItem>
+            <SelectItem value="medium">{formatMessage({ id: 'issues.discovery.findings.severity.medium' })}</SelectItem>
+            <SelectItem value="low">{formatMessage({ id: 'issues.discovery.findings.severity.low' })}</SelectItem>
           </SelectContent>
         </Select>
         {uniqueTypes.length > 0 && (

ccw/frontend/src/hooks/useProjectOverview.ts

@@ -38,7 +38,7 @@ export function useProjectOverview(options: UseProjectOverviewOptions = {}) {
   const queryEnabled = enabled && !!projectPath;
 
   const query = useQuery({
-    queryKey: projectOverviewKeys.detail(),
+    queryKey: projectOverviewKeys.detail(projectPath),
     queryFn: () => fetchProjectOverview(projectPath),
     staleTime,
     enabled: queryEnabled,

ccw/frontend/src/lib/api.ts

@@ -1077,6 +1077,12 @@ export interface Finding {
   created_at: string;
   issue_id?: string; // Associated issue ID if exported
   exported?: boolean; // Whether this finding has been exported as an issue
+  // Additional fields from discovery backend
+  category?: string;
+  suggested_issue?: string;
+  confidence?: number;
+  reference?: string;
+  perspective?: string;
 }
 
 export async function fetchDiscoveries(projectPath?: string): Promise<DiscoverySession[]> {
@@ -1131,7 +1137,11 @@ export async function fetchDiscoveryFindings(
     ? `/api/discoveries/${encodeURIComponent(sessionId)}/findings?path=${encodeURIComponent(projectPath)}`
     : `/api/discoveries/${encodeURIComponent(sessionId)}/findings`;
   const data = await fetchApi<{ findings?: Finding[] }>(url);
-  return data.findings ?? [];
+  // Map backend 'priority' to frontend 'severity' for compatibility
+  return (data.findings ?? []).map(f => ({
+    ...f,
+    severity: f.severity || (f as any).priority || 'medium'
+  }));
 }
 
 /**

ccw/frontend/src/locales/en/common.json

@@ -82,7 +82,8 @@
     "label": "Status",
     "openIssues": "Open Issues",
     "enabled": "Enabled",
-    "disabled": "Disabled"
+    "disabled": "Disabled",
+    "ready": "Ready"
   },
   "priority": {
     "low": "Low",

ccw/frontend/src/locales/en/issues.json

@@ -353,7 +353,14 @@
       "hasIssue": "Linked",
       "export": "Export",
       "selectAll": "Select All",
-      "deselectAll": "Deselect All"
+      "deselectAll": "Deselect All",
+      "description": "Description",
+      "location": "Location",
+      "codeSnippet": "Code Snippet",
+      "suggestedFix": "Suggested Fix",
+      "confidence": "Confidence",
+      "reference": "Reference",
+      "perspective": "Perspective"
     },
     "tabs": {
       "findings": "Findings",

ccw/frontend/src/locales/zh/common.json

@@ -82,7 +82,8 @@
     "label": "状态",
     "openIssues": "开放问题",
     "enabled": "已启用",
-    "disabled": "已禁用"
+    "disabled": "已禁用",
+    "ready": "就绪"
   },
   "priority": {
     "low": "低",

ccw/frontend/src/locales/zh/issues.json

@@ -353,7 +353,14 @@
       "hasIssue": "已关联",
       "export": "导出",
       "selectAll": "全选",
-      "deselectAll": "取消全选"
+      "deselectAll": "取消全选",
+      "description": "问题描述",
+      "location": "文件位置",
+      "codeSnippet": "代码片段",
+      "suggestedFix": "建议修复",
+      "confidence": "置信度",
+      "reference": "参考链接",
+      "perspective": "视角"
     },
     "tabs": {
       "findings": "发现",

ccw/frontend/src/pages/DiscoveryPage.tsx

@@ -7,7 +7,7 @@ import { useIntl } from 'react-intl';
 import { Radar, AlertCircle, Loader2 } from 'lucide-react';
 import { Card } from '@/components/ui/Card';
 import { Badge } from '@/components/ui/Badge';
-import { useIssueDiscovery } from '@/hooks/useIssues';
+import { useIssueDiscovery, useIssues } from '@/hooks/useIssues';
 import { DiscoveryCard } from '@/components/issue/discovery/DiscoveryCard';
 import { DiscoveryDetail } from '@/components/issue/discovery/DiscoveryDetail';
 
@@ -29,6 +29,12 @@ export function DiscoveryPage() {
     isExporting,
   } = useIssueDiscovery({ refetchInterval: 3000 });
 
+  // Fetch issues to find related ones when clicking findings
+  const { issues } = useIssues({
+    // Don't apply filters to get all issues for matching
+    filter: undefined
+  });
+
   if (error) {
     return (
       <div className="space-y-6">
@@ -167,6 +173,7 @@ export function DiscoveryPage() {
               onExport={exportFindings}
               onExportSelected={exportSelectedFindings}
               isExporting={isExporting}
+              issues={issues}
             />
           )}
         </div>

ccw/frontend/src/stores/queueSchedulerStore.ts

@@ -309,7 +309,7 @@ export const selectQueueSchedulerStatus = (state: QueueSchedulerStore): QueueSch
 
 /** Select all queue items */
 export const selectQueueItems = (state: QueueSchedulerStore): QueueItem[] =>
-  state?.items ?? [];
+  state?.items ?? EMPTY_ITEMS;
 
 /**
  * Select items that are ready to execute (status 'queued' or 'pending').
@@ -347,7 +347,7 @@ export const selectExecutingItems = (state: QueueSchedulerStore): QueueItem[] =>
  */
 export const selectSchedulerProgress = (state: QueueSchedulerStore): number => {
   if (!state) return 0;
-  const items = state.items ?? [];
+  const items = state.items ?? EMPTY_ITEMS;
   const total = items.length;
   if (total === 0) return 0;
   const terminal = items.filter(

ccw/src/core/a2ui/A2UIWebSocketHandler.ts

@@ -341,6 +341,77 @@ export class A2UIWebSocketHandler {
   ): boolean {
     const params = action.parameters ?? {};
     const questionId = typeof params.questionId === 'string' ? params.questionId : undefined;
+
+    // Handle submit-all first - it uses compositeId instead of questionId
+    if (action.actionId === 'submit-all') {
+      const compositeId = typeof params.compositeId === 'string' ? params.compositeId : undefined;
+      const questionIds = Array.isArray(params.questionIds) ? params.questionIds as string[] : undefined;
+      if (!compositeId || !questionIds) {
+        return false;
+      }
+
+      // DEBUG: NDJSON log for submit-all received
+      console.log(JSON.stringify({
+        timestamp: new Date().toISOString(),
+        level: 'DEBUG',
+        hid: 'H2',
+        event: 'submit_all_received_in_handleQuestionAction',
+        compositeId,
+        questionCount: questionIds.length
+      }));
+
+      // Collect answers for all sub-questions
+      const answers: QuestionAnswer[] = [];
+      for (const qId of questionIds) {
+        const singleSel = this.singleSelectSelections.get(qId);
+        const multiSel = this.multiSelectSelections.get(qId);
+        const inputVal = this.inputValues.get(qId);
+        const otherText = this.inputValues.get(`__other__:${qId}`);
+
+        if (singleSel !== undefined) {
+          const value = singleSel === '__other__' && otherText ? otherText : singleSel;
+          answers.push({ questionId: qId, value, cancelled: false });
+        } else if (multiSel !== undefined) {
+          const values = Array.from(multiSel).map(v =>
+            v === '__other__' && otherText ? otherText : v
+          );
+          answers.push({ questionId: qId, value: values, cancelled: false });
+        } else if (inputVal !== undefined) {
+          answers.push({ questionId: qId, value: inputVal, cancelled: false });
+        } else {
+          answers.push({ questionId: qId, value: '', cancelled: false });
+        }
+
+        // Cleanup per-question tracking
+        this.singleSelectSelections.delete(qId);
+        this.multiSelectSelections.delete(qId);
+        this.inputValues.delete(qId);
+        this.inputValues.delete(`__other__:${qId}`);
+      }
+
+      // Call multi-answer callback
+      let handled = false;
+      if (this.multiAnswerCallback) {
+        handled = this.multiAnswerCallback(compositeId, answers);
+      }
+      if (!handled) {
+        // Store for HTTP polling retrieval
+        this.resolvedMultiAnswers.set(compositeId, { compositeId, answers, timestamp: Date.now() });
+
+        console.log(JSON.stringify({
+          timestamp: new Date().toISOString(),
+          level: 'DEBUG',
+          hid: 'H2',
+          event: 'answer_stored_for_polling',
+          compositeId,
+          answerCount: answers.length
+        }));
+      }
+      this.activeSurfaces.delete(compositeId);
+      return true;
+    }
+
+    // For other actions, questionId is required
     if (!questionId) {
       return false;
     }
@@ -446,60 +517,6 @@ export class A2UIWebSocketHandler {
         return true;
       }
 
-      case 'submit-all': {
-        // Multi-question composite submit
-        const compositeId = typeof params.compositeId === 'string' ? params.compositeId : undefined;
-        const questionIds = Array.isArray(params.questionIds) ? params.questionIds as string[] : undefined;
-        if (!compositeId || !questionIds) {
-          return false;
-        }
-
-        // Collect answers for all sub-questions
-        const answers: QuestionAnswer[] = [];
-        for (const qId of questionIds) {
-          const singleSel = this.singleSelectSelections.get(qId);
-          const multiSel = this.multiSelectSelections.get(qId);
-          const inputVal = this.inputValues.get(qId);
-          const otherText = this.inputValues.get(`__other__:${qId}`);
-
-          if (singleSel !== undefined) {
-            // Resolve __other__ to actual text input
-            const value = singleSel === '__other__' && otherText ? otherText : singleSel;
-            answers.push({ questionId: qId, value, cancelled: false });
-          } else if (multiSel !== undefined) {
-            // Resolve __other__ in multi-select: replace with actual text
-            const values = Array.from(multiSel).map(v =>
-              v === '__other__' && otherText ? otherText : v
-            );
-            answers.push({ questionId: qId, value: values, cancelled: false });
-          } else if (inputVal !== undefined) {
-            answers.push({ questionId: qId, value: inputVal, cancelled: false });
-          } else {
-            // No value recorded — include empty
-            answers.push({ questionId: qId, value: '', cancelled: false });
-          }
-
-          // Cleanup per-question tracking
-          this.singleSelectSelections.delete(qId);
-          this.multiSelectSelections.delete(qId);
-          this.inputValues.delete(qId);
-          this.inputValues.delete(`__other__:${qId}`);
-        }
-
-        // Call multi-answer callback
-        let handled = false;
-        if (this.multiAnswerCallback) {
-          handled = this.multiAnswerCallback(compositeId, answers);
-        }
-        if (!handled) {
-          // Store for HTTP polling retrieval
-          this.resolvedMultiAnswers.set(compositeId, { compositeId, answers, timestamp: Date.now() });
-        }
-        // Always clean up UI state
-        this.activeSurfaces.delete(compositeId);
-        return true;
-      }
-
       default:
         return false;
     }

ccw/src/core/routes/discovery-routes.ts

@@ -262,7 +262,13 @@ function flattenFindings(perspectiveResults: any[]): any[] {
   const allFindings: any[] = [];
   for (const result of perspectiveResults) {
     if (result.findings) {
-      allFindings.push(...result.findings);
+      // Map backend 'priority' to frontend 'severity' for compatibility
+      const mappedFindings = result.findings.map((f: any) => ({
+        ...f,
+        severity: f.severity || f.priority || 'medium',
+        sessionId: f.discovery_id || result.discovery_id
+      }));
+      allFindings.push(...mappedFindings);
     }
   }
   return allFindings;

ccw/src/core/routes/orchestrator-routes.ts

@@ -31,6 +31,7 @@
  */
 
 import { join, dirname } from 'path';
+import { existsSync } from 'fs';
 import { randomBytes } from 'crypto';
 import { fileURLToPath } from 'url';
 import type { RouteContext } from './types.js';
@@ -635,10 +636,24 @@ function getTemplatesDir(workflowDir: string): string {
 
 /**
  * Get the builtin templates directory path (shipped with CCW)
+ * Returns null if the directory doesn't exist (e.g., in npm package without src/)
  */
-function getBuiltinTemplatesDir(): string {
-  // Resolve relative to this module's location
-  return join(__dirname, '..', '..', '..', 'templates', 'orchestrator');
+function getBuiltinTemplatesDir(): string | null {
+  // Try multiple possible locations for builtin templates
+  const possiblePaths = [
+    // From dist/core/routes/ -> ccw/templates/orchestrator/
+    join(__dirname, '..', '..', '..', 'templates', 'orchestrator'),
+    // From dist/core/routes/ -> ccw/src/templates/orchestrator/ (dev mode)
+    join(__dirname, '..', '..', '..', 'src', 'templates', 'orchestrator'),
+  ];
+
+  for (const path of possiblePaths) {
+    if (existsSync(path)) {
+      return path;
+    }
+  }
+
+  return null;
 }
 
 /**
@@ -745,10 +760,10 @@ async function listLocalTemplates(workflowDir: string): Promise<Template[]> {
  */
 async function listBuiltinTemplates(): Promise<Template[]> {
   const { readdir, readFile } = await import('fs/promises');
-  const { existsSync } = await import('fs');
   const builtinDir = getBuiltinTemplatesDir();
 
-  if (!existsSync(builtinDir)) {
+  // getBuiltinTemplatesDir() returns null if no builtin templates directory exists
+  if (!builtinDir) {
     return [];
   }
 

ccw/src/tools/ask-question.ts

@@ -611,8 +611,29 @@ export function handleAnswer(answer: QuestionAnswer): boolean {
  * @returns True if answer was processed
  */
 export function handleMultiAnswer(compositeId: string, answers: QuestionAnswer[]): boolean {
+  const handleStartTime = Date.now();
+
+  // DEBUG: NDJSON log for handleMultiAnswer start
+  console.log(JSON.stringify({
+    timestamp: new Date().toISOString(),
+    level: 'DEBUG',
+    hid: 'H3',
+    event: 'handle_multi_answer_start',
+    compositeId,
+    answerCount: answers.length
+  }));
+
   const pending = getPendingQuestion(compositeId);
   if (!pending) {
+    // DEBUG: NDJSON log for missing pending question
+    console.log(JSON.stringify({
+      timestamp: new Date().toISOString(),
+      level: 'DEBUG',
+      hid: 'H3',
+      event: 'handle_multi_answer_no_pending',
+      compositeId,
+      elapsedMs: Date.now() - handleStartTime
+    }));
     return false;
   }
 
@@ -625,6 +646,17 @@ export function handleMultiAnswer(compositeId: string, answers: QuestionAnswer[]
   });
 
   removePendingQuestion(compositeId);
+
+  // DEBUG: NDJSON log for handleMultiAnswer complete
+  console.log(JSON.stringify({
+    timestamp: new Date().toISOString(),
+    level: 'DEBUG',
+    hid: 'H3',
+    event: 'handle_multi_answer_complete',
+    compositeId,
+    elapsedMs: Date.now() - handleStartTime
+  }));
+
   return true;
 }
 
@@ -637,9 +669,24 @@ export function handleMultiAnswer(compositeId: string, answers: QuestionAnswer[]
  */
 function startAnswerPolling(questionId: string, isComposite: boolean = false): void {
   const pollPath = `/api/a2ui/answer?questionId=${encodeURIComponent(questionId)}&composite=${isComposite}`;
+  const startTime = Date.now();
+
+  // DEBUG: NDJSON log for polling start
+  console.log(JSON.stringify({
+    timestamp: new Date().toISOString(),
+    level: 'DEBUG',
+    hid: 'H1',
+    event: 'polling_start',
+    questionId,
+    isComposite,
+    port: DASHBOARD_PORT,
+    firstPollDelayMs: POLL_INTERVAL_MS
+  }));
 
   console.error(`[A2UI-Poll] Starting polling for questionId=${questionId}, composite=${isComposite}, port=${DASHBOARD_PORT}`);
 
+  let pollCount = 0;
+
   const poll = () => {
     // Stop if the question was already resolved or timed out
     if (!hasPendingQuestion(questionId)) {
@@ -647,6 +694,20 @@ function startAnswerPolling(questionId: string, isComposite: boolean = false): v
       return;
     }
 
+    pollCount++;
+    const pollStartTime = Date.now();
+
+    // DEBUG: NDJSON log for each poll attempt
+    console.log(JSON.stringify({
+      timestamp: new Date().toISOString(),
+      level: 'DEBUG',
+      hid: 'H1',
+      event: 'poll_attempt',
+      questionId,
+      pollCount,
+      elapsedMs: pollStartTime - startTime
+    }));
+
     const req = http.get({ hostname: '127.0.0.1', port: DASHBOARD_PORT, path: pollPath, timeout: 2000 }, (res) => {
       let data = '';
       res.on('data', (chunk: Buffer) => { data += chunk.toString(); });
@@ -667,6 +728,20 @@ function startAnswerPolling(questionId: string, isComposite: boolean = false): v
 
           console.error(`[A2UI-Poll] Answer received for questionId=${questionId}:`, JSON.stringify(parsed).slice(0, 200));
 
+          // DEBUG: NDJSON log for answer received
+          console.log(JSON.stringify({
+            timestamp: new Date().toISOString(),
+            level: 'DEBUG',
+            hid: 'H1',
+            event: 'answer_received',
+            questionId,
+            pollCount,
+            elapsedMs: Date.now() - startTime,
+            pollLatencyMs: Date.now() - pollStartTime,
+            isComposite,
+            answerPreview: JSON.stringify(parsed).slice(0, 100)
+          }));
+
           if (isComposite && Array.isArray(parsed.answers)) {
             const ok = handleMultiAnswer(questionId, parsed.answers as QuestionAnswer[]);
             console.error(`[A2UI-Poll] handleMultiAnswer result: ${ok}`);

ccw/src/tools/write-file.ts

@@ -105,8 +105,8 @@ export const schema: ToolSchema = {
   name: 'write_file',
   description: `Write content to file. Auto-creates parent directories.
 
-Usage: write_file(path="file.js", content="code here")
-Options: backup=true (backup before overwrite), createDirectories=false (disable auto-creation), encoding="utf8"`,
+Required: path (string), content (string)
+Options: backup=true, createDirectories=false, encoding="utf8"`,
   inputSchema: {
     type: 'object',
     properties: {

docs/.gitignore

@@ -0,0 +1,43 @@
+# Dependencies
+node_modules/
+npm-debug.log*
+yarn-debug.log*
+yarn-error.log*
+pnpm-debug.log*
+
+# Build outputs
+.vitepress/dist/
+.vitepress/cache/
+docs/.vitepress/dist/
+docs/.vitepress/cache/
+dist/
+public/search-index.*.json
+
+# Editor directories and files
+.vscode/
+.idea/
+*.suo
+*.ntvs*
+*.njsproj
+*.sln
+*.sw?
+.DS_Store
+
+# Temporary files
+*.tmp
+*.temp
+.cache/
+
+# Logs
+logs/
+*.log
+
+# Environment
+.env
+.env.local
+.env.*.local
+
+# OS
+.DS_Store
+Thumbs.db
+.ace-tool/

docs/.vitepress/config.ts

@@ -0,0 +1,428 @@
+import { defineConfig } from 'vitepress'
+
+const repoName = process.env.GITHUB_REPOSITORY?.split('/')[1]
+const isUserOrOrgSite = Boolean(repoName && repoName.endsWith('.github.io'))
+
+const base =
+  process.env.CCW_DOCS_BASE ||
+  (process.env.GITHUB_ACTIONS && repoName && !isUserOrOrgSite ? `/${repoName}/` : '/')
+
+export default defineConfig({
+  title: 'CCW Documentation',
+  description: 'Claude Code Workspace - Advanced AI-Powered Development Environment',
+  lang: 'zh-CN',
+  base,
+
+  // Ignore dead links for incomplete docs
+  ignoreDeadLinks: true,
+  head: [
+    ['link', { rel: 'icon', href: '/favicon.svg', type: 'image/svg+xml' }],
+    [
+      'script',
+      {},
+      `(() => {
+  try {
+    const theme = localStorage.getItem('ccw-theme') || 'blue'
+    document.documentElement.setAttribute('data-theme', theme)
+
+    const mode = localStorage.getItem('ccw-color-mode') || 'auto'
+    const prefersDark = window.matchMedia && window.matchMedia('(prefers-color-scheme: dark)').matches
+    const isDark = mode === 'dark' || (mode === 'auto' && prefersDark)
+    document.documentElement.classList.toggle('dark', isDark)
+  } catch {}
+})()`
+    ],
+    ['meta', { name: 'theme-color', content: '#3b82f6' }],
+    ['meta', { name: 'og:type', content: 'website' }],
+    ['meta', { name: 'og:locale', content: 'en_US' }],
+    ['meta', { name: 'og:locale:alternate', content: 'zh_CN' }]
+  ],
+
+  // Appearance
+  appearance: false,
+
+  // Vite build/dev optimizations
+  vite: {
+    optimizeDeps: {
+      include: ['flexsearch']
+    },
+    build: {
+      target: 'es2019',
+      cssCodeSplit: true
+    }
+  },
+
+  // Theme configuration
+  themeConfig: {
+    logo: '/logo.svg',
+
+    // Right-side table of contents (outline)
+    outline: {
+      level: [2, 3],
+      label: 'On this page'
+    },
+
+    // Navigation - 按照 Trellis 风格组织
+    nav: [
+      { text: 'Guide', link: '/guide/ch01-what-is-claude-dms3' },
+      { text: 'Commands', link: '/commands/claude/' },
+      { text: 'Skills', link: '/skills/' },
+      { text: 'Features', link: '/features/spec' },
+      {
+        text: 'Languages',
+        items: [
+          { text: '简体中文', link: '/zh/guide/ch01-what-is-claude-dms3' }
+        ]
+      }
+    ],
+
+    // Sidebar - 优化导航结构，增加二级标题和归类
+    sidebar: {
+      '/guide/': [
+        {
+          text: '📖 指南',
+          collapsible: false,
+          items: [
+            { text: 'What is Claude_dms3', link: '/guide/ch01-what-is-claude-dms3' },
+            { text: 'Getting Started', link: '/guide/ch02-getting-started' },
+            { text: 'Core Concepts', link: '/guide/ch03-core-concepts' },
+            { text: 'Workflow Basics', link: '/guide/ch04-workflow-basics' },
+            { text: 'Advanced Tips', link: '/guide/ch05-advanced-tips' },
+            { text: 'Best Practices', link: '/guide/ch06-best-practices' }
+          ]
+        },
+        {
+          text: '🚀 快速入口',
+          collapsible: true,
+          items: [
+            { text: 'Installation', link: '/guide/installation' },
+            { text: 'First Workflow', link: '/guide/first-workflow' },
+            { text: 'CLI Tools', link: '/guide/cli-tools' }
+          ]
+        }
+      ],
+      '/commands/': [
+        {
+          text: '🤖 Claude Commands',
+          collapsible: true,
+          items: [
+            { text: 'Overview', link: '/commands/claude/' },
+            { text: 'Core Orchestration', link: '/commands/claude/core-orchestration' },
+            { text: 'Workflow', link: '/commands/claude/workflow' },
+            { text: 'Session', link: '/commands/claude/session' },
+            { text: 'Issue', link: '/commands/claude/issue' },
+            { text: 'Memory', link: '/commands/claude/memory' },
+            { text: 'CLI', link: '/commands/claude/cli' },
+            { text: 'UI Design', link: '/commands/claude/ui-design' }
+          ]
+        },
+        {
+          text: '📝 Codex Prompts',
+          collapsible: true,
+          items: [
+            { text: 'Overview', link: '/commands/codex/' },
+            { text: 'Prep', link: '/commands/codex/prep' },
+            { text: 'Review', link: '/commands/codex/review' }
+          ]
+        }
+      ],
+      '/skills/': [
+        {
+          text: '⚡ Claude Skills',
+          collapsible: true,
+          items: [
+            { text: 'Overview', link: '/skills/claude-index' },
+            { text: 'Collaboration', link: '/skills/claude-collaboration' },
+            { text: 'Workflow', link: '/skills/claude-workflow' },
+            { text: 'Memory', link: '/skills/claude-memory' },
+            { text: 'Review', link: '/skills/claude-review' },
+            { text: 'Meta', link: '/skills/claude-meta' }
+          ]
+        },
+        {
+          text: '🔧 Codex Skills',
+          collapsible: true,
+          items: [
+            { text: 'Overview', link: '/skills/codex-index' },
+            { text: 'Lifecycle', link: '/skills/codex-lifecycle' },
+            { text: 'Workflow', link: '/skills/codex-workflow' },
+            { text: 'Specialized', link: '/skills/codex-specialized' }
+          ]
+        },
+        {
+          text: '🎨 Custom Skills',
+          collapsible: true,
+          items: [
+            { text: 'Overview', link: '/skills/custom' },
+            { text: 'Core Skills', link: '/skills/core-skills' },
+            { text: 'Reference', link: '/skills/reference' }
+          ]
+        }
+      ],
+      '/features/': [
+        {
+          text: '⚙️ Core Features',
+          collapsible: false,
+          items: [
+            { text: 'Spec System', link: '/features/spec' },
+            { text: 'Memory System', link: '/features/memory' },
+            { text: 'CLI Call', link: '/features/cli' },
+            { text: 'Dashboard', link: '/features/dashboard' },
+            { text: 'CodexLens', link: '/features/codexlens' }
+          ]
+        },
+        {
+          text: '🔌 Settings',
+          collapsible: true,
+          items: [
+            { text: 'API Settings', link: '/features/api-settings' },
+            { text: 'System Settings', link: '/features/system-settings' }
+          ]
+        }
+      ],
+      '/mcp/': [
+        {
+          text: '🔗 MCP Tools',
+          collapsible: true,
+          items: [
+            { text: 'Overview', link: '/mcp/tools' }
+          ]
+        }
+      ],
+      '/agents/': [
+        {
+          text: '🤖 Agents',
+          collapsible: true,
+          items: [
+            { text: 'Overview', link: '/agents/' },
+            { text: 'Built-in Agents', link: '/agents/builtin' },
+            { text: 'Custom Agents', link: '/agents/custom' }
+          ]
+        }
+      ],
+      '/workflows/': [
+        {
+          text: '🔄 Workflow System',
+          collapsible: true,
+          items: [
+            { text: 'Overview', link: '/workflows/' },
+            { text: '4-Level System', link: '/workflows/4-level' },
+            { text: 'Examples', link: '/workflows/examples' },
+            { text: 'Best Practices', link: '/workflows/best-practices' },
+            { text: 'Teams', link: '/workflows/teams' }
+          ]
+        }
+      ]
+    },
+
+    // Social links
+    socialLinks: [
+      { icon: 'github', link: 'https://github.com/catlog22/Claude-Code-Workflow' }
+    ],
+
+    // Footer
+    footer: {
+      message: 'Released under the MIT License.',
+      copyright: 'Copyright © 2025-present CCW Contributors'
+    },
+
+    // Edit link
+    editLink: {
+      pattern: 'https://github.com/catlog22/Claude-Code-Workflow/edit/main/docs/:path',
+      text: 'Edit this page on GitHub'
+    },
+
+    // Last updated
+    lastUpdated: {
+      text: 'Last updated',
+      formatOptions: {
+        dateStyle: 'full',
+        timeStyle: 'short'
+      }
+    },
+
+    // Search (handled by custom FlexSearch DocSearch component)
+    search: false
+  },
+
+  // Markdown configuration
+  markdown: {
+    lineNumbers: true,
+    theme: {
+      light: 'github-light',
+      dark: 'github-dark'
+    },
+    languages: [
+      'bash',
+      'shell',
+      'powershell',
+      'json',
+      'yaml',
+      'toml',
+      'javascript',
+      'typescript',
+      'jsx',
+      'tsx',
+      'vue',
+      'html',
+      'css',
+      'markdown',
+      'python',
+      'ruby',
+      'diff',
+      'xml',
+      'mermaid'
+    ],
+    config: (md) => {
+      // Add markdown-it plugins if needed
+    }
+  },
+
+  // locales
+  locales: {
+    root: {
+      label: 'English',
+      lang: 'en-US'
+    },
+    zh: {
+      label: '简体中文',
+      lang: 'zh-CN',
+      title: 'CCW 文档',
+      description: 'Claude Code Workspace - 高级 AI 驱动开发环境',
+      themeConfig: {
+        outline: {
+          level: [2, 3],
+          label: '本页目录'
+        },
+        nav: [
+          { text: '指南', link: '/zh/guide/ch01-what-is-claude-dms3' },
+          { text: '命令', link: '/zh/commands/claude/' },
+          { text: '技能', link: '/zh/skills/claude-index' },
+          { text: '功能', link: '/zh/features/spec' },
+          {
+            text: '语言',
+            items: [
+              { text: 'English', link: '/guide/ch01-what-is-claude-dms3' }
+            ]
+          }
+        ],
+        sidebar: {
+          '/zh/guide/': [
+            {
+              text: '📖 指南',
+              collapsible: false,
+              items: [
+                { text: 'Claude_dms3 是什么', link: '/zh/guide/ch01-what-is-claude-dms3' },
+                { text: '快速开始', link: '/zh/guide/ch02-getting-started' },
+                { text: '核心概念', link: '/zh/guide/ch03-core-concepts' },
+                { text: '工作流基础', link: '/zh/guide/ch04-workflow-basics' },
+                { text: '高级技巧', link: '/zh/guide/ch05-advanced-tips' },
+                { text: '最佳实践', link: '/zh/guide/ch06-best-practices' }
+              ]
+            },
+            {
+              text: '🚀 快速入口',
+              collapsible: true,
+              items: [
+                { text: '安装', link: '/zh/guide/installation' },
+                { text: '第一个工作流', link: '/zh/guide/first-workflow' },
+                { text: 'CLI 工具', link: '/zh/guide/cli-tools' }
+              ]
+            }
+          ],
+          '/zh/commands/': [
+            {
+              text: '🤖 Claude 命令',
+              collapsible: true,
+              items: [
+                { text: '概述', link: '/zh/commands/claude/' },
+                { text: '核心编排', link: '/zh/commands/claude/core-orchestration' },
+                { text: '工作流', link: '/zh/commands/claude/workflow' },
+                { text: '会话管理', link: '/zh/commands/claude/session' },
+                { text: 'Issue', link: '/zh/commands/claude/issue' },
+                { text: 'Memory', link: '/zh/commands/claude/memory' },
+                { text: 'CLI', link: '/zh/commands/claude/cli' },
+                { text: 'UI 设计', link: '/zh/commands/claude/ui-design' }
+              ]
+            },
+            {
+              text: '📝 Codex Prompts',
+              collapsible: true,
+              items: [
+                { text: '概述', link: '/zh/commands/codex/' },
+                { text: 'Prep', link: '/zh/commands/codex/prep' },
+                { text: 'Review', link: '/zh/commands/codex/review' }
+              ]
+            }
+          ],
+          '/zh/skills/': [
+            {
+              text: '⚡ Claude Skills',
+              collapsible: true,
+              items: [
+                { text: '概述', link: '/zh/skills/claude-index' },
+                { text: '协作', link: '/zh/skills/claude-collaboration' },
+                { text: '工作流', link: '/zh/skills/claude-workflow' },
+                { text: '记忆', link: '/zh/skills/claude-memory' },
+                { text: '审查', link: '/zh/skills/claude-review' },
+                { text: '元技能', link: '/zh/skills/claude-meta' }
+              ]
+            },
+            {
+              text: '🔧 Codex Skills',
+              collapsible: true,
+              items: [
+                { text: '概述', link: '/zh/skills/codex-index' },
+                { text: '生命周期', link: '/zh/skills/codex-lifecycle' },
+                { text: '工作流', link: '/zh/skills/codex-workflow' },
+                { text: '专项', link: '/zh/skills/codex-specialized' }
+              ]
+            },
+            {
+              text: '🎨 自定义技能',
+              collapsible: true,
+              items: [
+                { text: '概述', link: '/zh/skills/custom' },
+                { text: '核心技能', link: '/zh/skills/core-skills' },
+                { text: '参考', link: '/zh/skills/reference' }
+              ]
+            }
+          ],
+          '/zh/features/': [
+            {
+              text: '⚙️ 核心功能',
+              collapsible: false,
+              items: [
+                { text: 'Spec 规范系统', link: '/zh/features/spec' },
+                { text: 'Memory 记忆系统', link: '/zh/features/memory' },
+                { text: 'CLI 调用', link: '/zh/features/cli' },
+                { text: 'Dashboard 面板', link: '/zh/features/dashboard' },
+                { text: 'CodexLens', link: '/zh/features/codexlens' }
+              ]
+            },
+            {
+              text: '🔌 设置',
+              collapsible: true,
+              items: [
+                { text: 'API 设置', link: '/zh/features/api-settings' },
+                { text: '系统设置', link: '/zh/features/system-settings' }
+              ]
+            }
+          ],
+          '/zh/workflows/': [
+            {
+              text: '🔄 工作流系统',
+              collapsible: true,
+              items: [
+                { text: '概述', link: '/zh/workflows/' },
+                { text: '四级体系', link: '/zh/workflows/4-level' },
+                { text: '最佳实践', link: '/zh/workflows/best-practices' },
+                { text: '团队协作', link: '/zh/workflows/teams' }
+              ]
+            }
+          ]
+        }
+      }
+    }
+  }
+})

docs/.vitepress/search/flexsearch.mjs

@@ -0,0 +1,25 @@
+export const FLEXSEARCH_INDEX_VERSION = 1
+
+export function flexsearchEncode(text) {
+  const normalized = String(text ?? '')
+    .toLowerCase()
+    .normalize('NFKC')
+
+  const tokens = normalized.match(
+    /[a-z0-9]+|[\u3040-\u30ff\u3400-\u4dbf\u4e00-\u9fff\uac00-\ud7af]/g
+  )
+
+  return tokens ?? []
+}
+
+export const FLEXSEARCH_OPTIONS = {
+  tokenize: 'forward',
+  resolution: 9,
+  cache: 100,
+  encode: flexsearchEncode
+}
+
+export function createFlexSearchIndex(FlexSearch) {
+  return new FlexSearch.Index(FLEXSEARCH_OPTIONS)
+}
+

docs/.vitepress/theme/components/AgentOrchestration.vue

@@ -0,0 +1,216 @@
+<template>
+  <div class="agent-orchestration">
+    <div class="orchestration-title">🤖 Agent Orchestration</div>
+    
+    <div class="agent-flow">
+      <!-- CLI Layer -->
+      <div class="flow-layer cli-layer">
+        <div class="layer-label">CLI Tools</div>
+        <div class="agents-row">
+          <div class="agent-card cli" @mouseenter="showTooltip('cli-explore')" @mouseleave="hideTooltip">🔍 Explore</div>
+          <div class="agent-card cli" @mouseenter="showTooltip('cli-plan')" @mouseleave="hideTooltip">📋 Plan</div>
+          <div class="agent-card cli" @mouseenter="showTooltip('cli-exec')" @mouseleave="hideTooltip">⚡ Execute</div>
+          <div class="agent-card cli" @mouseenter="showTooltip('cli-discuss')" @mouseleave="hideTooltip">💬 Discuss</div>
+        </div>
+      </div>
+      
+      <!-- Flow Arrow -->
+      <div class="flow-arrow">▼</div>
+      
+      <!-- Development Layer -->
+      <div class="flow-layer dev-layer">
+        <div class="layer-label">Development</div>
+        <div class="agents-row">
+          <div class="agent-card dev" @mouseenter="showTooltip('code-dev')" @mouseleave="hideTooltip">👨‍💻 Code</div>
+          <div class="agent-card dev" @mouseenter="showTooltip('tdd')" @mouseleave="hideTooltip">🧪 TDD</div>
+          <div class="agent-card dev" @mouseenter="showTooltip('test-fix')" @mouseleave="hideTooltip">🔧 Fix</div>
+        </div>
+      </div>
+      
+      <!-- Flow Arrow -->
+      <div class="flow-arrow">▼</div>
+      
+      <!-- Output Layer -->
+      <div class="flow-layer output-layer">
+        <div class="layer-label">Output</div>
+        <div class="agents-row">
+          <div class="agent-card doc" @mouseenter="showTooltip('doc-gen')" @mouseleave="hideTooltip">📄 Docs</div>
+          <div class="agent-card ui" @mouseenter="showTooltip('ui-design')" @mouseleave="hideTooltip">🎨 UI</div>
+          <div class="agent-card universal" @mouseenter="showTooltip('universal')" @mouseleave="hideTooltip">🌐 Universal</div>
+        </div>
+      </div>
+    </div>
+    
+    <div class="tooltip" v-if="tooltip" :class="{ visible: tooltip }">
+      {{ tooltipText }}
+    </div>
+  </div>
+</template>
+
+<script setup>
+import { ref } from 'vue'
+
+const tooltip = ref(false)
+const tooltipText = ref('')
+
+const tooltips = {
+  'cli-explore': 'cli-explore-agent: 代码库探索和语义搜索',
+  'cli-plan': 'cli-planning-agent: 任务规划和分解',
+  'cli-exec': 'cli-execution-agent: 命令执行和结果处理',
+  'cli-discuss': 'cli-discuss-agent: 多视角讨论和共识达成',
+  'code-dev': 'code-developer: 代码实现和开发',
+  'tdd': 'tdd-developer: 测试驱动开发',
+  'test-fix': 'test-fix-agent: 测试修复循环',
+  'doc-gen': 'doc-generator: 文档自动生成',
+  'ui-design': 'ui-design-agent: UI设计和设计令牌',
+  'universal': 'universal-executor: 通用任务执行器'
+}
+
+function showTooltip(key) {
+  tooltipText.value = tooltips[key] || ''
+  tooltip.value = true
+}
+
+function hideTooltip() {
+  tooltip.value = false
+}
+</script>
+
+<style scoped>
+.agent-orchestration {
+  padding: 3rem 2rem;
+  background: var(--vp-c-bg-soft);
+  border: 1px solid var(--vp-c-divider);
+  border-radius: 24px;
+  margin: 2rem 0;
+  position: relative;
+  overflow: hidden;
+}
+
+.orchestration-title {
+  text-align: center;
+  font-size: 1.75rem;
+  font-weight: 700;
+  color: var(--vp-c-text-1);
+  margin-bottom: 2.5rem;
+}
+
+.agent-flow {
+  display: flex;
+  flex-direction: column;
+  align-items: center;
+  gap: 1rem;
+}
+
+.flow-layer {
+  width: 100%;
+  max-width: 600px;
+}
+
+.layer-label {
+  font-size: 0.7rem;
+  font-weight: 800;
+  text-transform: uppercase;
+  letter-spacing: 0.15em;
+  color: var(--vp-c-text-3);
+  margin-bottom: 0.75rem;
+  text-align: center;
+}
+
+.agents-row {
+  display: flex;
+  justify-content: center;
+  flex-wrap: wrap;
+  gap: 1rem;
+}
+
+.agent-card {
+  padding: 0.8rem 1.5rem;
+  border-radius: 12px;
+  font-size: 0.9rem;
+  font-weight: 600;
+  cursor: pointer;
+  transition: all 0.3s cubic-bezier(0.4, 0, 0.2, 1);
+  position: relative;
+  border: 1px solid transparent;
+}
+
+.agent-card.cli {
+  background: var(--vp-c-brand-soft);
+  color: var(--vp-c-brand-1);
+  border-color: rgba(59, 130, 246, 0.2);
+}
+
+.agent-card.dev {
+  background: rgba(16, 185, 129, 0.1);
+  color: #10B981;
+  border-color: rgba(16, 185, 129, 0.2);
+}
+
+.agent-card.doc {
+  background: rgba(139, 92, 246, 0.1);
+  color: #8B5CF6;
+  border-color: rgba(139, 92, 246, 0.2);
+}
+
+.agent-card.ui {
+  background: rgba(245, 158, 11, 0.1);
+  color: #F59E0B;
+  border-color: rgba(245, 158, 11, 0.2);
+}
+
+.agent-card.universal {
+  background: rgba(239, 68, 68, 0.1);
+  color: #EF4444;
+  border-color: rgba(239, 68, 68, 0.2);
+}
+
+.agent-card:hover {
+  transform: translateY(-3px);
+  box-shadow: 0 8px 24px rgba(0, 0, 0, 0.12);
+}
+
+.flow-arrow {
+  color: var(--vp-c-divider);
+  font-size: 1.25rem;
+  animation: bounce 2s infinite;
+}
+
+@keyframes bounce {
+  0%, 100% { transform: translateY(0); opacity: 0.5; }
+  50% { transform: translateY(8px); opacity: 1; }
+}
+
+.tooltip {
+  position: absolute;
+  top: 1rem;
+  right: 1rem;
+  background: var(--vp-c-bg);
+  border: 1px solid var(--vp-c-brand-1);
+  color: var(--vp-c-text-1);
+  padding: 0.75rem 1.25rem;
+  border-radius: 10px;
+  font-size: 0.85rem;
+  font-weight: 500;
+  opacity: 0;
+  transition: all 0.3s;
+  pointer-events: none;
+  box-shadow: var(--vp-shadow-md);
+  z-index: 10;
+}
+
+.tooltip.visible {
+  opacity: 1;
+}
+
+@media (max-width: 640px) {
+  .agent-orchestration {
+    padding: 2rem 1rem;
+  }
+  
+  .agent-card {
+    padding: 0.6rem 1rem;
+    font-size: 0.8rem;
+  }
+}
+</style>

docs/.vitepress/theme/components/Breadcrumb.vue

@@ -0,0 +1,123 @@
+<script setup lang="ts">
+import { computed } from 'vue'
+import { useData } from 'vitepress'
+
+const { page } = useData()
+
+interface BreadcrumbItem {
+  text: string
+  link?: string
+}
+
+const breadcrumbs = computed<BreadcrumbItem[]>(() => {
+  const items: BreadcrumbItem[] = [
+    { text: 'Home', link: '/' }
+  ]
+
+  const pathSegments = page.value.relativePath.split('/')
+  const fileName = pathSegments.pop()?.replace(/\.md$/, '')
+
+  // Build breadcrumb from path
+  let currentPath = ''
+  for (const segment of pathSegments) {
+    currentPath += `${segment}/`
+    items.push({
+      text: formatTitle(segment),
+      link: `/${currentPath}`
+    })
+  }
+
+  // Add current page
+  if (fileName && fileName !== 'index') {
+    items.push({
+      text: formatTitle(fileName)
+    })
+  }
+
+  return items
+})
+
+const formatTitle = (str: string): string => {
+  return str
+    .split(/[-_]/)
+    .map(word => word.charAt(0).toUpperCase() + word.slice(1))
+    .join(' ')
+}
+</script>
+
+<template>
+  <nav v-if="breadcrumbs.length > 1" class="breadcrumb" aria-label="Breadcrumb">
+    <ol class="breadcrumb-list">
+      <li v-for="(item, index) in breadcrumbs" :key="index" class="breadcrumb-item">
+        <router-link v-if="item.link && index < breadcrumbs.length - 1" :to="item.link" class="breadcrumb-link">
+          {{ item.text }}
+        </router-link>
+        <span v-else class="breadcrumb-current">{{ item.text }}</span>
+        <span v-if="index < breadcrumbs.length - 1" class="breadcrumb-separator">/</span>
+      </li>
+    </ol>
+  </nav>
+</template>
+
+<style scoped>
+.breadcrumb {
+  padding: 12px 0;
+  margin-bottom: 16px;
+}
+
+.breadcrumb-list {
+  display: flex;
+  flex-wrap: wrap;
+  align-items: center;
+  gap: 4px;
+  list-style: none;
+  margin: 0;
+  padding: 0;
+}
+
+.breadcrumb-item {
+  display: flex;
+  align-items: center;
+  gap: 4px;
+}
+
+.breadcrumb-link {
+  color: var(--vp-c-text-2);
+  font-size: var(--vp-font-size-sm);
+  text-decoration: none;
+  transition: color var(--vp-transition-color);
+}
+
+.breadcrumb-link:hover {
+  color: var(--vp-c-primary);
+  text-decoration: underline;
+}
+
+.breadcrumb-current {
+  color: var(--vp-c-text-1);
+  font-size: var(--vp-font-size-sm);
+  font-weight: 500;
+}
+
+.breadcrumb-separator {
+  color: var(--vp-c-text-3);
+  font-size: var(--vp-font-size-sm);
+}
+
+@media (max-width: 768px) {
+  .breadcrumb {
+    padding: 8px 0;
+    margin-bottom: 12px;
+  }
+
+  .breadcrumb-link,
+  .breadcrumb-current,
+  .breadcrumb-separator {
+    font-size: 12px;
+  }
+
+  .breadcrumb-list {
+    gap: 2px;
+  }
+}
+</style>

docs/.vitepress/theme/components/ColorSchemeSelector.vue

@@ -0,0 +1,72 @@
+<script setup lang="ts">
+// This component is integrated into ThemeSwitcher
+// Kept as separate component for modularity
+const emit = defineEmits<{
+  (e: 'select', scheme: string): void
+}>()
+
+const schemes = [
+  { id: 'blue', name: 'Blue', color: '#3b82f6' },
+  { id: 'green', name: 'Green', color: '#10b981' },
+  { id: 'orange', name: 'Orange', color: '#f59e0b' },
+  { id: 'purple', name: 'Purple', color: '#8b5cf6' }
+]
+
+const selectScheme = (schemeId: string) => {
+  emit('select', schemeId)
+}
+</script>
+
+<template>
+  <div class="color-scheme-selector">
+    <button
+      v-for="scheme in schemes"
+      :key="scheme.id"
+      :class="['scheme-option']"
+      :style="{ '--scheme-color': scheme.color }"
+      :aria-label="scheme.name"
+      @click="selectScheme(scheme.id)"
+    >
+      <span class="scheme-indicator"></span>
+      <span class="scheme-name">{{ scheme.name }}</span>
+    </button>
+  </div>
+</template>
+
+<style scoped>
+.color-scheme-selector {
+  display: flex;
+  flex-direction: column;
+  gap: 4px;
+}
+
+.scheme-option {
+  display: flex;
+  align-items: center;
+  gap: 12px;
+  padding: 8px 12px;
+  border: 1px solid var(--vp-c-border);
+  border-radius: var(--vp-radius-md);
+  background: var(--vp-c-bg);
+  cursor: pointer;
+  transition: all var(--vp-transition-color);
+}
+
+.scheme-option:hover {
+  border-color: var(--vp-c-primary);
+  background: var(--vp-c-bg-soft);
+}
+
+.scheme-indicator {
+  width: 16px;
+  height: 16px;
+  border-radius: var(--vp-radius-full);
+  background: var(--scheme-color);
+  border: 2px solid var(--vp-c-border);
+}
+
+.scheme-name {
+  font-size: var(--vp-font-size-sm);
+  color: var(--vp-c-text-1);
+}
+</style>

docs/.vitepress/theme/components/CopyCodeButton.vue

@@ -0,0 +1,135 @@
+<script setup lang="ts">
+import { ref } from 'vue'
+
+const props = defineProps<{
+  code: string
+}>()
+
+const emit = defineEmits<{
+  (e: 'copy'): void
+}>()
+
+const copied = ref(false)
+
+const copyToClipboard = async () => {
+  try {
+    await navigator.clipboard.writeText(props.code)
+    copied.value = true
+    emit('copy')
+    setTimeout(() => {
+      copied.value = false
+    }, 2000)
+  } catch (err) {
+    console.error('Failed to copy:', err)
+  }
+}
+
+// Also handle Ctrl+C
+const handleKeydown = (e: KeyboardEvent) => {
+  if (e.ctrlKey && e.key === 'c') {
+    copyToClipboard()
+  }
+}
+</script>
+
+<template>
+  <button
+    class="copy-code-button"
+    :class="{ copied }"
+    :aria-label="copied ? 'Copied!' : 'Copy code'"
+    :title="copied ? 'Copied!' : 'Copy code (Ctrl+C)'"
+    @click="copyToClipboard"
+    @keydown="handleKeydown"
+  >
+    <svg v-if="!copied" class="icon copy-icon" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2">
+      <rect x="9" y="9" width="13" height="13" rx="2" ry="2"/>
+      <path d="M5 15H4a2 2 0 0 1-2-2V4a2 2 0 0 1 2-2h9a2 2 0 0 1 2 2v1"/>
+    </svg>
+    <svg v-else class="icon check-icon" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2">
+      <polyline points="20 6 9 17 4 12"/>
+    </svg>
+    <span v-if="copied" class="copy-feedback">Copied!</span>
+  </button>
+</template>
+
+<style scoped>
+.copy-code-button {
+  position: absolute;
+  top: 12px;
+  right: 12px;
+  display: flex;
+  align-items: center;
+  gap: 6px;
+  padding: 6px 12px;
+  border: 1px solid var(--vp-c-border);
+  border-radius: var(--vp-radius-md);
+  background: var(--vp-c-bg);
+  color: var(--vp-c-text-2);
+  font-size: var(--vp-font-size-sm);
+  cursor: pointer;
+  opacity: 0;
+  transition: all var(--vp-transition-color);
+  z-index: 10;
+}
+
+.copy-code-button:hover {
+  background: var(--vp-c-bg-soft);
+  color: var(--vp-c-text-1);
+  border-color: var(--vp-c-primary);
+}
+
+.copy-code-button.copied {
+  background: var(--vp-c-secondary-500);
+  color: white;
+  border-color: var(--vp-c-secondary-500);
+}
+
+.copy-code-button .icon {
+  width: 16px;
+  height: 16px;
+}
+
+.copy-feedback {
+  position: absolute;
+  top: 100%;
+  right: 0;
+  margin-top: 4px;
+  padding: 4px 8px;
+  background: var(--vp-c-bg);
+  border: 1px solid var(--vp-c-border);
+  border-radius: var(--vp-radius-md);
+  font-size: 12px;
+  white-space: nowrap;
+  animation: fadeIn 0.2s ease;
+}
+
+@keyframes fadeIn {
+  from {
+    opacity: 0;
+    transform: translateY(-4px);
+  }
+  to {
+    opacity: 1;
+    transform: translateY(0);
+  }
+}
+
+/* Show button on code block hover */
+div[class*='language-']:hover .copy-code-button,
+.copy-code-button:focus {
+  opacity: 1;
+}
+
+@media (max-width: 768px) {
+  .copy-code-button {
+    opacity: 1;
+    top: 8px;
+    right: 8px;
+    padding: 8px;
+  }
+
+  .copy-feedback {
+    display: none;
+  }
+}
+</style>

docs/.vitepress/theme/components/DarkModeToggle.vue

@@ -0,0 +1,133 @@
+<script setup lang="ts">
+import { ref, onMounted } from 'vue'
+
+type ColorMode = 'light' | 'dark' | 'auto'
+
+const colorMode = ref<ColorMode>('auto')
+
+const modes: { id: ColorMode; name: string; icon: string }[] = [
+  { id: 'light', name: 'Light', icon: 'sun' },
+  { id: 'dark', name: 'Dark', icon: 'moon' },
+  { id: 'auto', name: 'Auto', icon: 'computer' }
+]
+
+const setMode = (mode: ColorMode) => {
+  colorMode.value = mode
+  localStorage.setItem('ccw-color-mode', mode)
+  applyMode(mode)
+}
+
+const applyMode = (mode: ColorMode) => {
+  const html = document.documentElement
+
+  if (mode === 'auto') {
+    const prefersDark = window.matchMedia('(prefers-color-scheme: dark)').matches
+    html.classList.toggle('dark', prefersDark)
+  } else {
+    html.classList.toggle('dark', mode === 'dark')
+  }
+}
+
+onMounted(() => {
+  const savedMode = localStorage.getItem('ccw-color-mode') as ColorMode
+  if (savedMode && modes.find(m => m.id === savedMode)) {
+    setMode(savedMode)
+  } else {
+    setMode('auto')
+  }
+
+  // Listen for system theme changes
+  window.matchMedia('(prefers-color-scheme: dark)').addEventListener('change', () => {
+    if (colorMode.value === 'auto') {
+      applyMode('auto')
+    }
+  })
+})
+</script>
+
+<template>
+  <div class="dark-mode-toggle">
+    <button
+      v-for="mode in modes"
+      :key="mode.id"
+      :class="['mode-button', { active: colorMode === mode.id }]"
+      :aria-label="`Switch to ${mode.name} mode`"
+      :title="mode.name"
+      @click="setMode(mode.id)"
+    >
+      <svg v-if="mode.icon === 'sun'" class="icon" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2">
+        <circle cx="12" cy="12" r="5"/>
+        <line x1="12" y1="1" x2="12" y2="3"/>
+        <line x1="12" y1="21" x2="12" y2="23"/>
+        <line x1="4.22" y1="4.22" x2="5.64" y2="5.64"/>
+        <line x1="18.36" y1="18.36" x2="19.78" y2="19.78"/>
+        <line x1="1" y1="12" x2="3" y2="12"/>
+        <line x1="21" y1="12" x2="23" y2="12"/>
+        <line x1="4.22" y1="19.78" x2="5.64" y2="18.36"/>
+        <line x1="18.36" y1="5.64" x2="19.78" y2="4.22"/>
+      </svg>
+      <svg v-else-if="mode.icon === 'moon'" class="icon" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2">
+        <path d="M21 12.79A9 9 0 1 1 11.21 3 7 7 0 0 0 21 12.79z"/>
+      </svg>
+      <svg v-else class="icon" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2">
+        <rect x="2" y="3" width="20" height="14" rx="2" ry="2"/>
+        <line x1="8" y1="21" x2="16" y2="21"/>
+        <line x1="12" y1="17" x2="12" y2="21"/>
+      </svg>
+    </button>
+  </div>
+</template>
+
+<style scoped>
+.dark-mode-toggle {
+  display: flex;
+  align-items: center;
+  gap: 2px;
+  padding: 4px;
+  background: var(--vp-c-bg-soft);
+  border-radius: var(--vp-radius-full);
+}
+
+.mode-button {
+  position: relative;
+  width: 36px;
+  height: 32px;
+  border: none;
+  background: transparent;
+  border-radius: var(--vp-radius-md);
+  cursor: pointer;
+  display: flex;
+  align-items: center;
+  justify-content: center;
+  color: var(--vp-c-text-2);
+  transition: all var(--vp-transition-color);
+}
+
+.mode-button:hover {
+  color: var(--vp-c-text-1);
+  background: var(--vp-c-bg-mute);
+}
+
+.mode-button.active {
+  background: var(--vp-c-bg);
+  color: var(--vp-c-primary);
+  box-shadow: var(--vp-shadow-sm);
+}
+
+.mode-button .icon {
+  width: 18px;
+  height: 18px;
+}
+
+@media (max-width: 768px) {
+  .mode-button {
+    width: 40px;
+    height: 36px;
+  }
+
+  .mode-button .icon {
+    width: 20px;
+    height: 20px;
+  }
+}
+</style>

docs/.vitepress/theme/components/DocSearch.vue

@@ -0,0 +1,492 @@
+<script setup lang="ts">
+import { computed, nextTick, onBeforeUnmount, onMounted, ref, shallowRef, watch } from 'vue'
+import { useData, useRouter, withBase } from 'vitepress'
+
+type LocaleKey = 'root' | 'zh'
+
+interface SearchDoc {
+  id: number
+  title: string
+  url: string
+  excerpt?: string
+}
+
+interface SearchIndexPayload {
+  version: number
+  locale: LocaleKey
+  index: Record<string, string>
+  docs: SearchDoc[]
+}
+
+const { page } = useData()
+const router = useRouter()
+
+const localeKey = computed<LocaleKey>(() =>
+  page.value.relativePath.startsWith('zh/') ? 'zh' : 'root'
+)
+
+const isOpen = ref(false)
+const isLoading = ref(false)
+const error = ref<string | null>(null)
+
+const query = ref('')
+const results = ref<SearchDoc[]>([])
+const activeIndex = ref(0)
+
+const inputRef = ref<HTMLInputElement | null>(null)
+const buttonRef = ref<HTMLButtonElement | null>(null)
+
+const modifierKey = ref('Ctrl')
+
+const loadedIndex = shallowRef<any | null>(null)
+const loadedDocsById = shallowRef<Map<number, SearchDoc> | null>(null)
+const cache = new Map<LocaleKey, { index: any; docsById: Map<number, SearchDoc> }>()
+
+const placeholder = computed(() => (localeKey.value === 'zh' ? '搜索文档' : 'Search docs'))
+const cancelText = computed(() => (localeKey.value === 'zh' ? '关闭' : 'Close'))
+const loadingText = computed(() => (localeKey.value === 'zh' ? '正在加载索引…' : 'Loading index…'))
+const hintText = computed(() => (localeKey.value === 'zh' ? '输入关键词开始搜索' : 'Type to start searching'))
+const noResultsText = computed(() => (localeKey.value === 'zh' ? '未找到结果' : 'No results'))
+
+function isEditableTarget(target: EventTarget | null) {
+  const el = target as HTMLElement | null
+  if (!el) return false
+  const tag = el.tagName?.toLowerCase()
+  if (!tag) return false
+  return tag === 'input' || tag === 'textarea' || tag === 'select' || el.isContentEditable
+}
+
+async function loadLocaleIndex(key: LocaleKey) {
+  const cached = cache.get(key)
+  if (cached) {
+    loadedIndex.value = cached.index
+    loadedDocsById.value = cached.docsById
+    return
+  }
+
+  isLoading.value = true
+  error.value = null
+
+  try {
+    const res = await fetch(withBase(`/search-index.${key}.json`))
+    if (!res.ok) throw new Error(`HTTP ${res.status}`)
+    const payload = (await res.json()) as SearchIndexPayload
+
+    if (!payload || payload.locale !== key) throw new Error('Invalid index payload')
+
+    const [{ default: FlexSearch }, { createFlexSearchIndex }] = await Promise.all([
+      import('flexsearch'),
+      import('../../search/flexsearch.mjs')
+    ])
+
+    const index = createFlexSearchIndex(FlexSearch)
+    await Promise.all(Object.entries(payload.index).map(([k, v]) => index.import(k, v)))
+
+    const docsById = new Map<number, SearchDoc>()
+    for (const doc of payload.docs) docsById.set(doc.id, doc)
+
+    cache.set(key, { index, docsById })
+    loadedIndex.value = index
+    loadedDocsById.value = docsById
+  } catch (e) {
+    error.value = e instanceof Error ? e.message : String(e)
+    loadedIndex.value = null
+    loadedDocsById.value = null
+  } finally {
+    isLoading.value = false
+  }
+}
+
+async function ensureReady() {
+  await loadLocaleIndex(localeKey.value)
+}
+
+let searchTimer: number | undefined
+async function runSearch() {
+  const q = query.value.trim()
+  if (!q) {
+    results.value = []
+    activeIndex.value = 0
+    return
+  }
+
+  await ensureReady()
+  if (!loadedIndex.value || !loadedDocsById.value) {
+    results.value = []
+    activeIndex.value = 0
+    return
+  }
+
+  const ids = loadedIndex.value.search(q, 12) as number[]
+  const docsById = loadedDocsById.value
+  results.value = ids
+    .map((id) => docsById.get(id))
+    .filter((d): d is SearchDoc => Boolean(d))
+  activeIndex.value = 0
+}
+
+function navigate(url: string) {
+  router.go(withBase(url))
+  close()
+}
+
+async function open() {
+  isOpen.value = true
+  document.body.style.overflow = 'hidden'
+  await ensureReady()
+  await nextTick()
+  inputRef.value?.focus()
+}
+
+function close() {
+  isOpen.value = false
+  query.value = ''
+  results.value = []
+  activeIndex.value = 0
+  error.value = null
+  document.body.style.overflow = ''
+  buttonRef.value?.focus()
+}
+
+function onInputKeydown(e: KeyboardEvent) {
+  if (e.key === 'Escape') {
+    e.preventDefault()
+    close()
+    return
+  }
+
+  if (e.key === 'ArrowDown') {
+    e.preventDefault()
+    if (results.value.length > 0) {
+      activeIndex.value = (activeIndex.value + 1) % results.value.length
+    }
+    return
+  }
+
+  if (e.key === 'ArrowUp') {
+    e.preventDefault()
+    if (results.value.length > 0) {
+      activeIndex.value =
+        (activeIndex.value - 1 + results.value.length) % results.value.length
+    }
+    return
+  }
+
+  if (e.key === 'Enter') {
+    const hit = results.value[activeIndex.value]
+    if (hit) {
+      e.preventDefault()
+      navigate(hit.url)
+    }
+  }
+}
+
+let onGlobalKeydown: ((e: KeyboardEvent) => void) | null = null
+
+onMounted(() => {
+  modifierKey.value = /mac/i.test(navigator.platform) ? '⌘' : 'Ctrl'
+
+  onGlobalKeydown = (e: KeyboardEvent) => {
+    if ((e.ctrlKey || e.metaKey) && e.key.toLowerCase() === 'k') {
+      e.preventDefault()
+      if (!isOpen.value) open()
+      return
+    }
+
+    if (e.key === '/' && !e.ctrlKey && !e.metaKey && !e.altKey) {
+      if (isEditableTarget(e.target)) return
+      e.preventDefault()
+      if (!isOpen.value) open()
+    }
+  }
+
+  window.addEventListener('keydown', onGlobalKeydown)
+})
+
+onBeforeUnmount(() => {
+  if (onGlobalKeydown) window.removeEventListener('keydown', onGlobalKeydown)
+})
+
+watch(query, () => {
+  if (searchTimer) window.clearTimeout(searchTimer)
+  searchTimer = window.setTimeout(() => {
+    runSearch()
+  }, 60)
+})
+
+watch(
+  () => page.value.relativePath,
+  () => {
+    if (isOpen.value) close()
+  }
+)
+</script>
+
+<template>
+  <div class="DocSearch">
+    <button
+      ref="buttonRef"
+      type="button"
+      class="DocSearch-Button"
+      :aria-label="placeholder"
+      @click="open"
+    >
+      <svg class="DocSearch-Button-Icon" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2">
+        <circle cx="11" cy="11" r="7" />
+        <line x1="21" y1="21" x2="16.65" y2="16.65" />
+      </svg>
+      <span class="DocSearch-Button-Placeholder">{{ placeholder }}</span>
+      <span class="DocSearch-Button-Keys" aria-hidden="true">
+        <kbd class="DocSearch-Button-Key">{{ modifierKey }}</kbd>
+        <kbd class="DocSearch-Button-Key">K</kbd>
+      </span>
+    </button>
+
+    <Teleport to="body">
+      <div v-if="isOpen" class="DocSearch-Modal">
+        <div class="DocSearch-Overlay" @click="close" />
+
+        <div class="DocSearch-Container" role="dialog" aria-modal="true">
+          <div class="DocSearch-SearchBar">
+            <svg class="DocSearch-SearchIcon" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2">
+              <circle cx="11" cy="11" r="7" />
+              <line x1="21" y1="21" x2="16.65" y2="16.65" />
+            </svg>
+            <input
+              ref="inputRef"
+              v-model="query"
+              class="DocSearch-Input"
+              type="search"
+              :placeholder="placeholder"
+              :aria-label="placeholder"
+              @keydown="onInputKeydown"
+            />
+            <button type="button" class="DocSearch-Cancel" @click="close">{{ cancelText }}</button>
+          </div>
+
+          <div class="DocSearch-Body">
+            <div v-if="isLoading" class="DocSearch-Status">{{ loadingText }}</div>
+            <div v-else-if="error" class="DocSearch-Status DocSearch-Status--error">{{ error }}</div>
+            <div v-else-if="query.trim().length === 0" class="DocSearch-Status">{{ hintText }}</div>
+            <div v-else>
+              <ul v-if="results.length > 0" class="DocSearch-Results">
+                <li
+                  v-for="(item, i) in results"
+                  :key="item.id"
+                  :class="['DocSearch-Result', { active: i === activeIndex }]"
+                  @mousemove="activeIndex = i"
+                >
+                  <a
+                    class="DocSearch-Result-Link"
+                    :href="withBase(item.url)"
+                    @click.prevent="navigate(item.url)"
+                  >
+                    <div class="DocSearch-Result-Title">{{ item.title }}</div>
+                    <div v-if="item.excerpt" class="DocSearch-Result-Excerpt">
+                      {{ item.excerpt }}
+                    </div>
+                  </a>
+                </li>
+              </ul>
+              <div v-else class="DocSearch-Status">{{ noResultsText }}</div>
+            </div>
+          </div>
+        </div>
+      </div>
+    </Teleport>
+  </div>
+</template>
+
+<style scoped>
+.DocSearch {
+  display: flex;
+  align-items: center;
+}
+
+.DocSearch-Button {
+  display: inline-flex;
+  align-items: center;
+  gap: 10px;
+  height: 36px;
+  padding: 0 12px;
+  border: 1px solid var(--vp-c-border);
+  border-radius: var(--vp-radius-full);
+  background: var(--vp-c-bg-soft);
+  color: var(--vp-c-text-2);
+  cursor: pointer;
+  transition: all var(--vp-transition-color);
+}
+
+.DocSearch-Button:hover {
+  border-color: var(--vp-c-primary);
+  color: var(--vp-c-text-1);
+  background: var(--vp-c-bg);
+}
+
+.DocSearch-Button-Icon {
+  width: 16px;
+  height: 16px;
+  flex: 0 0 auto;
+}
+
+.DocSearch-Button-Placeholder {
+  font-size: var(--vp-font-size-sm);
+  white-space: nowrap;
+}
+
+.DocSearch-Button-Keys {
+  display: inline-flex;
+  gap: 4px;
+  margin-left: 8px;
+}
+
+.DocSearch-Button-Key {
+  font-family: var(--vp-font-family-mono);
+  font-size: 12px;
+  line-height: 1;
+  padding: 4px 6px;
+  border-radius: var(--vp-radius-sm);
+  border: 1px solid var(--vp-c-divider);
+  background: var(--vp-c-bg);
+  color: var(--vp-c-text-2);
+}
+
+.DocSearch-Modal {
+  position: fixed;
+  inset: 0;
+  z-index: var(--vp-z-index-modal);
+}
+
+.DocSearch-Overlay {
+  position: absolute;
+  inset: 0;
+  background: rgba(0, 0, 0, 0.5);
+}
+
+.DocSearch-Container {
+  position: relative;
+  width: min(720px, calc(100vw - 32px));
+  margin: 10vh auto 0;
+  background: var(--vp-c-bg);
+  border: 1px solid var(--vp-c-divider);
+  border-radius: var(--vp-radius-xl);
+  box-shadow: var(--vp-shadow-xl);
+  overflow: hidden;
+}
+
+.DocSearch-SearchBar {
+  display: flex;
+  align-items: center;
+  gap: 12px;
+  padding: 12px 14px;
+  border-bottom: 1px solid var(--vp-c-divider);
+  background: var(--vp-c-bg-soft);
+}
+
+.DocSearch-SearchIcon {
+  width: 18px;
+  height: 18px;
+  color: var(--vp-c-text-3);
+  flex: 0 0 auto;
+}
+
+.DocSearch-Input {
+  flex: 1;
+  height: 40px;
+  padding: 0 12px;
+  border: 1px solid var(--vp-c-border);
+  border-radius: var(--vp-radius-md);
+  background: var(--vp-c-bg);
+  color: var(--vp-c-text-1);
+  font-size: 14px;
+}
+
+.DocSearch-Input:focus-visible {
+  outline: 2px solid var(--vp-c-primary);
+  outline-offset: 2px;
+}
+
+.DocSearch-Cancel {
+  height: 40px;
+  padding: 0 12px;
+  border: 1px solid var(--vp-c-border);
+  border-radius: var(--vp-radius-md);
+  background: var(--vp-c-bg);
+  color: var(--vp-c-text-2);
+  cursor: pointer;
+  transition: all var(--vp-transition-color);
+}
+
+.DocSearch-Cancel:hover {
+  border-color: var(--vp-c-primary);
+  color: var(--vp-c-text-1);
+  background: var(--vp-c-bg-soft);
+}
+
+.DocSearch-Body {
+  max-height: 60vh;
+  overflow: auto;
+}
+
+.DocSearch-Status {
+  padding: 18px 16px;
+  color: var(--vp-c-text-2);
+  font-size: var(--vp-font-size-sm);
+}
+
+.DocSearch-Status--error {
+  color: #ef4444;
+}
+
+.DocSearch-Results {
+  list-style: none;
+  margin: 0;
+  padding: 8px;
+}
+
+.DocSearch-Result {
+  border-radius: var(--vp-radius-lg);
+  transition: background var(--vp-transition-color);
+}
+
+.DocSearch-Result.active,
+.DocSearch-Result:hover {
+  background: var(--vp-c-bg-soft);
+}
+
+.DocSearch-Result-Link {
+  display: block;
+  padding: 12px 12px;
+  text-decoration: none;
+}
+
+.DocSearch-Result-Title {
+  font-weight: 600;
+  color: var(--vp-c-text-1);
+  font-size: 14px;
+}
+
+.DocSearch-Result-Excerpt {
+  margin-top: 4px;
+  color: var(--vp-c-text-3);
+  font-size: 12px;
+  line-height: 1.4;
+}
+
+@media (max-width: 768px) {
+  .DocSearch-Button-Placeholder,
+  .DocSearch-Button-Keys {
+    display: none;
+  }
+
+  .DocSearch-Button {
+    width: 40px;
+    justify-content: center;
+    padding: 0;
+  }
+
+  .DocSearch-Container {
+    margin-top: 6vh;
+  }
+}
+</style>

docs/.vitepress/theme/components/HeroAnimation.vue

@@ -0,0 +1,214 @@
+<template>
+  <div class="hero-animation-container" :class="{ 'is-visible': isVisible }">
+    <div class="glow-bg"></div>
+    <svg viewBox="0 0 400 320" class="hero-svg" preserveAspectRatio="xMidYMid meet">
+      <defs>
+        <filter id="glow" x="-20%" y="-20%" width="140%" height="140%">
+          <feGaussianBlur stdDeviation="3.5" result="coloredBlur"/>
+          <feMerge>
+            <feMergeNode in="coloredBlur"/>
+            <feMergeNode in="SourceGraphic"/>
+          </feMerge>
+        </filter>
+        <linearGradient id="pathGrad" x1="0%" y1="0%" x2="100%" y2="0%">
+          <stop offset="0%" stop-color="var(--vp-c-brand-1)" stop-opacity="0" />
+          <stop offset="50%" stop-color="var(--vp-c-brand-1)" stop-opacity="0.5" />
+          <stop offset="100%" stop-color="var(--vp-c-brand-1)" stop-opacity="0" />
+        </linearGradient>
+      </defs>
+
+      <!-- Connection Lines -->
+      <g class="data-paths">
+        <path v-for="(path, i) in paths" :key="'path-'+i" :d="path" class="connection-path" />
+        <circle v-for="(path, i) in paths" :key="'dot-'+i" r="2" class="data-pulse">
+          <animateMotion :dur="2 + i * 0.4 + 's'" repeatCount="indefinite" :path="path" />
+        </circle>
+      </g>
+
+      <!-- Orbit Rings -->
+      <g class="orbit-rings">
+        <circle cx="200" cy="160" r="130" class="orbit-ring ring-outer" />
+        <circle cx="200" cy="160" r="95" class="orbit-ring ring-inner" />
+      </g>
+
+      <!-- Agent Nodes -->
+      <g v-for="(agent, i) in agents" :key="'agent-'+i" class="agent-node" :style="{ '--delay': i * 0.4 + 's' }">
+        <g class="agent-group" :style="{ transform: `translate(${agent.x}px, ${agent.y}px)` }">
+          <circle r="8" :fill="agent.color" class="agent-circle" filter="url(#glow)" />
+          <circle r="12" :stroke="agent.color" fill="none" class="agent-halo" />
+          <text y="22" text-anchor="middle" class="agent-label">{{ agent.name }}</text>
+        </g>
+      </g>
+
+      <!-- Central Core -->
+      <g class="central-core" transform="translate(200, 160)">
+        <circle r="40" class="core-bg" />
+        <circle r="32" fill="var(--vp-c-brand-1)" filter="url(#glow)" class="core-inner" />
+        <text y="8" text-anchor="middle" class="core-text">CCW</text>
+        
+        <!-- Scanning Effect -->
+        <path d="M-32 0 A32 32 0 0 1 32 0" fill="none" stroke="white" stroke-width="2" stroke-linecap="round" class="core-scanner" />
+      </g>
+    </svg>
+  </div>
+</template>
+
+<script setup>
+import { ref, onMounted } from 'vue'
+
+const isVisible = ref(false)
+
+onMounted(() => {
+  setTimeout(() => {
+    isVisible.value = true
+  }, 100)
+})
+
+const agents = [
+  { name: 'Analyze', x: 200, y: 35, color: '#3B82F6' },
+  { name: 'Plan', x: 315, y: 110, color: '#10B981' },
+  { name: 'Code', x: 285, y: 245, color: '#8B5CF6' },
+  { name: 'Test', x: 115, y: 245, color: '#F59E0B' },
+  { name: 'Review', x: 85, y: 110, color: '#EF4444' }
+]
+
+const paths = [
+  'M200,160 L200,35',
+  'M200,160 L315,110',
+  'M200,160 L285,245',
+  'M200,160 L115,245',
+  'M200,160 L85,110',
+  'M200,35 Q260,35 315,110',
+  'M315,110 Q315,180 285,245',
+  'M285,245 Q200,285 115,245',
+  'M115,245 Q85,180 85,110',
+  'M85,110 Q85,35 200,35'
+]
+</script>
+
+<style scoped>
+.hero-animation-container {
+  width: 100%;
+  max-width: 480px;
+  position: relative;
+  opacity: 0;
+  transform: scale(0.95);
+  transition: all 1s cubic-bezier(0.2, 0.8, 0.2, 1);
+}
+
+.hero-animation-container.is-visible {
+  opacity: 1;
+  transform: scale(1);
+}
+
+.glow-bg {
+  position: absolute;
+  top: 50%;
+  left: 50%;
+  width: 150px;
+  height: 150px;
+  background: var(--vp-c-brand-1);
+  filter: blur(80px);
+  opacity: 0.15;
+  transform: translate(-50%, -50%);
+  pointer-events: none;
+}
+
+.hero-svg {
+  width: 100%;
+  height: auto;
+  overflow: visible;
+}
+
+.orbit-ring {
+  fill: none;
+  stroke: var(--vp-c-brand-1);
+  stroke-width: 0.5;
+  opacity: 0.1;
+  stroke-dasharray: 4 4;
+}
+
+.ring-outer { animation: rotate 60s linear infinite; transform-origin: 200px 160px; }
+.ring-inner { animation: rotate 40s linear infinite reverse; transform-origin: 200px 160px; }
+
+.connection-path {
+  fill: none;
+  stroke: var(--vp-c-brand-1);
+  stroke-width: 0.8;
+  opacity: 0.05;
+}
+
+.data-pulse {
+  fill: var(--vp-c-brand-2);
+  filter: drop-shadow(0 0 4px var(--vp-c-brand-2));
+}
+
+.agent-group {
+  transition: all 0.3s ease;
+}
+
+.agent-circle {
+  transition: all 0.3s ease;
+}
+
+.agent-halo {
+  opacity: 0.2;
+  animation: agent-pulse 2s ease-in-out infinite;
+  transform-origin: center;
+}
+
+.agent-label {
+  font-size: 10px;
+  fill: var(--vp-c-text-2);
+  font-weight: 600;
+  text-transform: uppercase;
+  letter-spacing: 0.05em;
+  opacity: 0.7;
+}
+
+.core-bg {
+  fill: var(--vp-c-bg-soft);
+  stroke: var(--vp-c-brand-soft);
+  stroke-width: 1;
+}
+
+.core-inner {
+  opacity: 0.8;
+}
+
+.core-text {
+  font-size: 14px;
+  font-weight: 800;
+  fill: white;
+  letter-spacing: 0.05em;
+}
+
+.core-scanner {
+  animation: rotate 3s linear infinite;
+  opacity: 0.6;
+}
+
+@keyframes rotate {
+  from { transform: rotate(0deg); }
+  to { transform: rotate(360deg); }
+}
+
+@keyframes agent-pulse {
+  0%, 100% { transform: scale(1); opacity: 0.2; }
+  50% { transform: scale(1.3); opacity: 0.1; }
+}
+
+.agent-node {
+  animation: agent-float 4s ease-in-out infinite;
+  animation-delay: var(--delay);
+}
+
+@keyframes agent-float {
+  0%, 100% { transform: translateY(0); }
+  50% { transform: translateY(-5px); }
+}
+
+.hero-animation-container:hover .agent-circle {
+  filter: blur(2px) brightness(1.5);
+}
+</style>

docs/.vitepress/theme/components/PageToc.vue

@@ -0,0 +1,184 @@
+<script setup lang="ts">
+import { computed, onMounted, ref } from 'vue'
+import { useData } from 'vitepress'
+
+const { page } = useData()
+
+interface TocItem {
+  id: string
+  text: string
+  level: number
+  children?: TocItem[]
+}
+
+const toc = computed<TocItem[]>(() => {
+  return page.value.headers || []
+})
+
+const activeId = ref('')
+
+const onItemClick = (id: string) => {
+  activeId.value = id
+  const element = document.getElementById(id)
+  if (element) {
+    element.scrollIntoView({ behavior: 'smooth', block: 'start' })
+  }
+}
+
+onMounted(() => {
+  // Update active heading on scroll
+  const observer = new IntersectionObserver(
+    (entries) => {
+      entries.forEach((entry) => {
+        if (entry.isIntersecting) {
+          activeId.value = entry.target.id
+        }
+      })
+    },
+    { rootMargin: '-80px 0px -80% 0px' }
+  )
+
+  page.value.headers.forEach((header) => {
+    const element = document.getElementById(header.id)
+    if (element) {
+      observer.observe(element)
+    }
+  })
+
+  return () => observer.disconnect()
+})
+</script>
+
+<template>
+  <nav v-if="toc.length > 0" class="page-toc" aria-label="Page navigation">
+    <div class="toc-header">On this page</div>
+    <ul class="toc-list">
+      <li
+        v-for="item in toc"
+        :key="item.id"
+        :class="['toc-item', `toc-level-${item.level}`, { active: item.id === activeId }]"
+      >
+        <a
+          :href="`#${item.id}`"
+          class="toc-link"
+          @click.prevent="onItemClick(item.id)"
+        >
+          {{ item.text }}
+        </a>
+        <ul v-if="item.children && item.children.length > 0" class="toc-list toc-sublist">
+          <li
+            v-for="child in item.children"
+            :key="child.id"
+            :class="['toc-item', `toc-level-${child.level}`, { active: child.id === activeId }]"
+          >
+            <a
+              :href="`#${child.id}`"
+              class="toc-link"
+              @click.prevent="onItemClick(child.id)"
+            >
+              {{ child.text }}
+            </a>
+          </li>
+        </ul>
+      </li>
+    </ul>
+  </nav>
+</template>
+
+<style scoped>
+.page-toc {
+  position: sticky;
+  top: calc(var(--vp-nav-height) + 24px);
+  max-height: calc(100vh - var(--vp-nav-height) - 48px);
+  overflow-y: auto;
+  padding: 16px;
+  background: var(--vp-c-bg-soft);
+  border-radius: var(--vp-radius-lg);
+  border: 1px solid var(--vp-c-divider);
+}
+
+.toc-header {
+  font-size: var(--vp-font-size-sm);
+  font-weight: 600;
+  color: var(--vp-c-text-1);
+  margin-bottom: 12px;
+  padding-bottom: 8px;
+  border-bottom: 1px solid var(--vp-c-divider);
+  text-transform: uppercase;
+  letter-spacing: 0.05em;
+}
+
+.toc-list {
+  list-style: none;
+  margin: 0;
+  padding: 0;
+}
+
+.toc-sublist {
+  margin-left: 12px;
+  padding-left: 12px;
+  border-left: 1px solid var(--vp-c-divider);
+}
+
+.toc-item {
+  margin: 4px 0;
+}
+
+.toc-link {
+  display: block;
+  padding: 4px 8px;
+  color: var(--vp-c-text-2);
+  font-size: var(--vp-font-size-sm);
+  text-decoration: none;
+  border-left: 2px solid transparent;
+  transition: all var(--vp-transition-color);
+  border-radius: 0 var(--vp-radius-sm) var(--vp-radius-sm) 0;
+}
+
+.toc-link:hover {
+  color: var(--vp-c-primary);
+  background: var(--vp-c-bg-soft);
+}
+
+.toc-item.active > .toc-link {
+  color: var(--vp-c-primary);
+  border-left-color: var(--vp-c-primary);
+  background: var(--vp-c-bg-mute);
+  font-weight: 500;
+}
+
+.toc-level-3 .toc-link {
+  font-size: 13px;
+  padding-left: 16px;
+}
+
+.toc-level-4 .toc-link {
+  font-size: 12px;
+  padding-left: 20px;
+}
+
+/* Hide on mobile */
+@media (max-width: 1024px) {
+  .page-toc {
+    display: none;
+  }
+}
+
+/* Scrollbar styling for TOC */
+.page-toc::-webkit-scrollbar {
+  width: 4px;
+}
+
+.page-toc::-webkit-scrollbar-track {
+  background: transparent;
+}
+
+.page-toc::-webkit-scrollbar-thumb {
+  background: var(--vp-c-divider);
+  border-radius: var(--vp-radius-full);
+}
+
+.page-toc::-webkit-scrollbar-thumb:hover {
+  background: var(--vp-c-text-3);
+}
+</style>

docs/.vitepress/theme/components/SkipLink.vue

@@ -0,0 +1,42 @@
+<script setup lang="ts">
+/**
+ * Skip Link Component
+ * Accessibility feature allowing keyboard users to skip to main content
+ */
+</script>
+
+<template>
+  <a href="#VPContent" class="skip-link">
+    Skip to main content
+  </a>
+</template>
+
+<style scoped>
+.skip-link {
+  position: absolute;
+  top: -100px;
+  left: 0;
+  padding: 8px 16px;
+  background: var(--vp-c-primary);
+  color: white;
+  text-decoration: none;
+  z-index: 9999;
+  transition: top 0.3s ease;
+  font-size: 14px;
+  font-weight: 500;
+}
+
+.skip-link:focus {
+  top: 0;
+}
+
+.skip-link:hover {
+  background: var(--vp-c-primary-600);
+}
+
+@media print {
+  .skip-link {
+    display: none;
+  }
+}
+</style>

docs/.vitepress/theme/components/ThemeSwitcher.vue

@@ -0,0 +1,108 @@
+<script setup lang="ts">
+import { ref, onMounted } from 'vue'
+
+const currentTheme = ref<'blue' | 'green' | 'orange' | 'purple'>('blue')
+
+const themes = [
+  { id: 'blue', name: 'Blue', color: '#3b82f6' },
+  { id: 'green', name: 'Green', color: '#10b981' },
+  { id: 'orange', name: 'Orange', color: '#f59e0b' },
+  { id: 'purple', name: 'Purple', color: '#8b5cf6' }
+]
+
+const setTheme = (themeId: typeof currentTheme.value) => {
+  currentTheme.value = themeId
+  document.documentElement.setAttribute('data-theme', themeId)
+  localStorage.setItem('ccw-theme', themeId)
+}
+
+onMounted(() => {
+  const savedTheme = localStorage.getItem('ccw-theme') as typeof currentTheme.value
+  if (savedTheme && themes.find(t => t.id === savedTheme)) {
+    setTheme(savedTheme)
+  }
+})
+</script>
+
+<template>
+  <div class="theme-switcher">
+    <div class="theme-buttons">
+      <button
+        v-for="theme in themes"
+        :key="theme.id"
+        :class="['theme-button', { active: currentTheme === theme.id }]"
+        :style="{ '--theme-color': theme.color }"
+        :aria-label="`Switch to ${theme.name} theme`"
+        :title="theme.name"
+        @click="setTheme(theme.id)"
+      >
+        <span class="theme-dot"></span>
+      </button>
+    </div>
+  </div>
+</template>
+
+<style scoped>
+.theme-switcher {
+  display: flex;
+  align-items: center;
+  gap: 8px;
+}
+
+.theme-buttons {
+  display: flex;
+  gap: 4px;
+  padding: 4px;
+  background: var(--vp-c-bg-soft);
+  border-radius: var(--vp-radius-full);
+}
+
+.theme-button {
+  position: relative;
+  width: 32px;
+  height: 32px;
+  border: none;
+  background: transparent;
+  border-radius: var(--vp-radius-full);
+  cursor: pointer;
+  display: flex;
+  align-items: center;
+  justify-content: center;
+  transition: all var(--vp-transition-color);
+}
+
+.theme-button:hover {
+  background: var(--vp-c-bg-mute);
+}
+
+.theme-button.active {
+  background: var(--vp-c-bg);
+  box-shadow: var(--vp-shadow-sm);
+}
+
+.theme-dot {
+  width: 16px;
+  height: 16px;
+  border-radius: var(--vp-radius-full);
+  background: var(--theme-color);
+  border: 2px solid transparent;
+  transition: all var(--vp-transition-color);
+}
+
+.theme-button.active .theme-dot {
+  border-color: var(--vp-c-text-1);
+  transform: scale(1.1);
+}
+
+@media (max-width: 768px) {
+  .theme-button {
+    width: 36px;
+    height: 36px;
+  }
+
+  .theme-dot {
+    width: 20px;
+    height: 20px;
+  }
+}
+</style>

docs/.vitepress/theme/components/WorkflowAnimation.vue

@@ -0,0 +1,202 @@
+<template>
+  <div class="workflow-animation">
+    <div class="workflow-container">
+      <div class="workflow-node coordinator">
+        <div class="node-icon">🎯</div>
+        <div class="node-label">Coordinator</div>
+      </div>
+      
+      <div class="workflow-paths">
+        <svg class="path-svg" viewBox="0 0 400 200">
+          <!-- Spec Path -->
+          <path class="flow-path path-spec" d="M50,100 Q150,20 250,50" fill="none" stroke="#3B82F6" stroke-width="2"/>
+          <circle class="flow-dot dot-spec" r="6" fill="#3B82F6">
+            <animateMotion dur="3s" repeatCount="indefinite" path="M50,100 Q150,20 250,50"/>
+          </circle>
+          
+          <!-- Impl Path -->
+          <path class="flow-path path-impl" d="M50,100 Q150,100 250,100" fill="none" stroke="#10B981" stroke-width="2"/>
+          <circle class="flow-dot dot-impl" r="6" fill="#10B981">
+            <animateMotion dur="2.5s" repeatCount="indefinite" path="M50,100 Q150,100 250,100"/>
+          </circle>
+          
+          <!-- Test Path -->
+          <path class="flow-path path-test" d="M50,100 Q150,180 250,150" fill="none" stroke="#F59E0B" stroke-width="2"/>
+          <circle class="flow-dot dot-test" r="6" fill="#F59E0B">
+            <animateMotion dur="3.5s" repeatCount="indefinite" path="M50,100 Q150,180 250,150"/>
+          </circle>
+        </svg>
+      </div>
+      
+      <div class="workflow-nodes">
+        <div class="workflow-node analyst">
+          <div class="node-icon">📊</div>
+          <div class="node-label">Analyst</div>
+        </div>
+        <div class="workflow-node writer">
+          <div class="node-icon">✍️</div>
+          <div class="node-label">Writer</div>
+        </div>
+        <div class="workflow-node executor">
+          <div class="node-icon">⚡</div>
+          <div class="node-label">Executor</div>
+        </div>
+        <div class="workflow-node tester">
+          <div class="node-icon">🧪</div>
+          <div class="node-label">Tester</div>
+        </div>
+      </div>
+    </div>
+    
+    <div class="workflow-legend">
+      <div class="legend-item"><span class="dot spec"></span> Spec Phase</div>
+      <div class="legend-item"><span class="dot impl"></span> Impl Phase</div>
+      <div class="legend-item"><span class="dot test"></span> Test Phase</div>
+    </div>
+  </div>
+</template>
+
+<script setup>
+import { onMounted, onUnmounted } from 'vue'
+
+onMounted(() => {
+  // Add animation class after mount
+  document.querySelector('.workflow-animation')?.classList.add('animate')
+})
+</script>
+
+<style scoped>
+.workflow-animation {
+  padding: 2rem;
+  background: var(--vp-c-bg-soft);
+  border: 1px solid var(--vp-c-divider);
+  border-radius: 24px;
+  margin: 2rem 0;
+  overflow: hidden;
+}
+
+.workflow-container {
+  display: flex;
+  align-items: center;
+  justify-content: space-around;
+  flex-wrap: wrap;
+  gap: 2rem;
+  min-height: 200px;
+}
+
+.workflow-paths {
+  flex: 1.2;
+  min-width: 280px;
+  max-width: 450px;
+}
+
+.path-svg {
+  width: 100%;
+  height: auto;
+}
+
+.flow-path {
+  stroke-dasharray: 6, 6;
+  opacity: 0.3;
+  animation: dash 30s linear infinite;
+}
+
+@keyframes dash {
+  to {
+    stroke-dashoffset: -120;
+  }
+}
+
+.flow-dot {
+  filter: drop-shadow(0 0 4px currentColor);
+}
+
+.workflow-nodes {
+  display: grid;
+  grid-template-columns: repeat(2, 1fr);
+  gap: 1rem;
+  flex: 0.8;
+  min-width: 200px;
+}
+
+.workflow-node {
+  display: flex;
+  flex-direction: column;
+  align-items: center;
+  padding: 1.25rem;
+  background: var(--vp-c-bg);
+  border: 1px solid var(--vp-c-divider);
+  border-radius: 16px;
+  box-shadow: var(--vp-shadow-sm);
+  transition: all 0.3s cubic-bezier(0.4, 0, 0.2, 1);
+}
+
+.workflow-node:hover {
+  transform: translateY(-4px);
+  border-color: var(--vp-c-brand-1);
+  box-shadow: var(--vp-shadow-md);
+}
+
+.node-icon {
+  font-size: 2rem;
+  margin-bottom: 0.5rem;
+}
+
+.node-label {
+  font-size: 0.85rem;
+  font-weight: 600;
+  color: var(--vp-c-text-1);
+}
+
+.workflow-node.coordinator {
+  grid-column: span 2;
+  background: linear-gradient(135deg, var(--vp-c-brand-1), var(--vp-c-brand-2));
+  border: none;
+  color: white;
+}
+
+.workflow-node.coordinator .node-label {
+  color: white;
+}
+
+.workflow-legend {
+  display: flex;
+  justify-content: center;
+  gap: 2.5rem;
+  margin-top: 2rem;
+  flex-wrap: wrap;
+}
+
+.legend-item {
+  display: flex;
+  align-items: center;
+  gap: 0.75rem;
+  font-size: 0.85rem;
+  font-weight: 500;
+  color: var(--vp-c-text-2);
+}
+
+.dot {
+  width: 10px;
+  height: 10px;
+  border-radius: 50%;
+}
+
+.dot.spec { background: var(--vp-c-brand-1); }
+.dot.impl { background: var(--vp-c-secondary-500); }
+.dot.test { background: var(--vp-c-accent-400); }
+
+@media (max-width: 768px) {
+  .workflow-animation {
+    padding: 1.5rem;
+  }
+  
+  .workflow-container {
+    flex-direction: column;
+  }
+  
+  .workflow-nodes {
+    width: 100%;
+  }
+}
+</style>

docs/.vitepress/theme/index.ts

@@ -0,0 +1,27 @@
+import DefaultTheme from 'vitepress/theme'
+import ThemeSwitcher from './components/ThemeSwitcher.vue'
+import DocSearch from './components/DocSearch.vue'
+import DarkModeToggle from './components/DarkModeToggle.vue'
+import CopyCodeButton from './components/CopyCodeButton.vue'
+import Breadcrumb from './components/Breadcrumb.vue'
+import PageToc from './components/PageToc.vue'
+import ProfessionalHome from './components/ProfessionalHome.vue'
+import Layout from './layouts/Layout.vue'
+import './styles/variables.css'
+import './styles/custom.css'
+import './styles/mobile.css'
+
+export default {
+  extends: DefaultTheme,
+  Layout,
+  enhanceApp({ app, router, siteData }) {
+    // Register global components
+    app.component('ThemeSwitcher', ThemeSwitcher)
+    app.component('DocSearch', DocSearch)
+    app.component('DarkModeToggle', DarkModeToggle)
+    app.component('CopyCodeButton', CopyCodeButton)
+    app.component('Breadcrumb', Breadcrumb)
+    app.component('PageToc', PageToc)
+    app.component('ProfessionalHome', ProfessionalHome)
+  }
+}

docs/.vitepress/theme/layouts/Layout.vue

@@ -0,0 +1,153 @@
+<script setup lang="ts">
+import DefaultTheme from 'vitepress/theme'
+import { onBeforeUnmount, onMounted } from 'vue'
+
+let mediaQuery: MediaQueryList | null = null
+let systemThemeChangeHandler: (() => void) | null = null
+let storageHandler: ((e: StorageEvent) => void) | null = null
+
+function applyTheme() {
+  const savedTheme = localStorage.getItem('ccw-theme') || 'blue'
+  document.documentElement.setAttribute('data-theme', savedTheme)
+}
+
+function applyColorMode() {
+  const mode = localStorage.getItem('ccw-color-mode') || 'auto'
+  const prefersDark = window.matchMedia('(prefers-color-scheme: dark)').matches
+  const isDark = mode === 'dark' || (mode === 'auto' && prefersDark)
+  document.documentElement.classList.toggle('dark', isDark)
+}
+
+onMounted(() => {
+  applyTheme()
+  applyColorMode()
+
+  mediaQuery = window.matchMedia('(prefers-color-scheme: dark)')
+  systemThemeChangeHandler = () => {
+    const mode = localStorage.getItem('ccw-color-mode') || 'auto'
+    if (mode === 'auto') applyColorMode()
+  }
+  mediaQuery.addEventListener('change', systemThemeChangeHandler)
+
+  storageHandler = (e: StorageEvent) => {
+    if (e.key === 'ccw-theme') applyTheme()
+    if (e.key === 'ccw-color-mode') applyColorMode()
+  }
+  window.addEventListener('storage', storageHandler)
+})
+
+onBeforeUnmount(() => {
+  if (mediaQuery && systemThemeChangeHandler) {
+    mediaQuery.removeEventListener('change', systemThemeChangeHandler)
+  }
+  if (storageHandler) window.removeEventListener('storage', storageHandler)
+})
+</script>
+
+<template>
+  <DefaultTheme.Layout>
+    <template #home-hero-after>
+      <div class="hero-extensions">
+        <div class="hero-stats">
+          <div class="stat-item">
+            <div class="stat-value">27+</div>
+            <div class="stat-label">Built-in Skills</div>
+          </div>
+          <div class="stat-item">
+            <div class="stat-value">10+</div>
+            <div class="stat-label">Agent Types</div>
+          </div>
+          <div class="stat-item">
+            <div class="stat-value">4</div>
+            <div class="stat-label">Workflow Levels</div>
+          </div>
+        </div>
+      </div>
+    </template>
+
+    <template #layout-top>
+      <a href="#VPContent" class="skip-link">Skip to main content</a>
+    </template>
+
+    <template #nav-bar-content-after>
+      <div class="nav-extensions">
+        <DocSearch />
+        <DarkModeToggle />
+        <ThemeSwitcher />
+      </div>
+    </template>
+  </DefaultTheme.Layout>
+</template>
+
+<style scoped>
+.hero-extensions {
+  margin-top: 40px;
+  text-align: center;
+}
+
+.hero-stats {
+  display: flex;
+  justify-content: center;
+  gap: 48px;
+  flex-wrap: wrap;
+}
+
+.stat-item {
+  text-align: center;
+}
+
+.stat-value {
+  font-size: 32px;
+  font-weight: 700;
+  color: var(--vp-c-primary);
+}
+
+.stat-label {
+  font-size: 14px;
+  color: var(--vp-c-text-2);
+  margin-top: 4px;
+}
+
+.nav-extensions {
+  display: flex;
+  align-items: center;
+  gap: 12px;
+  margin-left: auto;
+  padding-left: 16px;
+}
+
+.skip-link {
+  position: absolute;
+  top: -100px;
+  left: 0;
+  padding: 8px 16px;
+  background: var(--vp-c-primary);
+  color: white;
+  text-decoration: none;
+  z-index: 9999;
+  transition: top 0.3s;
+}
+
+.skip-link:focus {
+  top: 0;
+}
+
+@media (max-width: 768px) {
+  .hero-stats {
+    gap: 24px;
+  }
+
+  .stat-value {
+    font-size: 24px;
+  }
+
+  .stat-label {
+    font-size: 12px;
+  }
+
+  .nav-extensions {
+    gap: 8px;
+    padding-left: 8px;
+  }
+}
+</style>

docs/.vitepress/theme/styles/custom.css

@@ -0,0 +1,352 @@
+/**
+ * VitePress Custom Styles
+ * Overrides and extensions for default VitePress theme
+ * Design System: ui-ux-pro-max — dark-mode-first, developer-focused
+ */
+
+/* ============================================
+ * Global Theme Variables
+ * ============================================ */
+:root {
+  --vp-c-brand: var(--vp-c-primary);
+  --vp-c-brand-light: var(--vp-c-primary-300);
+  --vp-c-brand-lighter: var(--vp-c-primary-200);
+  --vp-c-brand-dark: var(--vp-c-primary-700);
+  --vp-c-brand-darker: var(--vp-c-primary-800);
+
+  --vp-home-hero-name-color: var(--vp-c-primary);
+  --vp-home-hero-name-background: linear-gradient(120deg, var(--vp-c-primary-500) 30%, var(--vp-c-secondary-500));
+
+  --vp-button-brand-bg: var(--vp-c-primary);
+  --vp-button-brand-hover-bg: var(--vp-c-primary-600);
+  --vp-button-brand-active-bg: var(--vp-c-primary-700);
+
+  --vp-custom-block-tip-bg: var(--vp-c-primary-50);
+  --vp-custom-block-tip-border: var(--vp-c-primary-200);
+  --vp-custom-block-tip-text: var(--vp-c-primary-700);
+
+  --vp-custom-block-warning-bg: var(--vp-c-accent-50);
+  --vp-custom-block-warning-border: var(--vp-c-accent-200);
+  --vp-custom-block-warning-text: var(--vp-c-accent-700);
+
+  --vp-custom-block-danger-bg: #fef2f2;
+  --vp-custom-block-danger-border: #fecaca;
+  --vp-custom-block-danger-text: #b91c1c;
+
+  /* Layout Width Adjustments */
+  --vp-layout-max-width: 1600px;
+  --vp-content-width: 1000px;
+  --vp-sidebar-width: 272px;
+}
+
+.dark {
+  --vp-custom-block-tip-bg: rgba(59, 130, 246, 0.1);
+  --vp-custom-block-tip-border: rgba(59, 130, 246, 0.3);
+  --vp-custom-block-tip-text: var(--vp-c-primary-300);
+
+  --vp-custom-block-warning-bg: rgba(217, 119, 6, 0.1);
+  --vp-custom-block-warning-border: rgba(217, 119, 6, 0.3);
+  --vp-custom-block-warning-text: var(--vp-c-accent-300);
+
+  --vp-custom-block-danger-bg: rgba(185, 28, 28, 0.1);
+  --vp-custom-block-danger-border: rgba(185, 28, 28, 0.3);
+  --vp-custom-block-danger-text: #fca5a5;
+}
+
+/* ============================================
+ * Layout Container Adjustments
+ * ============================================ */
+.VPDoc .content-container {
+  max-width: var(--vp-content-width);
+  padding: 0 32px;
+}
+
+/* Adjust sidebar and content layout */
+.VPDoc {
+  padding-left: var(--vp-sidebar-width);
+}
+
+/* Right side outline (TOC) adjustments */
+.VPDocOutline {
+  padding-left: 24px;
+}
+
+.VPDocOutline .outline-link {
+  font-size: 13px;
+  line-height: 1.6;
+  padding: 4px 8px;
+}
+
+/* ============================================
+ * Home Page Override
+ * ============================================ */
+.VPHome {
+  padding-bottom: 0;
+}
+
+.VPHomeHero {
+  padding: 80px 24px;
+  background: linear-gradient(180deg, var(--vp-c-bg-soft) 0%, var(--vp-c-bg) 100%);
+}
+
+/* ============================================
+ * Documentation Content Typography
+ * ============================================ */
+.vp-doc h1 {
+  font-weight: 800;
+  letter-spacing: -0.02em;
+  margin-bottom: 1.5rem;
+}
+
+.vp-doc h2 {
+  font-weight: 700;
+  margin-top: 3rem;
+  padding-top: 2rem;
+  letter-spacing: -0.01em;
+  border-top: 1px solid var(--vp-c-divider);
+}
+
+.vp-doc h2:first-of-type {
+  margin-top: 1.5rem;
+  border-top: none;
+}
+
+.vp-doc h3 {
+  font-weight: 600;
+  margin-top: 2.5rem;
+}
+
+.vp-doc h4 {
+  font-weight: 600;
+  margin-top: 2rem;
+}
+
+.vp-doc p {
+  line-height: 1.8;
+  margin: 1.25rem 0;
+}
+
+.vp-doc ul,
+.vp-doc ol {
+  margin: 1.25rem 0;
+  padding-left: 1.5rem;
+}
+
+.vp-doc li {
+  line-height: 1.8;
+  margin: 0.5rem 0;
+}
+
+.vp-doc li + li {
+  margin-top: 0.5rem;
+}
+
+/* Better spacing for code blocks in lists */
+.vp-doc li > code {
+  margin: 0 2px;
+}
+
+/* ============================================
+ * Command Reference Specific Styles
+ * ============================================ */
+.vp-doc h3[id^="ccw"],
+.vp-doc h3[id^="workflow"],
+.vp-doc h3[id^="issue"],
+.vp-doc h3[id^="cli"],
+.vp-doc h3[id^="memory"] {
+  scroll-margin-top: 80px;
+  position: relative;
+}
+
+/* Add subtle separator between command sections */
+.vp-doc hr {
+  border: none;
+  border-top: 1px solid var(--vp-c-divider);
+  margin: 3rem 0;
+}
+
+/* ============================================
+ * Custom Container Blocks
+ * ============================================ */
+.custom-container {
+  margin: 20px 0;
+  padding: 16px 20px;
+  border-radius: 12px;
+  border-left: 4px solid;
+}
+
+.custom-container.info {
+  background: var(--vp-c-bg-soft);
+  border-color: var(--vp-c-primary);
+}
+
+.custom-container.success {
+  background: var(--vp-c-secondary-50);
+  border-color: var(--vp-c-secondary);
+}
+
+.dark .custom-container.success {
+  background: rgba(16, 185, 129, 0.1);
+}
+
+.custom-container.tip {
+  border-radius: 12px;
+}
+
+.custom-container.warning {
+  border-radius: 12px;
+}
+
+.custom-container.danger {
+  border-radius: 12px;
+}
+
+/* ============================================
+ * Code Block Improvements
+ * ============================================ */
+.vp-code-group {
+  margin: 20px 0;
+  border-radius: 12px;
+  overflow: hidden;
+}
+
+.vp-code-group .tabs {
+  background: var(--vp-c-bg-soft);
+  border-bottom: 1px solid var(--vp-c-divider);
+}
+
+.vp-code-group div[class*='language-'] {
+  margin: 0;
+  border-radius: 0;
+}
+
+div[class*='language-'] {
+  border-radius: 12px;
+  margin: 20px 0;
+}
+
+div[class*='language-'] pre {
+  line-height: 1.65;
+}
+
+/* Inline code */
+.vp-doc :not(pre) > code {
+  border-radius: 6px;
+  padding: 2px 6px;
+  font-size: 0.875em;
+  font-weight: 500;
+}
+
+/* ============================================
+ * Table Styling
+ * ============================================ */
+table {
+  border-collapse: collapse;
+  width: 100%;
+  margin: 20px 0;
+  border-radius: 12px;
+  overflow: hidden;
+}
+
+table th,
+table td {
+  padding: 12px 16px;
+  border: 1px solid var(--vp-c-divider);
+  text-align: left;
+}
+
+table th {
+  background: var(--vp-c-bg-soft);
+  font-weight: 600;
+  font-size: 0.875rem;
+  text-transform: uppercase;
+  letter-spacing: 0.03em;
+  color: var(--vp-c-text-2);
+}
+
+table tr:hover {
+  background: var(--vp-c-bg-soft);
+}
+
+/* ============================================
+ * Sidebar Polish
+ * ============================================ */
+.VPSidebar .group + .group {
+  margin-top: 0.5rem;
+  padding-top: 0.5rem;
+  border-top: 1px solid var(--vp-c-divider);
+}
+
+/* ============================================
+ * Scrollbar Styling
+ * ============================================ */
+::-webkit-scrollbar {
+  width: 6px;
+  height: 6px;
+}
+
+::-webkit-scrollbar-track {
+  background: transparent;
+}
+
+::-webkit-scrollbar-thumb {
+  background: var(--vp-c-surface-2);
+  border-radius: var(--vp-radius-full);
+}
+
+::-webkit-scrollbar-thumb:hover {
+  background: var(--vp-c-surface-3);
+}
+
+/* ============================================
+ * Link Improvements
+ * ============================================ */
+a {
+  text-decoration: none;
+  transition: color 0.2s ease;
+}
+
+a:hover {
+  text-decoration: underline;
+}
+
+/* ============================================
+ * Focus States — Accessibility
+ * ============================================ */
+:focus-visible {
+  outline: 2px solid var(--vp-c-primary);
+  outline-offset: 2px;
+}
+
+/* ============================================
+ * Skip Link — Accessibility
+ * ============================================ */
+.skip-link {
+  position: absolute;
+  top: -100px;
+  left: 0;
+  background: var(--vp-c-bg);
+  padding: 8px 16px;
+  z-index: 9999;
+  transition: top 0.3s;
+}
+
+.skip-link:focus {
+  top: 0;
+}
+
+/* ============================================
+ * Print Styles
+ * ============================================ */
+@media print {
+  .VPNav,
+  .VPSidebar,
+  .skip-link {
+    display: none;
+  }
+
+  .VPContent {
+    margin: 0 !important;
+    padding: 0 !important;
+  }
+}

docs/.vitepress/theme/styles/mobile.css

@@ -0,0 +1,346 @@
+/**
+ * Mobile-Responsive Styles
+ * Breakpoints: 320px-768px (mobile), 768px-1024px (tablet), 1024px+ (desktop)
+ * WCAG 2.1 AA compliant
+ */
+
+/* ============================================
+ * Mobile First Approach
+ * ============================================ */
+
+/* Base Mobile Styles (320px+) */
+@media (max-width: 768px) {
+  /* Typography */
+  :root {
+    --vp-font-size-base: 14px;
+    --vp-content-width: 100%;
+  }
+
+  /* Container */
+  .container {
+    padding: 0 16px;
+  }
+
+  /* Navigation */
+  .VPNav {
+    height: 56px;
+  }
+
+  .VPNavBar {
+    padding: 0 16px;
+  }
+
+  /* Sidebar */
+  .VPSidebar {
+    width: 100%;
+    max-width: 320px;
+  }
+
+  /* Content */
+  .VPContent {
+    padding: 16px;
+  }
+
+  /* Doc content adjustments */
+  .VPDoc .content-container {
+    padding: 0 16px;
+  }
+
+  /* Hide outline on mobile */
+  .VPDocOutline {
+    display: none;
+  }
+
+  /* Hero Section */
+  .VPHomeHero {
+    padding: 40px 16px;
+  }
+
+  .VPHomeHero h1 {
+    font-size: 28px;
+    line-height: 1.2;
+  }
+
+  .VPHomeHero p {
+    font-size: 14px;
+  }
+
+  /* Code Blocks */
+  div[class*='language-'] {
+    margin: 12px -16px;
+    border-radius: 0;
+  }
+
+  div[class*='language-'] pre {
+    padding: 12px 16px;
+    font-size: 12px;
+  }
+
+  /* Tables - make them scrollable */
+  .vp-doc table {
+    display: block;
+    width: 100%;
+    overflow-x: auto;
+    -webkit-overflow-scrolling: touch;
+  }
+
+  table {
+    font-size: 12px;
+  }
+
+  table th,
+  table td {
+    padding: 8px 12px;
+  }
+
+  /* Buttons */
+  .VPButton {
+    padding: 8px 16px;
+    font-size: 14px;
+  }
+
+  /* Cards */
+  .VPFeature {
+    padding: 16px;
+  }
+
+  /* Touch-friendly tap targets (min 44x44px per WCAG) */
+  button,
+  a,
+  input,
+  select,
+  textarea {
+    min-height: 44px;
+    min-width: 44px;
+  }
+
+  /* Search */
+  .DocSearch {
+    width: 100%;
+  }
+
+  /* Theme Switcher */
+  .theme-switcher {
+    padding: 12px;
+  }
+
+  /* Breadcrumbs */
+  .breadcrumb {
+    padding: 8px 0;
+    font-size: 12px;
+  }
+
+  /* Table of Contents - hidden on mobile */
+  .page-toc {
+    display: none;
+  }
+
+  /* Typography adjustments for mobile */
+  .vp-doc h1 {
+    font-size: 1.75rem;
+    margin-bottom: 1rem;
+  }
+
+  .vp-doc h2 {
+    font-size: 1.375rem;
+    margin-top: 2rem;
+    padding-top: 1.5rem;
+  }
+
+  .vp-doc h3 {
+    font-size: 1.125rem;
+    margin-top: 1.5rem;
+  }
+
+  .vp-doc p {
+    line-height: 1.7;
+    margin: 1rem 0;
+  }
+
+  .vp-doc ul,
+  .vp-doc ol {
+    margin: 1rem 0;
+    padding-left: 1.25rem;
+  }
+
+  .vp-doc li {
+    margin: 0.375rem 0;
+  }
+}
+
+/* ============================================
+ * Tablet Styles (768px - 1024px)
+ * ============================================ */
+@media (min-width: 768px) and (max-width: 1024px) {
+  :root {
+    --vp-content-width: 760px;
+    --vp-sidebar-width: 240px;
+  }
+
+  .VPContent {
+    padding: 24px;
+  }
+
+  .VPDoc .content-container {
+    padding: 0 24px;
+    max-width: var(--vp-content-width);
+  }
+
+  .VPHomeHero {
+    padding: 60px 24px;
+  }
+
+  .VPHomeHero h1 {
+    font-size: 36px;
+  }
+
+  div[class*='language-'] {
+    margin: 12px 0;
+  }
+
+  /* Outline visible but narrower */
+  .VPDocOutline {
+    width: 200px;
+    padding-left: 16px;
+  }
+
+  .VPDocOutline .outline-link {
+    font-size: 12px;
+  }
+}
+
+/* ============================================
+ * Desktop Styles (1024px+)
+ * ============================================ */
+@media (min-width: 1024px) {
+  :root {
+    --vp-layout-max-width: 1600px;
+    --vp-content-width: 960px;
+    --vp-sidebar-width: 272px;
+  }
+
+  .VPContent {
+    padding: 32px 48px;
+    max-width: var(--vp-layout-max-width);
+  }
+
+  .VPDoc .content-container {
+    max-width: var(--vp-content-width);
+    padding: 0 40px;
+  }
+
+  /* Outline - sticky on desktop with good width */
+  .VPDocOutline {
+    position: sticky;
+    top: calc(var(--vp-nav-height) + 24px);
+    width: 256px;
+    padding-left: 24px;
+    max-height: calc(100vh - var(--vp-nav-height) - 48px);
+    overflow-y: auto;
+  }
+
+  .VPDocOutline .outline-marker {
+    display: block;
+  }
+
+  .VPDocOutline .outline-link {
+    font-size: 13px;
+    line-height: 1.6;
+    padding: 4px 12px;
+    transition: color 0.2s ease;
+  }
+
+  .VPDocOutline .outline-link:hover {
+    color: var(--vp-c-primary);
+  }
+
+  /* Two-column layout for content + TOC */
+  .content-with-toc {
+    display: grid;
+    grid-template-columns: 1fr 280px;
+    gap: 32px;
+  }
+}
+
+/* ============================================
+ * Large Desktop (1440px+)
+ * ============================================ */
+@media (min-width: 1440px) {
+  :root {
+    --vp-content-width: 1040px;
+    --vp-sidebar-width: 280px;
+  }
+
+  .VPDoc .content-container {
+    padding: 0 48px;
+  }
+
+  .VPDocOutline {
+    width: 280px;
+  }
+}
+
+/* ============================================
+ * Landscape Orientation
+ * ============================================ */
+@media (max-height: 500px) and (orientation: landscape) {
+  .VPNav {
+    height: 48px;
+  }
+
+  .VPHomeHero {
+    padding: 20px 16px;
+  }
+}
+
+/* ============================================
+ * High DPI Displays
+ * ============================================ */
+@media (-webkit-min-device-pixel-ratio: 2),
+       (min-resolution: 192dpi) {
+  /* Optimize images for retina displays */
+  img {
+    image-rendering: -webkit-optimize-contrast;
+    image-rendering: crisp-edges;
+  }
+}
+
+/* ============================================
+ * Reduced Motion (Accessibility)
+ * ============================================ */
+@media (prefers-reduced-motion: reduce) {
+  *,
+  *::before,
+  *::after {
+    animation-duration: 0.01ms !important;
+    animation-iteration-count: 1 !important;
+    transition-duration: 0.01ms !important;
+  }
+}
+
+/* ============================================
+ * Dark Mode Specific
+ * ============================================ */
+@media (max-width: 768px) {
+  .dark {
+    --vp-c-bg: #0f172a;
+    --vp-c-text-1: #f1f5f9;
+  }
+}
+
+/* ============================================
+ * Print Styles for Mobile
+ * ============================================ */
+@media print and (max-width: 768px) {
+  .VPContent {
+    font-size: 10pt;
+  }
+
+  h1 {
+    font-size: 14pt;
+  }
+
+  h2 {
+    font-size: 12pt;
+  }
+}

docs/.vitepress/theme/styles/variables.css

@@ -0,0 +1,221 @@
+/**
+ * Design Tokens for CCW Documentation
+ * Based on ui-ux-pro-max design system
+ * 8 themes: 4 colors × 2 modes (light/dark)
+ */
+
+:root {
+  /* ============================================
+   * Color Scheme: Blue (Default)
+   * ============================================ */
+
+  /* Primary Colors */
+  --vp-c-primary-50: #eff6ff;
+  --vp-c-primary-100: #dbeafe;
+  --vp-c-primary-200: #bfdbfe;
+  --vp-c-primary-300: #93c5fd;
+  --vp-c-primary-400: #60a5fa;
+  --vp-c-primary-500: #3b82f6;
+  --vp-c-primary-600: #2563eb;
+  --vp-c-primary-700: #1d4ed8;
+  --vp-c-primary-800: #1e40af;
+  --vp-c-primary-900: #1e3a8a;
+  --vp-c-primary: var(--vp-c-primary-500);
+  --vp-c-brand-1: var(--vp-c-primary-500);
+  --vp-c-brand-2: var(--vp-c-secondary-500);
+  --vp-c-brand-soft: rgba(59, 130, 246, 0.1);
+
+  /* Secondary Colors (Green) */
+  --vp-c-secondary-50: #ecfdf5;
+  --vp-c-secondary-100: #d1fae5;
+  --vp-c-secondary-200: #a7f3d0;
+  --vp-c-secondary-300: #6ee7b7;
+  --vp-c-secondary-400: #34d399;
+  --vp-c-secondary-500: #10b981;
+  --vp-c-secondary-600: #059669;
+  --vp-c-secondary-700: #047857;
+  --vp-c-secondary-800: #065f46;
+  --vp-c-secondary-900: #064e3b;
+  --vp-c-secondary: var(--vp-c-secondary-500);
+
+  /* Accent Colors */
+  --vp-c-accent-50: #fef3c7;
+  --vp-c-accent-100: #fde68a;
+  --vp-c-accent-200: #fcd34d;
+  --vp-c-accent-300: #fbbf24;
+  --vp-c-accent-400: #f59e0b;
+  --vp-c-accent-500: #d97706;
+  --vp-c-accent: var(--vp-c-accent-400);
+
+  /* Background Colors (Light Mode) */
+  --vp-c-bg: #ffffff;
+  --vp-c-bg-soft: #f9fafb;
+  --vp-c-bg-mute: #f3f4f6;
+  --vp-c-bg-alt: #ffffff;
+
+  /* Surface Colors */
+  --vp-c-surface: #f9fafb;
+  --vp-c-surface-1: #f3f4f6;
+  --vp-c-surface-2: #e5e7eb;
+  --vp-c-surface-3: #d1d5db;
+
+  /* Border Colors */
+  --vp-c-border: #e5e7eb;
+  --vp-c-border-soft: #f3f4f6;
+  --vp-c-divider: #e5e7eb;
+
+  /* Text Colors */
+  --vp-c-text-1: #111827;
+  --vp-c-text-2: #374151;
+  --vp-c-text-3: #6b7280;
+  --vp-c-text-4: #9ca3af;
+  --vp-c-text-code: #ef4444;
+
+  /* Typography */
+  --vp-font-family-base: 'Inter', -apple-system, BlinkMacSystemFont, 'Segoe UI', Roboto, Oxygen, Ubuntu, Cantarell, 'Fira Sans', 'Droid Sans', 'Helvetica Neue', sans-serif;
+  --vp-font-family-mono: 'Fira Code', 'Cascadia Code', 'JetBrains Mono', Consolas, 'Courier New', monospace;
+
+  /* Font Sizes */
+  --vp-font-size-base: 16px;
+  --vp-font-size-sm: 14px;
+  --vp-font-size-lg: 18px;
+  --vp-font-size-xl: 20px;
+
+  /* Spacing */
+  --vp-spacing-xs: 0.25rem;   /* 4px */
+  --vp-spacing-sm: 0.5rem;    /* 8px */
+  --vp-spacing-md: 1rem;      /* 16px */
+  --vp-spacing-lg: 1.5rem;    /* 24px */
+  --vp-spacing-xl: 2rem;      /* 32px */
+  --vp-spacing-2xl: 3rem;     /* 48px */
+
+  /* Border Radius */
+  --vp-radius-sm: 0.25rem;    /* 4px */
+  --vp-radius-md: 0.375rem;   /* 6px */
+  --vp-radius-lg: 0.5rem;     /* 8px */
+  --vp-radius-xl: 0.75rem;    /* 12px */
+  --vp-radius-full: 9999px;
+
+  /* Shadows */
+  --vp-shadow-sm: 0 1px 2px 0 rgb(0 0 0 / 0.05);
+  --vp-shadow-md: 0 4px 6px -1px rgb(0 0 0 / 0.1);
+  --vp-shadow-lg: 0 10px 15px -3px rgb(0 0 0 / 0.1);
+  --vp-shadow-xl: 0 20px 25px -5px rgb(0 0 0 / 0.1);
+
+  /* Transitions */
+  --vp-transition-color: 0.2s ease;
+  --vp-transition-transform: 0.2s ease;
+  --vp-transition-all: 0.3s cubic-bezier(0.4, 0, 0.2, 1);
+
+  /* Z-Index */
+  --vp-z-index-base: 1;
+  --vp-z-index-dropdown: 10;
+  --vp-z-index-sticky: 20;
+  --vp-z-index-fixed: 50;
+  --vp-z-index-modal: 100;
+  --vp-z-index-toast: 200;
+}
+
+/* ============================================
+ * Dark Mode
+ * ============================================ */
+.dark {
+  /* Background Colors (Dark Mode) */
+  --vp-c-bg: #111827;
+  --vp-c-bg-soft: #1f2937;
+  --vp-c-bg-mute: #374151;
+  --vp-c-bg-alt: #0f172a;
+
+  /* Surface Colors */
+  --vp-c-surface: #1f2937;
+  --vp-c-surface-1: #374151;
+  --vp-c-surface-2: #4b5563;
+  --vp-c-surface-3: #6b7280;
+
+  /* Border Colors */
+  --vp-c-border: #374151;
+  --vp-c-border-soft: #1f2937;
+  --vp-c-divider: #374151;
+
+  /* Text Colors */
+  --vp-c-text-1: #f9fafb;
+  --vp-c-text-2: #e5e7eb;
+  --vp-c-text-3: #d1d5db;
+  --vp-c-text-4: #9ca3af;
+  --vp-c-text-code: #fca5a5;
+
+  /* Primary Colors (adjusted for dark mode) */
+  --vp-c-primary-50: #1e3a8a;
+  --vp-c-primary-100: #1e40af;
+  --vp-c-primary-200: #1d4ed8;
+  --vp-c-primary-300: #2563eb;
+  --vp-c-primary-400: #3b82f6;
+  --vp-c-primary-500: #60a5fa;
+  --vp-c-primary-600: #93c5fd;
+  --vp-c-primary-700: #bfdbfe;
+  --vp-c-primary-800: #dbeafe;
+  --vp-c-primary-900: #eff6ff;
+  --vp-c-primary: var(--vp-c-primary-400);
+  --vp-c-brand-1: var(--vp-c-primary-400);
+  --vp-c-brand-2: var(--vp-c-secondary-400);
+  --vp-c-brand-soft: rgba(96, 165, 250, 0.2);
+
+  /* Secondary Colors (adjusted for dark mode) */
+  --vp-c-secondary-50: #064e3b;
+  --vp-c-secondary-100: #065f46;
+  --vp-c-secondary-200: #047857;
+  --vp-c-secondary-300: #059669;
+  --vp-c-secondary-400: #10b981;
+  --vp-c-secondary-500: #34d399;
+  --vp-c-secondary-600: #6ee7b7;
+  --vp-c-secondary-700: #a7f3d0;
+  --vp-c-secondary-800: #d1fae5;
+  --vp-c-secondary-900: #ecfdf5;
+  --vp-c-secondary: var(--vp-c-secondary-400);
+}
+
+/* ============================================
+ * Color Scheme: Green
+ * ============================================ */
+[data-theme="green"] {
+  --vp-c-primary: var(--vp-c-secondary-500);
+}
+[data-theme="green"].dark {
+  --vp-c-primary: var(--vp-c-secondary-400);
+}
+
+/* ============================================
+ * Color Scheme: Orange
+ * ============================================ */
+[data-theme="orange"] {
+  --vp-c-primary: var(--vp-c-accent-500);
+}
+[data-theme="orange"].dark {
+  --vp-c-primary: var(--vp-c-accent-400);
+}
+
+/* ============================================
+ * Color Scheme: Purple
+ * ============================================ */
+[data-theme="purple"] {
+  --vp-c-primary-500: #8b5cf6;
+  --vp-c-primary-600: #7c3aed;
+  --vp-c-primary-700: #6d28d9;
+  --vp-c-primary: var(--vp-c-primary-500);
+}
+[data-theme="purple"].dark {
+  --vp-c-primary-400: #a78bfa;
+  --vp-c-primary-500: #8b5cf6;
+  --vp-c-primary: var(--vp-c-primary-400);
+}
+
+/* ============================================
+ * Utility Classes
+ * ============================================ */
+.text-primary { color: var(--vp-c-primary); }
+.text-secondary { color: var(--vp-c-secondary); }
+.bg-surface { background-color: var(--vp-c-surface); }
+.border-primary { border-color: var(--vp-c-primary); }
+.rounded-md { border-radius: var(--vp-radius-md); }
+.shadow-md { box-shadow: var(--vp-shadow-md); }
+.transition { transition: var(--vp-transition-all); }

docs/agents/builtin.md

@@ -0,0 +1,524 @@
+# Built-in Agents
+
+CCW includes **21 specialized agents** organized across 5 categories, each designed for specific development tasks. Agents can work independently or be orchestrated together for complex workflows.
+
+## Categories Overview
+
+| Category | Count | Primary Use |
+|----------|-------|-------------|
+| [CLI](#cli-agents) | 6 | CLI-based interactions, exploration, and planning |
+| [Development](#development-agents) | 5 | Code implementation and debugging |
+| [Planning](#planning-agents) | 4 | Strategic planning and issue management |
+| [Testing](#testing-agents) | 3 | Test generation, execution, and quality assurance |
+| [Documentation](#documentation-agents) | 3 | Documentation and design systems |
+
+---
+
+## CLI Agents
+
+### cli-explore-agent
+
+**Purpose**: Specialized CLI exploration with 3 analysis modes
+
+**Capabilities**:
+- Quick-scan (Bash only)
+- Deep-scan (Bash + Gemini)
+- Dependency-map (graph construction)
+- 4-phase workflow: Task Understanding → Analysis Execution → Schema Validation → Output Generation
+
+**Tools**: `Bash`, `Read`, `Grep`, `Glob`, `ccw cli (gemini/qwen/codex)`, `ACE search_context`
+
+```javascript
+Task({
+  subagent_type: "cli-explore-agent",
+  prompt: "Analyze authentication module dependencies"
+})
+```
+
+### cli-discuss-agent
+
+**Purpose**: Multi-CLI collaborative discussion with cross-verification
+
+**Capabilities**:
+- 5-phase workflow: Context Prep → CLI Execution → Cross-Verify → Synthesize → Output
+- Loads discussion history
+- Maintains context across sessions
+
+**Tools**: `Read`, `Grep`, `Glob`, `ccw cli`
+
+**Calls**: `cli-explore-agent` for codebase discovery before discussions
+
+```javascript
+Task({
+  subagent_type: "cli-discuss-agent",
+  prompt: "Discuss architecture patterns for microservices"
+})
+```
+
+### cli-execution-agent
+
+**Purpose**: Intelligent CLI execution with automated context discovery
+
+**Capabilities**:
+- 5-phase workflow: Task Understanding → Context Discovery → Prompt Enhancement → Tool Execution → Output Routing
+- Background execution support
+- Result polling
+
+**Tools**: `Bash`, `Read`, `Grep`, `Glob`, `ccw cli`, `TaskOutput`
+
+**Calls**: `cli-explore-agent` for discovery before execution
+
+```javascript
+Task({
+  subagent_type: "cli-execution-agent",
+  prompt: "Execute security scan on authentication module"
+})
+```
+
+### cli-lite-planning-agent
+
+**Purpose**: Lightweight planning for quick task breakdowns
+
+**Capabilities**:
+- Creates simplified task JSONs without complex schema validation
+- For straightforward implementation tasks
+
+**Tools**: `Read`, `Write`, `Bash`, `Grep`
+
+```javascript
+Task({
+  subagent_type: "cli-lite-planning-agent",
+  prompt: "Plan user registration feature"
+})
+```
+
+### cli-planning-agent
+
+**Purpose**: Full-featured planning for complex implementations
+
+**Capabilities**:
+- 6-field schema with context loading
+- Flow control and artifact integration
+- Comprehensive task JSON generation
+
+**Tools**: `Read`, `Write`, `Bash`, `Grep`, `Glob`, `mcp__ace-tool__search_context`
+
+```javascript
+Task({
+  subagent_type: "cli-planning-agent",
+  prompt: "Plan microservices architecture migration"
+})
+```
+
+### cli-roadmap-plan-agent
+
+**Purpose**: Strategic planning for roadmap and milestone generation
+
+**Capabilities**:
+- Creates long-term project plans
+- Generates epics, milestones, and delivery timelines
+- Issue creation via ccw
+
+**Tools**: `Read`, `Write`, `Bash`, `Grep`
+
+```javascript
+Task({
+  subagent_type: "cli-roadmap-plan-agent",
+  prompt: "Create Q1 roadmap for payment system"
+})
+```
+
+---
+
+## Development Agents
+
+### code-developer
+
+**Purpose**: Core code execution for any implementation task
+
+**Capabilities**:
+- Adapts to any domain while maintaining quality standards
+- Supports analysis, implementation, documentation, research
+- Complex multi-step workflows
+
+**Tools**: `Read`, `Edit`, `Write`, `Bash`, `Grep`, `Glob`, `Task`, `mcp__ccw-tools__edit_file`, `mcp__ccw-tools__write_file`
+
+```javascript
+Task({
+  subagent_type: "code-developer",
+  prompt: "Implement user authentication with JWT"
+})
+```
+
+### tdd-developer
+
+**Purpose**: TDD-aware code execution for Red-Green-Refactor workflows
+
+**Capabilities**:
+- Extends code-developer with TDD cycle awareness
+- Automatic test-fix iteration
+- CLI session resumption
+
+**Tools**: `Read`, `Edit`, `Write`, `Bash`, `Grep`, `Glob`, `ccw cli`
+
+**Extends**: `code-developer`
+
+```javascript
+Task({
+  subagent_type: "tdd-developer",
+  prompt: "Implement payment processing with TDD"
+})
+```
+
+### context-search-agent
+
+**Purpose**: Specialized context collector for brainstorming workflows
+
+**Capabilities**:
+- Analyzes existing codebase
+- Identifies patterns
+- Generates standardized context packages
+
+**Tools**: `mcp__ace-tool__search_context`, `mcp__ccw-tools__smart_search`, `Read`, `Grep`, `Glob`, `Bash`
+
+```javascript
+Task({
+  subagent_type: "context-search-agent",
+  prompt: "Gather context for API refactoring"
+})
+```
+
+### debug-explore-agent
+
+**Purpose**: Debugging specialist for code analysis and problem diagnosis
+
+**Capabilities**:
+- Hypothesis-driven debugging with NDJSON logging
+- CLI-assisted analysis
+- Iterative verification
+- Traces execution flow, identifies failure points, analyzes state at failure
+
+**Tools**: `Read`, `Grep`, `Bash`, `ccw cli`
+
+**Workflow**: Bug Analysis → Hypothesis Generation → Instrumentation → Log Analysis → Fix Verification
+
+```javascript
+Task({
+  subagent_type: "debug-explore-agent",
+  prompt: "Debug memory leak in connection handler"
+})
+```
+
+### universal-executor
+
+**Purpose**: Versatile execution for implementing any task efficiently
+
+**Capabilities**:
+- Adapts to any domain while maintaining quality standards
+- Handles analysis, implementation, documentation, research
+- Complex multi-step workflows
+
+**Tools**: `Read`, `Edit`, `Write`, `Bash`, `Grep`, `Glob`, `Task`, `mcp__ace-tool__search_context`, `mcp__exa__web_search_exa`
+
+```javascript
+Task({
+  subagent_type: "universal-executor",
+  prompt: "Implement GraphQL API with authentication"
+})
+```
+
+---
+
+## Planning Agents
+
+### action-planning-agent
+
+**Purpose**: Pure execution agent for creating implementation plans
+
+**Capabilities**:
+- Transforms requirements and brainstorming artifacts into structured plans
+- Quantified deliverables and measurable acceptance criteria
+- Control flags for execution modes
+
+**Tools**: `Read`, `Write`, `Bash`, `Grep`, `Glob`, `mcp__ace-tool__search_context`, `mcp__ccw-tools__smart_search`
+
+```javascript
+Task({
+  subagent_type: "action-planning-agent",
+  prompt: "Create implementation plan for user dashboard"
+})
+```
+
+### conceptual-planning-agent
+
+**Purpose**: High-level planning for architectural and conceptual design
+
+**Capabilities**:
+- Creates system designs
+- Architecture patterns
+- Technical strategies
+
+**Tools**: `Read`, `Write`, `Bash`, `Grep`, `ccw cli`
+
+```javascript
+Task({
+  subagent_type: "conceptual-planning-agent",
+  prompt: "Design event-driven architecture for order system"
+})
+```
+
+### issue-plan-agent
+
+**Purpose**: Issue resolution planning with closed-loop exploration
+
+**Capabilities**:
+- Analyzes issues and generates solution plans
+- Creates task JSONs with dependencies and acceptance criteria
+- 5-phase tasks from exploration to solution
+
+**Tools**: `Read`, `Write`, `Bash`, `Grep`, `mcp__ace-tool__search_context`
+
+```javascript
+Task({
+  subagent_type: "issue-plan-agent",
+  prompt: "Plan resolution for issue #123"
+})
+```
+
+### issue-queue-agent
+
+**Purpose**: Solution ordering agent for queue formation
+
+**Capabilities**:
+- Receives solutions from bound issues
+- Uses Gemini for intelligent conflict detection
+- Produces ordered execution queue
+
+**Tools**: `Read`, `Write`, `Bash`, `ccw cli (gemini)`, `mcp__ace-tool__search_context`, `mcp__ccw-tools__smart_search`
+
+**Calls**: `issue-plan-agent`
+
+```javascript
+Task({
+  subagent_type: "issue-queue-agent",
+  prompt: "Form execution queue for issues #101, #102, #103"
+})
+```
+
+---
+
+## Testing Agents
+
+### test-action-planning-agent
+
+**Purpose**: Specialized agent for test planning documents
+
+**Capabilities**:
+- Extends action-planning-agent for test planning
+- Progressive L0-L3 test layers (Static, Unit, Integration, E2E)
+- AI code issue detection (L0.5) with CRITICAL/ERROR/WARNING severity
+- Project-specific templates
+- Test anti-pattern detection with quality gates
+
+**Tools**: `Read`, `Write`, `Bash`, `Grep`, `Glob`
+
+**Extends**: `action-planning-agent`
+
+```javascript
+Task({
+  subagent_type: "test-action-planning-agent",
+  prompt: "Create test plan for payment module"
+})
+```
+
+### test-context-search-agent
+
+**Purpose**: Specialized context collector for test generation workflows
+
+**Capabilities**:
+- Analyzes test coverage
+- Identifies missing tests
+- Loads implementation context from source sessions
+- Generates standardized test-context packages
+
+**Tools**: `mcp__ccw-tools__codex_lens`, `Read`, `Glob`, `Bash`, `Grep`
+
+```javascript
+Task({
+  subagent_type: "test-context-search-agent",
+  prompt: "Gather test context for authentication module"
+})
+```
+
+### test-fix-agent
+
+**Purpose**: Execute tests, diagnose failures, and fix code until all tests pass
+
+**Capabilities**:
+- Multi-layered test execution (L0-L3)
+- Analyzes failures and modifies source code
+- Quality gate for passing tests
+
+**Tools**: `Bash`, `Read`, `Edit`, `Write`, `Grep`, `ccw cli`
+
+```javascript
+Task({
+  subagent_type: "test-fix-agent",
+  prompt: "Run tests for user service and fix failures"
+})
+```
+
+---
+
+## Documentation Agents
+
+### doc-generator
+
+**Purpose**: Documentation generation for technical docs, API references, and code comments
+
+**Capabilities**:
+- Synthesizes context from multiple sources
+- Produces comprehensive documentation
+- Flow_control-based task execution
+
+**Tools**: `Read`, `Write`, `Bash`, `Grep`, `Glob`
+
+```javascript
+Task({
+  subagent_type: "doc-generator",
+  prompt: "Generate API documentation for REST endpoints"
+})
+```
+
+### memory-bridge
+
+**Purpose**: Documentation update coordinator for complex projects
+
+**Capabilities**:
+- Orchestrates parallel CLAUDE.md updates
+- Uses ccw tool exec update_module_claude
+- Processes every module path
+
+**Tools**: `Bash`, `ccw tool exec`, `TodoWrite`
+
+```javascript
+Task({
+  subagent_type: "memory-bridge",
+  prompt: "Update CLAUDE.md for all modules"
+})
+```
+
+### ui-design-agent
+
+**Purpose**: UI design token management and prototype generation
+
+**Capabilities**:
+- W3C Design Tokens Format compliance
+- State-based component definitions (default, hover, focus, active, disabled)
+- Complete component library coverage (12+ interactive components)
+- Animation-component state integration
+- WCAG AA compliance validation
+- Token-driven prototype generation
+
+**Tools**: `Read`, `Write`, `Edit`, `Bash`, `mcp__exa__web_search_exa`, `mcp__exa__get_code_context_exa`
+
+```javascript
+Task({
+  subagent_type: "ui-design-agent",
+  prompt: "Generate design tokens for dashboard components"
+})
+```
+
+---
+
+## Orchestration Patterns
+
+Agents can be combined using these orchestration patterns:
+
+### Inheritance Chain
+
+Agent extends another agent's capabilities:
+
+| Parent | Child | Extension |
+|--------|-------|-----------|
+| code-developer | tdd-developer | Adds TDD Red-Green-Refactor workflow, test-fix cycle |
+| action-planning-agent | test-action-planning-agent | Adds L0-L3 test layers, AI issue detection |
+
+### Sequential Delegation
+
+Agent calls another agent for preprocessing:
+
+| Caller | Callee | Purpose |
+|--------|--------|---------|
+| cli-discuss-agent | cli-explore-agent | Codebase discovery before discussion |
+| cli-execution-agent | cli-explore-agent | Discovery before CLI command execution |
+
+### Queue Formation
+
+Agent collects outputs from multiple agents and orders them:
+
+| Collector | Source | Purpose |
+|-----------|--------|---------|
+| issue-queue-agent | issue-plan-agent | Collect solutions, detect conflicts, produce ordered queue |
+
+### Context Loading Chain
+
+Agent generates context packages used by execution agents:
+
+| Context Provider | Consumer | Purpose |
+|------------------|----------|---------|
+| context-search-agent | code-developer | Provides brainstorming context packages |
+| test-context-search-agent | test-fix-agent | Provides test context packages |
+
+### Quality Gate Chain
+
+Sequential execution through validation gates:
+
+```
+code-developer (IMPL-001)
+  → test-fix-agent (IMPL-001.3 validation)
+  → test-fix-agent (IMPL-001.5 review)
+  → test-fix-agent (IMPL-002 fix)
+```
+
+---
+
+## Agent Selection Guide
+
+| Task | Recommended Agent | Alternative |
+|------|------------------|-------------|
+| Explore codebase | cli-explore-agent | context-search-agent |
+| Implement code | code-developer | tdd-developer |
+| Debug issues | debug-explore-agent | cli-execution-agent |
+| Plan implementation | cli-planning-agent | action-planning-agent |
+| Generate tests | test-action-planning-agent | test-fix-agent |
+| Review code | test-fix-agent | doc-generator |
+| Create documentation | doc-generator | ui-design-agent |
+| UI design | ui-design-agent | - |
+| Manage issues | issue-plan-agent | issue-queue-agent |
+
+---
+
+## Tool Dependencies
+
+### Core Tools
+
+All agents have access to: `Read`, `Write`, `Edit`, `Bash`, `Grep`, `Glob`
+
+### MCP Tools
+
+Specialized agents use: `mcp__ace-tool__search_context`, `mcp__ccw-tools__smart_search`, `mcp__ccw-tools__edit_file`, `mcp__ccw-tools__write_file`, `mcp__ccw-tools__codex_lens`, `mcp__exa__web_search_exa`
+
+### CLI Tools
+
+CLI-capable agents use: `ccw cli`, `ccw tool exec`
+
+### Workflow Tools
+
+Coordinating agents use: `Task`, `TaskCreate`, `TaskUpdate`, `TaskList`, `TaskOutput`, `TodoWrite`, `SendMessage`
+
+::: info See Also
+- [Agents Overview](./index.md) - Agent system introduction
+- [Custom Agents](./custom.md) - Create custom agents
+- [Team Skills](../skills/core-skills.md#team-skills) - Multi-agent team skills
+:::

docs/agents/custom.md

@@ -0,0 +1,263 @@
+# Custom Agents
+
+Guide to creating and configuring custom CCW agents.
+
+## Agent Structure
+
+```
+~/.claude/agents/my-agent/
+├── AGENT.md          # Agent definition
+├── index.ts          # Agent logic
+├── tools/            # Agent-specific tools
+└── examples/         # Usage examples
+```
+
+## Creating an Agent
+
+### 1. Define Agent
+
+Create `AGENT.md`:
+
+```markdown
+---
+name: my-agent
+type: development
+version: 1.0.0
+capabilities: [react, typescript, testing]
+---
+
+# My Custom Agent
+
+Specialized agent for React component development with TypeScript.
+
+## Capabilities
+
+- Generate React components with hooks
+- TypeScript type definitions
+- Vitest testing setup
+- Tailwind CSS styling
+
+## Usage
+
+\`\`\`javascript
+Task({
+  subagent_type: "my-agent",
+  prompt: "Create a user profile component"
+})
+\`\`\`
+```
+
+### 2. Implement Agent Logic
+
+Create `index.ts`:
+
+```typescript
+import type { AgentContext, AgentResult } from '@ccw/types'
+
+export async function execute(
+  prompt: string,
+  context: AgentContext
+): Promise<AgentResult> {
+  // Analyze request
+  const intent = analyzeIntent(prompt)
+
+  // Execute based on intent
+  switch (intent.type) {
+    case 'generate-component':
+      return await generateComponent(intent.options)
+    case 'add-tests':
+      return await addTests(intent.options)
+    default:
+      return await handleGeneral(prompt)
+  }
+}
+
+function analyzeIntent(prompt: string) {
+  // Parse user intent from prompt
+  // Return structured intent object
+}
+```
+
+## Agent Capabilities
+
+### Code Generation
+
+```typescript
+async function generateComponent(options: ComponentOptions) {
+  return {
+    files: [
+      {
+        path: 'src/components/UserProfile.tsx',
+        content: generateReactComponent(options)
+      },
+      {
+        path: 'src/components/UserProfile.test.tsx',
+        content: generateTests(options)
+      }
+    ]
+  }
+}
+```
+
+### Analysis
+
+```typescript
+async function analyzeCodebase(context: AgentContext) {
+  const files = await context.filesystem.read('src/**/*.ts')
+  const patterns = identifyPatterns(files)
+  return {
+    patterns,
+    recommendations: generateRecommendations(patterns)
+  }
+}
+```
+
+### Testing
+
+```typescript
+async function generateTests(options: TestOptions) {
+  return {
+    framework: 'vitest',
+    files: [
+      {
+        path: `${options.file}.test.ts`,
+        content: generateTestCode(options)
+      }
+    ]
+  }
+}
+```
+
+## Agent Tools
+
+Agents can define custom tools:
+
+```typescript
+export const tools = {
+  'my-tool': {
+    description: 'My custom tool',
+    parameters: {
+      type: 'object',
+      properties: {
+        input: { type: 'string' }
+      }
+    },
+    execute: async (params) => {
+      // Tool implementation
+    }
+  }
+}
+```
+
+## Agent Communication
+
+Agents communicate via message bus:
+
+```typescript
+// Send message to another agent
+await context.messaging.send({
+  to: 'tester',
+  type: 'task-complete',
+  data: { files: generatedFiles }
+})
+
+// Receive messages
+context.messaging.on('task-complete', async (message) => {
+  if (message.from === 'executor') {
+    await startTesting(message.data.files)
+  }
+})
+```
+
+## Agent Configuration
+
+Configure agents in `~/.claude/agents/config.json`:
+
+```json
+{
+  "my-agent": {
+    "enabled": true,
+    "priority": 10,
+    "capabilities": {
+      "frameworks": ["react", "vue"],
+      "languages": ["typescript", "javascript"],
+      "tools": ["vitest", "playwright"]
+    },
+    "limits": {
+      "maxFiles": 100,
+      "maxSize": "10MB"
+    }
+  }
+}
+```
+
+## Agent Best Practices
+
+### 1. Clear Purpose
+
+Define specific, focused capabilities:
+
+```markdown
+# Good: Focused
+name: react-component-agent
+purpose: Generate React components with TypeScript
+
+# Bad: Too broad
+name: fullstack-agent
+purpose: Handle everything
+```
+
+### 2. Tool Selection
+
+Use appropriate tools for tasks:
+
+```typescript
+// File operations
+context.filesystem.read(path)
+context.filesystem.write(path, content)
+
+// Code analysis
+context.codebase.search(query)
+context.codebase.analyze(pattern)
+
+// Communication
+context.messaging.send(to, type, data)
+```
+
+### 3. Error Handling
+
+```typescript
+try {
+  const result = await executeTask(prompt)
+  return { success: true, result }
+} catch (error) {
+  return {
+    success: false,
+    error: error.message,
+    recovery: suggestRecovery(error)
+  }
+}
+```
+
+## Testing Agents
+
+```typescript
+import { describe, it, expect } from 'vitest'
+import { execute } from '../index'
+
+describe('my-agent', () => {
+  it('should generate component', async () => {
+    const result = await execute(
+      'Create a UserCard component',
+      mockContext
+    )
+    expect(result.success).toBe(true)
+    expect(result.files).toHaveLength(2) // component + test
+  })
+})
+```
+
+::: info See Also
+- [Built-in Agents](./builtin.md) - Pre-configured agents
+- [Agents Overview](./index.md) - Agent system introduction
+:::

docs/agents/index.md

@@ -0,0 +1,315 @@
+# Agents
+
+CCW provides specialized agents for different development workflows.
+
+## What are Agents?
+
+Agents are specialized AI assistants with specific expertise and tools for different aspects of software development. They are invoked via the `Task` tool in skills and workflows.
+
+## Built-in Agents
+
+### Execution Agents
+
+#### code-developer
+
+Pure code execution agent for implementing programming tasks and writing tests.
+
+**Expertise:**
+- Feature implementation
+- Code generation and modification
+- Test writing
+- Bug fixes
+- All programming languages and frameworks
+
+```javascript
+Task({
+  subagent_type: "code-developer",
+  prompt: "Implement user authentication API",
+  run_in_background: false
+})
+```
+
+#### tdd-developer
+
+TDD-aware execution agent supporting Red-Green-Refactor workflow.
+
+**Expertise:**
+- Test-first development
+- Red-Green-Refactor cycle
+- Test-driven implementation
+- Refactoring with test safety
+
+```javascript
+Task({
+  subagent_type: "tdd-developer",
+  prompt: "Execute TDD task IMPL-1 with test-first development",
+  run_in_background: false
+})
+```
+
+#### test-fix-agent
+
+Executes tests, diagnoses failures, and fixes code until tests pass.
+
+**Expertise:**
+- Test execution and analysis
+- Failure diagnosis
+- Automated fixing
+- Iterative test-fix cycles
+
+```javascript
+Task({
+  subagent_type: "test-fix-agent",
+  prompt: "Run tests and fix any failures",
+  run_in_background: false
+})
+```
+
+#### universal-executor
+
+Universal executor for general-purpose execution tasks.
+
+**Expertise:**
+- General task execution
+- Document generation
+- Multi-step workflows
+- Cross-domain tasks
+
+```javascript
+Task({
+  subagent_type: "universal-executor",
+  prompt: "Generate documentation for the API",
+  run_in_background: false
+})
+```
+
+### Analysis Agents
+
+#### context-search-agent
+
+Intelligent context collector for development tasks.
+
+**Expertise:**
+- Codebase exploration
+- Pattern discovery
+- Context gathering
+- File relationship analysis
+
+```javascript
+Task({
+  subagent_type: "context-search-agent",
+  prompt: "Gather context for implementing user authentication",
+  run_in_background: false
+})
+```
+
+#### debug-explore-agent
+
+Hypothesis-driven debugging agent with NDJSON logging.
+
+**Expertise:**
+- Root cause analysis
+- Hypothesis generation and testing
+- Debug logging
+- Systematic troubleshooting
+
+```javascript
+Task({
+  subagent_type: "debug-explore-agent",
+  prompt: "Debug the WebSocket connection timeout issue",
+  run_in_background: false
+})
+```
+
+#### cli-explore-agent
+
+CLI-based code exploration agent.
+
+**Expertise:**
+- CLI code analysis
+- External tool integration
+- Shell-based exploration
+- Command-line workflows
+
+```javascript
+Task({
+  subagent_type: "cli-explore-agent",
+  prompt: "Explore codebase for authentication patterns",
+  run_in_background: false
+})
+```
+
+### Planning Agents
+
+#### action-planning-agent
+
+Creates implementation plans based on requirements and control flags.
+
+**Expertise:**
+- Task breakdown
+- Implementation planning
+- Dependency analysis
+- Priority sequencing
+
+```javascript
+Task({
+  subagent_type: "action-planning-agent",
+  prompt: "Create implementation plan for OAuth2 authentication",
+  run_in_background: false
+})
+```
+
+#### issue-plan-agent
+
+Planning agent specialized for issue resolution.
+
+**Expertise:**
+- Issue analysis
+- Solution planning
+- Task generation
+- Impact assessment
+
+```javascript
+Task({
+  subagent_type: "issue-plan-agent",
+  prompt: "Plan solution for GitHub issue #123",
+  run_in_background: false
+})
+```
+
+### Specialized Agents
+
+#### team-worker
+
+Unified team worker agent for role-based collaboration.
+
+**Expertise:**
+- Multi-role execution (analyst, writer, planner, executor, tester, reviewer)
+- Team coordination
+- Lifecycle management
+- Inter-role communication
+
+```javascript
+Task({
+  subagent_type: "team-worker",
+  description: "Spawn executor worker",
+  team_name: "my-team",
+  name: "executor",
+  run_in_background: true,
+  prompt: "## Role Assignment\nrole: executor\n..."
+})
+```
+
+#### doc-generator
+
+Documentation generation agent.
+
+**Expertise:**
+- API documentation
+- User guides
+- Technical writing
+- Diagram generation
+
+```javascript
+Task({
+  subagent_type: "doc-generator",
+  prompt: "Generate documentation for the REST API",
+  run_in_background: false
+})
+```
+
+## Agent Categories
+
+| Category | Agents | Purpose |
+|----------|--------|---------|
+| **Execution** | code-developer, tdd-developer, test-fix-agent, universal-executor | Implement code and run tasks |
+| **Analysis** | context-search-agent, debug-explore-agent, cli-explore-agent | Explore and analyze code |
+| **Planning** | action-planning-agent, issue-plan-agent, cli-planning-agent | Create plans and strategies |
+| **Specialized** | team-worker, doc-generator, ui-design-agent | Domain-specific tasks |
+
+## Agent Communication
+
+Agents can communicate and coordinate with each other:
+
+```javascript
+// Agent sends message
+SendMessage({
+  type: "message",
+  recipient: "tester",
+  content: "Feature implementation complete, ready for testing"
+})
+
+// Agent receives message via system
+```
+
+## Team Workflows
+
+Multiple agents can work together on complex tasks:
+
+```
+[analyst] -> RESEARCH (requirements analysis)
+    |
+    v
+[writer] -> DRAFT (specification creation)
+    |
+    v
+[planner] -> PLAN (implementation planning)
+    |
+    +--[executor] -> IMPL (code implementation)
+    |               |
+    |               v
+    +-----------[tester] -> TEST (testing)
+    |
+    v
+[reviewer] -> REVIEW (code review)
+```
+
+## Using Agents
+
+### Via Task Tool
+
+```javascript
+// Foreground execution
+Task({
+  subagent_type: "code-developer",
+  prompt: "Implement user dashboard",
+  run_in_background: false
+})
+
+// Background execution
+Task({
+  subagent_type: "code-developer",
+  prompt: "Implement user dashboard",
+  run_in_background: true
+})
+```
+
+### Via Team Skills
+
+Team skills automatically coordinate multiple agents:
+
+```javascript
+Skill({
+  skill: "team-lifecycle",
+  args: "Build user authentication system"
+})
+```
+
+### Configuration
+
+Agent behavior is configured via role-spec files in team workflows:
+
+```markdown
+---
+role: executor
+prefix: IMPL
+inner_loop: true
+subagents: [explore]
+---
+```
+
+::: info See Also
+- [Skills](../skills/) - Reusable skill library
+- [Workflows](../workflows/) - Orchestration system
+- [Teams](../workflows/teams.md) - Team workflow reference
+:::

docs/cli/commands.md

@@ -0,0 +1,889 @@
+# CLI Commands Reference
+
+Complete reference for all **43 CCW commands** organized by category, with **7 workflow chains** for common development scenarios.
+
+## Command Categories
+
+| Category | Commands | Description |
+|----------|----------|-------------|
+| [Orchestrators](#orchestrators) | 3 | Main workflow orchestration |
+| [Workflow](#workflow-commands) | 10 | Project initialization and management |
+| [Session](#session-commands) | 6 | Session lifecycle management |
+| [Analysis](#analysis-commands) | 4 | Code analysis and debugging |
+| [Planning](#planning-commands) | 3 | Brainstorming and planning |
+| [Execution](#execution-commands) | 1 | Universal execution engine |
+| [UI Design](#ui-design-commands) | 10 | Design token extraction and prototyping |
+| [Issue](#issue-commands) | 8 | Issue discovery and resolution |
+| [Memory](#memory-commands) | 2 | Memory and context management |
+| [CLI](#cli-commands) | 2 | CLI configuration and review |
+
+---
+
+## Orchestrators
+
+### ccw
+
+**Purpose**: Main workflow orchestrator - analyze intent, select workflow, execute command chain
+
+**Description**: Analyzes user intent, selects appropriate workflow, and executes command chain in main process.
+
+**Flags**:
+- `-y, --yes` - Skip all confirmations
+
+**Mapped Skills**:
+- workflow-lite-plan, workflow-plan, workflow-execute, workflow-tdd
+- workflow-test-fix, workflow-multi-cli-plan, review-cycle, brainstorm
+- team-planex, team-iterdev, team-lifecycle, team-issue
+- team-testing, team-quality-assurance, team-brainstorm, team-uidesign
+
+```bash
+ccw -y
+```
+
+### ccw-coordinator
+
+**Purpose**: Command orchestration tool with external CLI execution
+
+**Description**: Analyzes requirements, recommends chain, executes sequentially with state persistence. Uses background tasks with hook callbacks.
+
+**Tools**: `Task`, `AskUserQuestion`, `Read`, `Write`, `Bash`, `Glob`, `Grep`
+
+```bash
+ccw-coordinator
+```
+
+### flow-create
+
+**Purpose**: Generate workflow templates for meta-skill/flow-coordinator
+
+```bash
+flow-create
+```
+
+---
+
+## Workflow Commands
+
+### workflow init
+
+**Purpose**: Initialize project-level state with intelligent project analysis
+
+**Description**: Uses cli-explore-agent for intelligent project analysis, generating project-tech.json and specification files.
+
+**Flags**:
+- `--regenerate` - Force regeneration
+- `--skip-specs` - Skip specification generation
+
+**Output**:
+- `.workflow/project-tech.json`
+- `.workflow/specs/*.md`
+
+**Delegates to**: `cli-explore-agent`
+
+```bash
+workflow init --regenerate
+```
+
+### workflow init-specs
+
+**Purpose**: Interactive wizard for creating individual specs or personal constraints
+
+**Flags**:
+- `--scope <global|project>` - Scope selection
+- `--dimension <specs|personal>` - Dimension selection
+- `--category <general|exploration|planning|execution>` - Category selection
+
+```bash
+workflow init-specs --scope project --dimension specs
+```
+
+### workflow init-guidelines
+
+**Purpose**: Interactive wizard to fill specs/*.md based on project analysis
+
+**Flags**:
+- `--reset` - Reset existing guidelines
+
+```bash
+workflow init-guidelines --reset
+```
+
+### workflow clean
+
+**Purpose**: Intelligent code cleanup with mainline detection
+
+**Description**: Discovers stale artifacts and executes safe cleanup operations.
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `--dry-run` - Preview without changes
+
+**Delegates to**: `cli-explore-agent`
+
+```bash
+workflow clean --dry-run
+```
+
+### workflow unified-execute-with-file
+
+**Purpose**: Universal execution engine for any planning/brainstorm/analysis output
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `-p, --plan <path>` - Plan file path
+- `--auto-commit` - Auto-commit after execution
+
+**Execution Methods**: Agent, CLI-Codex, CLI-Gemini, Auto
+
+**Output**: `.workflow/.execution/{session-id}/execution-events.md`
+
+```bash
+workflow unified-execute-with-file -p plan.json --auto-commit
+```
+
+### workflow brainstorm-with-file
+
+**Purpose**: Interactive brainstorming with multi-CLI collaboration
+
+**Description**: Documents thought evolution with idea expansion.
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `-c, --continue` - Continue previous session
+- `-m, --mode <creative|structured>` - Brainstorming mode
+
+**Delegates to**: `cli-explore-agent`, `Multi-CLI (Gemini/Codex/Claude)`
+
+**Output**: `.workflow/.brainstorm/{session-id}/synthesis.json`
+
+```bash
+workflow brainstorm-with-file -m creative
+```
+
+### workflow analyze-with-file
+
+**Purpose**: Interactive collaborative analysis with documented discussions
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `-c, --continue` - Continue previous session
+
+**Delegates to**: `cli-explore-agent`, `Gemini/Codex`
+
+**Output**: `.workflow/.analysis/{session-id}/discussion.md`
+
+```bash
+workflow analyze-with-file
+```
+
+### workflow debug-with-file
+
+**Purpose**: Interactive hypothesis-driven debugging
+
+**Description**: Documents exploration with Gemini-assisted correction.
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+
+**Output**: `.workflow/.debug/{session-id}/understanding.md`
+
+```bash
+workflow debug-with-file
+```
+
+### workflow collaborative-plan-with-file
+
+**Purpose**: Collaborative planning with Plan Note
+
+**Description**: Parallel agents fill pre-allocated sections with conflict detection.
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `--max-agents=5` - Maximum parallel agents
+
+**Output**: `.workflow/.planning/{session-id}/plan-note.md`
+
+```bash
+workflow collaborative-plan-with-file --max-agents=5
+```
+
+### workflow roadmap-with-file
+
+**Purpose**: Strategic requirement roadmap with iterative decomposition
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `-c, --continue` - Continue previous session
+- `-m, --mode <progressive|direct|auto>` - Decomposition mode
+
+**Output**:
+- `.workflow/.roadmap/{session-id}/roadmap.md`
+- `.workflow/issues/issues.jsonl`
+
+**Handoff to**: `team-planex`
+
+```bash
+workflow roadmap-with-file -m progressive
+```
+
+---
+
+## Session Commands
+
+### workflow session start
+
+**Purpose**: Discover existing sessions or start new workflow session
+
+**Flags**:
+- `--type <workflow|review|tdd|test|docs>` - Session type
+- `--auto|--new` - Auto-discover or force new
+
+**Calls first**: `workflow init`
+
+```bash
+workflow session start --type tdd
+```
+
+### workflow session resume
+
+**Purpose**: Resume the most recently paused workflow session
+
+```bash
+workflow session resume
+```
+
+### workflow session list
+
+**Purpose**: List all workflow sessions with status filtering
+
+```bash
+workflow session list
+```
+
+### workflow session sync
+
+**Purpose**: Quick-sync session work to specs/*.md and project-tech
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+
+**Updates**: `.workflow/specs/*.md`, `.workflow/project-tech.json`
+
+```bash
+workflow session sync -y
+```
+
+### workflow session solidify
+
+**Purpose**: Crystallize session learnings into permanent project guidelines
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `--type <convention|constraint|learning|compress>` - Solidification type
+- `--category <category>` - Category for guidelines
+- `--limit <N>` - Limit for compress mode
+
+```bash
+workflow session solidify --type learning
+```
+
+### workflow session complete
+
+**Purpose**: Mark active workflow session as complete
+
+**Description**: Archives with lessons learned, auto-calls sync.
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `--detailed` - Detailed completion report
+
+**Auto-calls**: `workflow session sync -y`
+
+```bash
+workflow session complete --detailed
+```
+
+---
+
+## Analysis Commands
+
+### workflow integration-test-cycle
+
+**Purpose**: Self-iterating integration test workflow
+
+**Description**: Autonomous test-fix cycles with reflection-driven adjustment.
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `-c, --continue` - Continue previous session
+- `--max-iterations=N` - Maximum iterations
+
+**Output**: `.workflow/.integration-test/{session-id}/reflection-log.md`
+
+```bash
+workflow integration-test-cycle --max-iterations=5
+```
+
+### workflow refactor-cycle
+
+**Purpose**: Tech debt discovery and self-iterating refactoring
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `-c, --continue` - Continue previous session
+- `--scope=module|project` - Refactoring scope
+
+**Output**: `.workflow/.refactor/{session-id}/reflection-log.md`
+
+```bash
+workflow refactor-cycle --scope project
+```
+
+---
+
+## Planning Commands
+
+### workflow req-plan-with-file
+
+**Purpose**: Requirement-level progressive roadmap planning with issue creation
+
+**Description**: Decomposes requirements into convergent layers or task sequences.
+
+```bash
+workflow req-plan-with-file
+```
+
+---
+
+## Execution Commands
+
+### workflow execute
+
+**Purpose**: Coordinate agent execution for workflow tasks
+
+**Description**: Automatic session discovery, parallel task processing, and status tracking.
+
+**Triggers**: `workflow:execute`
+
+```bash
+workflow execute
+```
+
+---
+
+## UI Design Commands
+
+### workflow ui-design style-extract
+
+**Purpose**: Extract design style from reference images or text prompts
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `--design-id <id>` - Design identifier
+- `--session <id>` - Session identifier
+- `--images <glob>` - Image file pattern
+- `--prompt <desc>` - Text description
+- `--variants <count>` - Number of variants
+- `--interactive` - Interactive mode
+- `--refine` - Refinement mode
+
+**Modes**: Exploration, Refinement
+
+**Output**: `style-extraction/style-{id}/design-tokens.json`
+
+```bash
+workflow ui-design style-extract --images "design/*.png" --variants 3
+```
+
+### workflow ui-design layout-extract
+
+**Purpose**: Extract structural layout from reference images or text prompts
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `--design-id <id>` - Design identifier
+- `--session <id>` - Session identifier
+- `--images <glob>` - Image file pattern
+- `--prompt <desc>` - Text description
+- `--targets <list>` - Target components
+- `--variants <count>` - Number of variants
+- `--device-type <desktop|mobile|tablet|responsive>` - Device type
+- `--interactive` - Interactive mode
+- `--refine` - Refinement mode
+
+**Delegates to**: `ui-design-agent`
+
+**Output**: `layout-extraction/layout-*.json`
+
+```bash
+workflow ui-design layout-extract --prompt "dashboard layout" --device-type responsive
+```
+
+### workflow ui-design generate
+
+**Purpose**: Assemble UI prototypes by combining layout templates with design tokens
+
+**Flags**:
+- `--design-id <id>` - Design identifier
+- `--session <id>` - Session identifier
+
+**Delegates to**: `ui-design-agent`
+
+**Prerequisites**: `workflow ui-design style-extract`, `workflow ui-design layout-extract`
+
+```bash
+workflow ui-design generate --design-id dashboard-001
+```
+
+### workflow ui-design animation-extract
+
+**Purpose**: Extract animation and transition patterns
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `--design-id <id>` - Design identifier
+- `--session <id>` - Session identifier
+- `--images <glob>` - Image file pattern
+- `--focus <types>` - Animation types
+- `--interactive` - Interactive mode
+- `--refine` - Refinement mode
+
+**Delegates to**: `ui-design-agent`
+
+**Output**: `animation-extraction/animation-tokens.json`
+
+```bash
+workflow ui-design animation-extract --focus "transition,keyframe"
+```
+
+### workflow ui-design import-from-code
+
+**Purpose**: Import design system from code files
+
+**Description**: Automatic file discovery for CSS/JS/HTML/SCSS.
+
+**Flags**:
+- `--design-id <id>` - Design identifier
+- `--session <id>` - Session identifier
+- `--source <path>` - Source path
+
+**Delegates to**: Style Agent, Animation Agent, Layout Agent
+
+**Output**: `style-extraction`, `animation-extraction`, `layout-extraction`
+
+```bash
+workflow ui-design import-from-code --source src/styles
+```
+
+### workflow ui-design codify-style
+
+**Purpose**: Extract styles from code and generate shareable reference package
+
+**Flags**:
+- `--package-name <name>` - Package name
+- `--output-dir <path>` - Output directory
+- `--overwrite` - Overwrite existing
+
+**Orchestrates**: `workflow ui-design import-from-code`, `workflow ui-design reference-page-generator`
+
+**Output**: `.workflow/reference_style/{package-name}/`
+
+```bash
+workflow ui-design codify-style --package-name my-design-system
+```
+
+### workflow ui-design reference-page-generator
+
+**Purpose**: Generate multi-component reference pages from design run extraction
+
+**Flags**:
+- `--design-run <path>` - Design run path
+- `--package-name <name>` - Package name
+- `--output-dir <path>` - Output directory
+
+**Output**: `.workflow/reference_style/{package-name}/preview.html`
+
+```bash
+workflow ui-design reference-page-generator --design-run .workflow/design-run-001
+```
+
+### workflow ui-design design-sync
+
+**Purpose**: Synchronize finalized design system references to brainstorming artifacts
+
+**Flags**:
+- `--session <session_id>` - Session identifier
+- `--selected-prototypes <list>` - Selected prototypes
+
+**Updates**: Role analysis documents, context-package.json
+
+```bash
+workflow ui-design design-sync --session design-001
+```
+
+### workflow ui-design explore-auto
+
+**Purpose**: Interactive exploratory UI design with style-centric batch generation
+
+**Flags**:
+- `--input <value>` - Input source
+- `--targets <list>` - Target components
+- `--target-type <page|component>` - Target type
+- `--session <id>` - Session identifier
+- `--style-variants <count>` - Style variants
+- `--layout-variants <count>` - Layout variants
+
+**Orchestrates**: `import-from-code`, `style-extract`, `animation-extract`, `layout-extract`, `generate`
+
+```bash
+workflow ui-design explore-auto --input "dashboard" --style-variants 3
+```
+
+### workflow ui-design imitate-auto
+
+**Purpose**: UI design workflow with direct code/image input
+
+**Flags**:
+- `--input <value>` - Input source
+- `--session <id>` - Session identifier
+
+**Orchestrates**: Same as explore-auto
+
+```bash
+workflow ui-design imitate-auto --input ./reference.png
+```
+
+---
+
+## Issue Commands
+
+### issue new
+
+**Purpose**: Create structured issue from GitHub URL or text description
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `--priority 1-5` - Issue priority
+
+**Features**: Clarity detection
+
+**Output**: `.workflow/issues/issues.jsonl`
+
+```bash
+issue new --priority 3
+```
+
+### issue discover
+
+**Purpose**: Discover potential issues from multiple perspectives
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `--perspectives=bug,ux,...` - Analysis perspectives
+- `--external` - Include external research
+
+**Perspectives**: bug, ux, test, quality, security, performance, maintainability, best-practices
+
+**Delegates to**: `cli-explore-agent`
+
+**Output**: `.workflow/issues/discoveries/{discovery-id}/`
+
+```bash
+issue discover --perspectives=bug,security,performance
+```
+
+### issue discover-by-prompt
+
+**Purpose**: Discover issues from user prompt with Gemini-planned exploration
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `--scope=src/**` - File scope
+- `--depth=standard|deep` - Analysis depth
+- `--max-iterations=5` - Maximum iterations
+
+**Delegates to**: `Gemini CLI`, `ACE search`, `multi-agent exploration`
+
+```bash
+issue discover-by-prompt --depth deep --scope "src/auth/**"
+```
+
+### issue plan
+
+**Purpose**: Batch plan issue resolution using issue-plan-agent
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `--all-pending` - Plan all pending issues
+- `--batch-size 3` - Batch size
+
+**Delegates to**: `issue-plan-agent`
+
+**Output**: `.workflow/issues/solutions/{issue-id}.jsonl`
+
+```bash
+issue plan --all-pending --batch-size 5
+```
+
+### issue queue
+
+**Purpose**: Form execution queue from bound solutions
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `--queues <n>` - Number of queues
+- `--issue <id>` - Specific issue
+
+**Delegates to**: `issue-queue-agent`
+
+**Output**: `.workflow/issues/queues/QUE-xxx.json`
+
+```bash
+issue queue --queues 2
+```
+
+### issue execute
+
+**Purpose**: Execute queue with DAG-based parallel orchestration
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `--queue <queue-id>` - Queue identifier
+- `--worktree [<path>]` - Use worktree isolation
+
+**Executors**: Codex (recommended), Gemini, Agent
+
+```bash
+issue execute --queue QUE-001 --worktree
+```
+
+### issue convert-to-plan
+
+**Purpose**: Convert planning artifacts to issue solutions
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `--issue <id>` - Issue identifier
+- `--supplement` - Supplement existing solution
+
+**Sources**: lite-plan, workflow-session, markdown, json
+
+```bash
+issue convert-to-plan --issue 123 --supplement
+```
+
+### issue from-brainstorm
+
+**Purpose**: Convert brainstorm session ideas into issue with executable solution
+
+**Flags**:
+- `-y, --yes` - Skip confirmation
+- `--idea=<index>` - Idea index
+- `--auto` - Auto-select best idea
+
+**Input Sources**: `synthesis.json`, `perspectives.json`, `.brainstorming/**`
+
+**Output**: `issues.jsonl`, `solutions/{issue-id}.jsonl`
+
+```bash
+issue from-brainstorm --auto
+```
+
+---
+
+## Memory Commands
+
+### memory prepare
+
+**Purpose**: Delegate to universal-executor agent for project analysis
+
+**Description**: Returns JSON core content package for memory loading.
+
+**Flags**:
+- `--tool gemini|qwen` - AI tool selection
+
+**Delegates to**: `universal-executor agent`
+
+```bash
+memory prepare --tool gemini
+```
+
+### memory style-skill-memory
+
+**Purpose**: Generate SKILL memory package from style reference
+
+**Flags**:
+- `--regenerate` - Force regeneration
+
+**Input**: `.workflow/reference_style/{package-name}/`
+
+**Output**: `.claude/skills/style-{package-name}/SKILL.md`
+
+```bash
+memory style-skill-memory --regenerate
+```
+
+---
+
+## CLI Commands
+
+### cli init
+
+**Purpose**: Generate .gemini/ and .qwen/ config directories
+
+**Description**: Creates settings.json and ignore files based on workspace technology detection.
+
+**Flags**:
+- `--tool gemini|qwen|all` - Tool selection
+- `--output path` - Output path
+- `--preview` - Preview without writing
+
+**Output**: `.gemini/`, `.qwen/`, `.geminiignore`, `.qwenignore`
+
+```bash
+cli init --tool all --preview
+```
+
+### cli codex-review
+
+**Purpose**: Interactive code review using Codex CLI
+
+**Flags**:
+- `--uncommitted` - Review uncommitted changes
+- `--base <branch>` - Compare to branch
+- `--commit <sha>` - Review specific commit
+- `--model <model>` - Model selection
+- `--title <title>` - Review title
+
+```bash
+cli codex-review --base main --title "Security Review"
+```
+
+---
+
+## Workflow Chains
+
+Pre-defined command combinations for common development scenarios:
+
+### 1. Project Initialization Chain
+
+**Purpose**: Initialize project state and guidelines
+
+```bash
+workflow init
+workflow init-specs --scope project
+workflow init-guidelines
+```
+
+**Output**: `.workflow/project-tech.json`, `.workflow/specs/*.md`
+
+---
+
+### 2. Session Lifecycle Chain
+
+**Purpose**: Complete session management workflow
+
+```bash
+workflow session start --type workflow
+# ... work on tasks ...
+workflow session sync -y
+workflow session solidify --type learning
+workflow session complete --detailed
+```
+
+---
+
+### 3. Issue Workflow Chain
+
+**Purpose**: Full issue discovery to execution cycle
+
+```bash
+issue discover --perspectives=bug,security
+issue plan --all-pending
+issue queue --queues 2
+issue execute --queue QUE-001
+```
+
+---
+
+### 4. Brainstorm to Issue Chain
+
+**Purpose**: Convert brainstorm to executable issue
+
+```bash
+workflow brainstorm-with-file -m creative
+issue from-brainstorm --auto
+issue queue
+issue execute
+```
+
+---
+
+### 5. UI Design Full Cycle
+
+**Purpose**: Complete UI design workflow
+
+```bash
+workflow ui-design style-extract --images "design/*.png"
+workflow ui-design layout-extract --images "design/*.png"
+workflow ui-design generate --design-id main-001
+```
+
+---
+
+### 6. UI Design from Code Chain
+
+**Purpose**: Extract design system from existing code
+
+```bash
+workflow ui-design import-from-code --source src/styles
+workflow ui-design reference-page-generator --design-run .workflow/style-extraction
+```
+
+---
+
+### 7. Roadmap to Team Execution Chain
+
+**Purpose**: Strategic planning to team execution
+
+```bash
+workflow roadmap-with-file -m progressive
+# Handoff to team-planex skill
+```
+
+---
+
+## Command Dependencies
+
+Some commands have prerequisites or call other commands:
+
+| Command | Depends On |
+|---------|------------|
+| `workflow session start` | `workflow init` |
+| `workflow session complete` | `workflow session sync` |
+| `workflow ui-design generate` | `style-extract`, `layout-extract` |
+| `workflow ui-design codify-style` | `import-from-code`, `reference-page-generator` |
+| `issue from-brainstorm` | `workflow brainstorm-with-file` |
+| `issue queue` | `issue plan` |
+| `issue execute` | `issue queue` |
+| `memory style-skill-memory` | `workflow ui-design codify-style` |
+
+---
+
+## Agent Delegations
+
+Commands delegate work to specialized agents:
+
+| Agent | Commands |
+|-------|----------|
+| `cli-explore-agent` | `workflow init`, `workflow clean`, `workflow brainstorm-with-file`, `workflow analyze-with-file`, `issue discover` |
+| `universal-executor` | `memory prepare` |
+| `issue-plan-agent` | `issue plan` |
+| `issue-queue-agent` | `issue queue` |
+| `ui-design-agent` | `workflow ui-design layout-extract`, `generate`, `animation-extract` |
+
+::: info See Also
+- [CLI Tools Configuration](../guide/cli-tools.md) - Configure CLI tools
+- [Skills Library](../skills/core-skills.md) - Built-in skills
+- [Agents](../agents/builtin.md) - Specialized agents
+:::

docs/commands/claude/cli.md

@@ -0,0 +1,152 @@
+# CLI Tool Commands
+
+## One-Liner
+
+**CLI tool commands are the bridge to external model invocation** — integrating Gemini, Qwen, Codex and other multi-model capabilities into workflows.
+
+## Core Concepts
+
+| Concept | Description | Configuration |
+|----------|-------------|---------------|
+| **CLI Tool** | External AI model invocation interface | `cli-tools.json` |
+| **Endpoint** | Available model services | gemini, qwen, codex, claude |
+| **Mode** | analysis / write / review | Permission level |
+
+## Command List
+
+| Command | Function | Syntax |
+|---------|----------|--------|
+| [`cli-init`](#cli-init) | Generate configuration directory and settings files | `/cli:cli-init [--tool gemini\|qwen\|all] [--output path] [--preview]` |
+| [`codex-review`](#codex-review) | Interactive code review using Codex CLI | `/cli:codex-review [--uncommitted\|--base <branch>\|--commit <sha>] [prompt]` |
+
+## Command Details
+
+### cli-init
+
+**Function**: Generate `.gemini/` and `.qwen/` configuration directories based on workspace tech detection, including settings.json and ignore files.
+
+**Syntax**:
+```
+/cli:cli-init [--tool gemini|qwen|all] [--output path] [--preview]
+```
+
+**Options**:
+- `--tool=tool`: gemini, qwen or all
+- `--output=path`: Output directory
+- `--preview`: Preview mode (don't actually create)
+
+**Generated File Structure**:
+```
+.gemini/
+├── settings.json      # Gemini configuration
+└── ignore            # Ignore patterns
+
+.qwen/
+├── settings.json      # Qwen configuration
+└── ignore            # Ignore patterns
+```
+
+**Tech Detection**:
+
+| Detection Item | Generated Config |
+|----------------|------------------|
+| TypeScript | tsconfig-related config |
+| React | React-specific config |
+| Vue | Vue-specific config |
+| Python | Python-specific config |
+
+**Examples**:
+```bash
+# Initialize all tools
+/cli:cli-init --tool all
+
+# Initialize specific tool
+/cli:cli-init --tool gemini
+
+# Specify output directory
+/cli:cli-init --output ./configs
+
+# Preview mode
+/cli:cli-init --preview
+```
+
+### codex-review
+
+**Function**: Interactive code review using Codex CLI via ccw endpoint, supporting configurable review targets, models, and custom instructions.
+
+**Syntax**:
+```
+/cli:codex-review [--uncommitted|--base <branch>|--commit <sha>] [--model <model>] [--title <title>] [prompt]
+```
+
+**Options**:
+- `--uncommitted`: Review uncommitted changes
+- `--base <branch>`: Compare with branch
+- `--commit <sha>`: Review specific commit
+- `--model <model>`: Specify model
+- `--title <title>`: Review title
+
+**Note**: Target flags and prompt are mutually exclusive
+
+**Examples**:
+```bash
+# Review uncommitted changes
+/cli:codex-review --uncommitted
+
+# Compare with main branch
+/cli:codex-review --base main
+
+# Review specific commit
+/cli:codex-review --commit abc123
+
+# With custom instructions
+/cli:codex-review --uncommitted "focus on security issues"
+
+# Specify model and title
+/cli:codex-review --model gpt-5.2 --title "Authentication module review"
+```
+
+## CLI Tool Configuration
+
+### cli-tools.json Structure
+
+```json
+{
+  "version": "3.3.0",
+  "tools": {
+    "gemini": {
+      "enabled": true,
+      "primaryModel": "gemini-2.5-flash",
+      "secondaryModel": "gemini-2.5-flash",
+      "tags": ["analysis", "Debug"],
+      "type": "builtin"
+    },
+    "qwen": {
+      "enabled": true,
+      "primaryModel": "coder-model",
+      "tags": [],
+      "type": "builtin"
+    },
+    "codex": {
+      "enabled": true,
+      "primaryModel": "gpt-5.2",
+      "tags": [],
+      "type": "builtin"
+    }
+  }
+}
+```
+
+### Mode Descriptions
+
+| Mode | Permission | Use Cases |
+|------|------------|-----------|
+| `analysis` | Read-only | Code review, architecture analysis, pattern discovery |
+| `write` | Create/modify/delete | Feature implementation, bug fixes, documentation creation |
+| `review` | Git-aware code review | Review uncommitted changes, branch diffs, specific commits |
+
+## Related Documentation
+
+- [CLI Invocation System](../../features/cli.md)
+- [Core Orchestration](./core-orchestration.md)
+- [Workflow Commands](./workflow.md)

docs/commands/claude/core-orchestration.md

@@ -0,0 +1,166 @@
+# Core Orchestration Commands
+
+## One-Liner
+
+**Core orchestration commands are the workflow brain of Claude_dms3** — analyzing task intent, selecting appropriate workflows, and automatically executing command chains.
+
+## Command List
+
+| Command | Function | Syntax |
+|---------|----------|--------|
+| [`/ccw`](#ccw) | Main workflow orchestrator - intent analysis -> workflow selection -> command chain execution | `/ccw "task description"` |
+| [`/ccw-coordinator`](#ccw-coordinator) | Command orchestration tool - chained command execution and state persistence | `/ccw-coordinator "task description"` |
+
+## Command Details
+
+### /ccw
+
+**Function**: Main workflow orchestrator - intent analysis -> workflow selection -> command chain execution
+
+**Syntax**:
+```
+/ccw "task description"
+```
+
+**Options**:
+- `--yes` / `-y`: Auto mode, skip confirmation steps
+
+**Workflow**:
+
+```mermaid
+graph TD
+    A[User Input] --> B[Analyze Intent]
+    B --> C{Clarity Score}
+    C -->|>=2| D[Select Workflow Directly]
+    C -->|<2| E[Clarify Requirements]
+    E --> D
+    D --> F[Build Command Chain]
+    F --> G{User Confirm?}
+    G -->|Yes| H[Execute Command Chain]
+    G -->|Cancel| I[End]
+    H --> J{More Steps?}
+    J -->|Yes| F
+    J -->|No| K[Complete]
+```
+
+**Task Type Detection**:
+
+| Type | Trigger Keywords | Workflow |
+|------|------------------|----------|
+| **Bug Fix** | urgent, production, critical + fix, bug | lite-fix |
+| **Brainstorming** | brainstorm, ideation | brainstorm-with-file |
+| **Debug Document** | debug document, hypothesis | debug-with-file |
+| **Collaborative Analysis** | analyze document | analyze-with-file |
+| **Collaborative Planning** | collaborative plan | collaborative-plan-with-file |
+| **Requirements Roadmap** | roadmap | req-plan-with-file |
+| **Integration Test** | integration test | integration-test-cycle |
+| **Refactoring** | refactor | refactor-cycle |
+| **Team Workflow** | team + keywords | corresponding team workflow |
+| **TDD** | tdd, test-first | tdd-plan -> execute |
+| **Test Fix** | test fix, failing test | test-fix-gen -> test-cycle-execute |
+
+**Examples**:
+
+```bash
+# Basic usage - auto-select workflow
+/ccw "implement user authentication"
+
+# Bug fix
+/ccw "fix login failure bug"
+
+# TDD development
+/ccw "implement payment using TDD"
+
+# Team collaboration
+/ccw "team-planex implement user notification system"
+```
+
+### /ccw-coordinator
+
+**Function**: Command orchestration tool - analyze tasks, recommend command chains, sequential execution, state persistence
+
+**Syntax**:
+```
+/ccw-coordinator "task description"
+```
+
+**Minimal Execution Units**:
+
+| Unit Name | Command Chain | Output |
+|-----------|---------------|--------|
+| **Quick Implementation** | lite-plan -> lite-execute | Working code |
+| **Multi-CLI Planning** | multi-cli-plan -> lite-execute | Working code |
+| **Bug Fix** | lite-plan (--bugfix) -> lite-execute | Fixed code |
+| **Full Plan+Execute** | plan -> execute | Working code |
+| **Verified Plan+Execute** | plan -> plan-verify -> execute | Working code |
+| **TDD Plan+Execute** | tdd-plan -> execute | Working code |
+| **Test Gen+Execute** | test-gen -> execute | Generated tests |
+| **Review Cycle** | review-session-cycle -> review-cycle-fix | Fixed code |
+| **Issue Workflow** | discover -> plan -> queue -> execute | Completed issue |
+
+**Workflow**:
+
+```mermaid
+graph TD
+    A[Task Analysis] --> B[Discover Commands]
+    B --> C[Recommend Command Chain]
+    C --> D{User Confirm?}
+    D -->|Yes| E[Sequential Execute]
+    D -->|Modify| F[Adjust Command Chain]
+    F --> D
+    E --> G[Persist State]
+    G --> H{More Steps?}
+    H -->|Yes| B
+    H -->|No| I[Complete]
+```
+
+**Examples**:
+
+```bash
+# Auto-orchestrate bug fix
+/ccw-coordinator "production login failure"
+
+# Auto-orchestrate feature implementation
+/ccw-coordinator "add user avatar upload"
+```
+
+## Auto Mode
+
+Both commands support the `--yes` flag for auto mode:
+
+```bash
+# Auto mode - skip all confirmations
+/ccw "implement user authentication" --yes
+/ccw-coordinator "fix login bug" --yes
+```
+
+**Auto mode behavior**:
+- Skip requirement clarification
+- Skip user confirmation
+- Execute command chain directly
+
+## Related Skills
+
+| Skill | Function |
+|-------|----------|
+| `workflow-lite-plan` | Lightweight planning workflow |
+| `workflow-plan` | Full planning workflow |
+| `workflow-execute` | Execution workflow |
+| `workflow-tdd` | TDD workflow |
+| `review-cycle` | Code review cycle |
+
+## Comparison
+
+| Feature | /ccw | /ccw-coordinator |
+|---------|------|------------------|
+| **Execution Location** | Main process | External CLI + background tasks |
+| **State Persistence** | No | Yes |
+| **Hook Callbacks** | Not supported | Supported |
+| **Complex Workflows** | Simple chains | Supports parallel, dependencies |
+| **Use Cases** | Daily development | Complex projects, team collaboration |
+
+## Related Documentation
+
+- [Workflow Commands](./workflow.md)
+- [Session Management](./session.md)
+- [CLI Invocation System](../features/cli.md)

docs/commands/claude/index.md

@@ -0,0 +1,222 @@
+# Claude Commands
+
+## One-Liner
+
+**Claude Commands is the core command system of Claude_dms3** — invoking various workflows, tools, and collaboration features through slash commands.
+
+## Core Concepts
+
+| Category | Command Count | Description |
+|----------|---------------|-------------|
+| **Core Orchestration** | 2 | Main workflow orchestrators (ccw, ccw-coordinator) |
+| **Workflow** | 20+ | Planning, execution, review, TDD, testing workflows |
+| **Session Management** | 6 | Session creation, listing, resuming, completion |
+| **Issue Workflow** | 7 | Issue discovery, planning, queue, execution |
+| **Memory** | 8 | Memory capture, update, document generation |
+| **CLI Tools** | 2 | CLI initialization, Codex review |
+| **UI Design** | 10 | UI design prototype generation, style extraction |
+
+## Command Categories
+
+### 1. Core Orchestration Commands
+
+| Command | Function | Difficulty |
+|---------|----------|------------|
+| [`/ccw`](./core-orchestration.md#ccw) | Main workflow orchestrator - intent analysis -> workflow selection -> command chain execution | Intermediate |
+| [`/ccw-coordinator`](./core-orchestration.md#ccw-coordinator) | Command orchestration tool - chained command execution and state persistence | Intermediate |
+
+### 2. Workflow Commands
+
+| Command | Function | Difficulty |
+|---------|----------|------------|
+| [`/workflow:lite-lite-lite`](./workflow.md#lite-lite-lite) | Ultra-lightweight multi-tool analysis and direct execution | Intermediate |
+| [`/workflow:lite-plan`](./workflow.md#lite-plan) | Lightweight interactive planning workflow | Intermediate |
+| [`/workflow:lite-execute`](./workflow.md#lite-execute) | Execute tasks based on in-memory plan | Intermediate |
+| [`/workflow:lite-fix`](./workflow.md#lite-fix) | Lightweight bug diagnosis and fix | Intermediate |
+| [`/workflow:plan`](./workflow.md#plan) | 5-phase planning workflow | Intermediate |
+| [`/workflow:execute`](./workflow.md#execute) | Coordinate agent execution of workflow tasks | Intermediate |
+| [`/workflow:replan`](./workflow.md#replan) | Interactive workflow replanning | Intermediate |
+| [`/workflow:multi-cli-plan`](./workflow.md#multi-cli-plan) | Multi-CLI collaborative planning | Intermediate |
+| [`/workflow:review`](./workflow.md#review) | Post-implementation review | Intermediate |
+| [`/workflow:clean`](./workflow.md#clean) | Smart code cleanup | Intermediate |
+| [`/workflow:init`](./workflow.md#init) | Initialize project state | Intermediate |
+| [`/workflow:brainstorm-with-file`](./workflow.md#brainstorm-with-file) | Interactive brainstorming | Intermediate |
+| [`/workflow:analyze-with-file`](./workflow.md#analyze-with-file) | Interactive collaborative analysis | Beginner |
+| [`/workflow:debug-with-file`](./workflow.md#debug-with-file) | Interactive hypothesis-driven debugging | Intermediate |
+| [`/workflow:unified-execute-with-file`](./workflow.md#unified-execute-with-file) | Universal execution engine | Intermediate |
+
+### 3. Session Management Commands
+
+| Command | Function | Difficulty |
+|---------|----------|------------|
+| [`/workflow:session:start`](./session.md#start) | Discover existing sessions or start new workflow session | Intermediate |
+| [`/workflow:session:list`](./session.md#list) | List all workflow sessions | Beginner |
+| [`/workflow:session:resume`](./session.md#resume) | Resume most recently paused workflow session | Intermediate |
+| [`/workflow:session:complete`](./session.md#complete) | Mark active workflow session as completed | Intermediate |
+| [`/workflow:session:solidify`](./session.md#solidify) | Crystallize session learnings into project guidelines | Intermediate |
+
+### 4. Issue Workflow Commands
+
+| Command | Function | Difficulty |
+|---------|----------|------------|
+| [`/issue:new`](./issue.md#new) | Create structured issue from GitHub URL or text description | Intermediate |
+| [`/issue:discover`](./issue.md#discover) | Discover potential issues from multiple perspectives | Intermediate |
+| [`/issue:discover-by-prompt`](./issue.md#discover-by-prompt) | Discover issues via user prompt | Intermediate |
+| [`/issue:plan`](./issue.md#plan) | Batch plan issue solutions | Intermediate |
+| [`/issue:queue`](./issue.md#queue) | Form execution queue | Intermediate |
+| [`/issue:execute`](./issue.md#execute) | Execute queue | Intermediate |
+| [`/issue:convert-to-plan`](./issue.md#convert-to-plan) | Convert planning artifact to issue solution | Intermediate |
+
+### 5. Memory Commands
+
+| Command | Function | Difficulty |
+|---------|----------|------------|
+| [`/memory:compact`](./memory.md#compact) | Compress current session memory to structured text | Intermediate |
+| [`/memory:tips`](./memory.md#tips) | Quick note-taking | Beginner |
+| [`/memory:load`](./memory.md#load) | Load task context via CLI project analysis | Intermediate |
+| [`/memory:update-full`](./memory.md#update-full) | Update all CLAUDE.md files | Intermediate |
+| [`/memory:update-related`](./memory.md#update-related) | Update CLAUDE.md for git-changed modules | Intermediate |
+| [`/memory:docs-full-cli`](./memory.md#docs-full-cli) | Generate full project documentation using CLI | Intermediate |
+| [`/memory:docs-related-cli`](./memory.md#docs-related-cli) | Generate documentation for git-changed modules | Intermediate |
+| [`/memory:style-skill-memory`](./memory.md#style-skill-memory) | Generate SKILL memory package from style reference | Intermediate |
+
+### 6. CLI Tool Commands
+
+| Command | Function | Difficulty |
+|---------|----------|------------|
+| [`/cli:cli-init`](./cli.md#cli-init) | Generate configuration directory and settings files | Intermediate |
+| [`/cli:codex-review`](./cli.md#codex-review) | Interactive code review using Codex CLI | Intermediate |
+
+### 7. UI Design Commands
+
+| Command | Function | Difficulty |
+|---------|----------|------------|
+| [`/workflow:ui-design:explore-auto`](./ui-design.md#explore-auto) | Interactive exploratory UI design workflow | Intermediate |
+| [`/workflow:ui-design:imitate-auto`](./ui-design.md#imitate-auto) | Direct code/image input UI design | Intermediate |
+| [`/workflow:ui-design:style-extract`](./ui-design.md#style-extract) | Extract design styles from reference images or prompts | Intermediate |
+| [`/workflow:ui-design:layout-extract`](./ui-design.md#layout-extract) | Extract layout information from reference images | Intermediate |
+| [`/workflow:ui-design:animation-extract`](./ui-design.md#animation-extract) | Extract animation and transition patterns | Intermediate |
+| [`/workflow:ui-design:codify-style`](./ui-design.md#codify-style) | Extract styles from code and generate shareable reference package | Intermediate |
+| [`/workflow:ui-design:generate`](./ui-design.md#generate) | Combine layout templates with design tokens to generate prototypes | Intermediate |
+
+## Auto Mode
+
+Most commands support the `--yes` or `-y` flag to enable auto mode and skip confirmation steps.
+
+```bash
+# Standard mode - requires confirmation
+/ccw "implement user authentication"
+
+# Auto mode - execute directly without confirmation
+/ccw "implement user authentication" --yes
+```
+
+## Usage Examples
+
+### Quick Analysis
+
+```bash
+# Analyze codebase structure
+/ccw "Analyze the authentication module architecture"
+
+# Quick bug diagnosis
+/ccw "Diagnose why the login timeout issue occurs"
+```
+
+### Planning & Implementation
+
+```bash
+# Create implementation plan
+/workflow:plan "Add OAuth2 authentication with Google and GitHub providers"
+
+# Execute with auto mode
+/workflow:execute --yes
+```
+
+### Code Review
+
+```bash
+# Review current changes
+/cli:codex-review
+
+# Focus on specific area
+/cli:codex-review "Focus on security vulnerabilities in auth module"
+```
+
+### Session Management
+
+```bash
+# List all sessions
+/workflow:session:list
+
+# Resume a paused session
+/workflow:session:resume "WFS-001"
+
+# Mark session as complete
+/workflow:session:complete "WFS-001"
+```
+
+### Issue Workflow
+
+```bash
+# Discover issues from codebase
+/issue:discover
+
+# Create plan for specific issue
+/issue:plan "ISSUE-001"
+
+# Execute the fix
+/issue:execute --commit
+```
+
+### Memory Management
+
+```bash
+# Capture current session learnings
+/memory:capture "Key insights from authentication refactoring"
+
+# List all memories
+/memory:list
+
+# Search memories
+/memory:search "authentication patterns"
+```
+
+### CLI Tool Invocation
+
+```bash
+# Initialize CLI configuration
+/cli:cli-init
+
+# Run Gemini analysis
+ccw cli -p "Analyze code patterns in src/auth" --tool gemini --mode analysis
+
+# Run with specific rule template
+ccw cli -p "Review code quality" --tool gemini --mode analysis --rule analysis-review-code-quality
+```
+
+### UI Design Workflow
+
+```bash
+# Extract styles from reference image
+/workflow:ui-design:style-extract --input "path/to/reference.png"
+
+# Generate prototype
+/workflow:ui-design:generate --layout "dashboard" --tokens "design-tokens.json"
+```
+
+## Tips
+
+1. **Use Auto Mode Sparingly**: Only use `--yes` or `-y` for routine tasks. Keep manual confirmation for complex decisions.
+
+2. **Session Persistence**: Always complete sessions with `/workflow:session:complete` to preserve learnings.
+
+3. **Memory Capture**: Regularly capture important insights with `/memory:capture` to build project knowledge.
+
+4. **CLI Tool Selection**: Let `/ccw` auto-select the appropriate tool, or explicitly specify with `--tool gemini|qwen|codex`.
+
+## Related Documentation
+
+- [Skills Reference](../skills/)
+- [CLI Invocation System](../features/cli.md)
+- [Workflow Guide](../guide/ch04-workflow-basics.md)

docs/commands/claude/issue.md

@@ -0,0 +1,294 @@
+# Issue Workflow Commands
+
+## One-Liner
+
+**Issue workflow commands are the closed-loop system for issue management** — from discovery, planning to execution, fully tracking the issue resolution process.
+
+## Core Concepts
+
+| Concept | Description | Location |
+|---------|-------------|----------|
+| **Issue** | Structured issue definition | `.workflow/issues/ISS-*.json` |
+| **Solution** | Execution plan | `.workflow/solutions/SOL-*.json` |
+| **Queue** | Execution queue | `.workflow/queues/QUE-*.json` |
+| **Execution State** | Progress tracking | State within queue |
+
+## Command List
+
+| Command | Function | Syntax |
+|---------|----------|--------|
+| [`new`](#new) | Create structured issue from GitHub URL or text description | `/issue:new [-y] <github-url \| description> [--priority 1-5]` |
+| [`discover`](#discover) | Discover potential issues from multiple perspectives | `/issue:discover [-y] <path pattern> [--perspectives=dimensions] [--external]` |
+| [`discover-by-prompt`](#discover-by-prompt) | Discover issues via user prompt | `/issue:discover-by-prompt [-y] <prompt> [--scope=src/**]` |
+| [`plan`](#plan) | Batch plan issue solutions | `/issue:plan [-y] --all-pending <issue-id>[,...] [--batch-size 3]` |
+| [`queue`](#queue) | Form execution queue | `/issue:queue [-y] [--queues N] [--issue id]` |
+| [`execute`](#execute) | Execute queue | `/issue:execute [-y] --queue <queue-id> [--worktree [path]]` |
+| [`convert-to-plan`](#convert-to-plan) | Convert planning artifact to issue solution | `/issue:convert-to-plan [-y] [--issue id] [--supplement] <source>` |
+
+## Command Details
+
+### new
+
+**Function**: Create structured issue from GitHub URL or text description, supporting requirement clarity detection.
+
+**Syntax**:
+```
+/issue:new [-y|--yes] <github-url | text description> [--priority 1-5]
+```
+
+**Options**:
+- `--priority 1-5`: Priority (1=critical, 5=low)
+
+**Clarity Detection**:
+
+| Input Type | Clarity | Behavior |
+|------------|---------|----------|
+| GitHub URL | 3 | Direct creation |
+| Structured text | 2 | Direct creation |
+| Long text | 1 | Partial clarification |
+| Short text | 0 | Full clarification |
+
+**Issue Structure**:
+```typescript
+interface Issue {
+  id: string;                    // GH-123 or ISS-YYYYMMDD-HHMMSS
+  title: string;
+  status: 'registered' | 'planned' | 'queued' | 'in_progress' | 'completed' | 'failed';
+  priority: number;              // 1-5
+  context: string;               // Issue description (single source of truth)
+  source: 'github' | 'text' | 'discovery';
+  source_url?: string;
+
+  // Binding
+  bound_solution_id: string | null;
+
+  // Feedback history
+  feedback?: Array<{
+    type: 'failure' | 'clarification' | 'rejection';
+    stage: string;
+    content: string;
+    created_at: string;
+  }>;
+}
+```
+
+**Examples**:
+```bash
+# Create from GitHub
+/issue:new https://github.com/owner/repo/issues/123
+
+# Create from text (structured)
+/issue:new "login failed: expected success, actual 500 error"
+
+# Create from text (vague - will ask)
+/issue:new "auth has problems"
+
+# Specify priority
+/issue:new --priority 2 "payment timeout issue"
+```
+
+### discover
+
+**Function**: Discover potential issues from multiple perspectives (Bug, UX, Test, Quality, Security, Performance, Maintainability, Best Practices).
+
+**Syntax**:
+```
+/issue:discover [-y|--yes] <path pattern> [--perspectives=bug,ux,...] [--external]
+```
+
+**Options**:
+- `--perspectives=dimensions`: Analysis dimensions
+  - `bug`: Potential bugs
+  - `ux`: UX issues
+  - `test`: Test coverage
+  - `quality`: Code quality
+  - `security`: Security issues
+  - `performance`: Performance issues
+  - `maintainability`: Maintainability
+  - `best-practices`: Best practices
+- `--external`: Use Exa external research (security, best practices)
+
+**Examples**:
+```bash
+# Comprehensive scan
+/issue:discover src/
+
+# Specific dimensions
+/issue:discover src/auth/ --perspectives=security,bug
+
+# With external research
+/issue:discover src/payment/ --perspectives=security --external
+```
+
+### discover-by-prompt
+
+**Function**: Discover issues via user prompt, using Gemini-planned iterative multi-agent exploration, supporting cross-module comparison.
+
+**Syntax**:
+```
+/issue:discover-by-prompt [-y|--yes] <prompt> [--scope=src/**] [--depth=standard|deep] [--max-iterations=5]
+```
+
+**Options**:
+- `--scope=path`: Scan scope
+- `--depth=depth`: standard or deep
+- `--max-iterations=N`: Maximum iteration count
+
+**Examples**:
+```bash
+# Standard scan
+/issue:discover-by-prompt "find auth module issues"
+
+# Deep scan
+/issue:discover-by-prompt "analyze API performance bottlenecks" --depth=deep
+
+# Specify scope
+/issue:discover-by-prompt "check database query optimization" --scope=src/db/
+```
+
+### plan
+
+**Function**: Batch plan issue solutions, using issue-plan-agent (explore + plan closed loop).
+
+**Syntax**:
+```
+/issue:plan [-y|--yes] --all-pending <issue-id>[,<issue-id>,...] [--batch-size 3]
+```
+
+**Options**:
+- `--all-pending`: Plan all pending issues
+- `--batch-size=N`: Issues per batch
+
+**Examples**:
+```bash
+# Plan specific issues
+/issue:plan ISS-20240115-001,ISS-20240115-002
+
+# Plan all pending issues
+/issue:plan --all-pending
+
+# Specify batch size
+/issue:plan --all-pending --batch-size 5
+```
+
+### queue
+
+**Function**: Form execution queue from bound solutions, using issue-queue-agent (solution level).
+
+**Syntax**:
+```
+/issue:queue [-y|--yes] [--queues <n>] [--issue <id>]
+```
+
+**Options**:
+- `--queues N`: Number of queues to create
+- `--issue id`: Specific issue
+
+**Examples**:
+```bash
+# Form queue
+/issue:queue
+
+# Create multiple queues
+/issue:queue --queues 3
+
+# Specific issue
+/issue:queue --issue ISS-20240115-001
+```
+
+### execute
+
+**Function**: Execute queue, using DAG parallel orchestration (one commit per solution).
+
+**Syntax**:
+```
+/issue:execute [-y|--yes] --queue <queue-id> [--worktree [<existing-path>]]
+```
+
+**Options**:
+- `--queue id`: Queue ID
+- `--worktree [path]`: Optional worktree path
+
+**Examples**:
+```bash
+# Execute queue
+/issue:execute --queue QUE-20240115-001
+
+# Use worktree
+/issue:execute --queue QUE-20240115-001 --worktree ../feature-branch
+```
+
+### convert-to-plan
+
+**Function**: Convert planning artifact (lite-plan, workflow session, markdown) to issue solution.
+
+**Syntax**:
+```
+/issue:convert-to-plan [-y|--yes] [--issue <id>] [--supplement] <source>
+```
+
+**Options**:
+- `--issue id`: Bind to existing issue
+- `--supplement`: Supplement mode (add to existing solution)
+
+**Source Types**:
+- lite-plan artifact
+- workflow session
+- Markdown file
+
+**Examples**:
+```bash
+# Convert from lite-plan
+/issue:convert-to-plan .workflow/sessions/WFS-xxx/artifacts/lite-plan.md
+
+# Bind to issue
+/issue:convert-to-plan --issue ISS-20240115-001 plan.md
+
+# Supplement mode
+/issue:convert-to-plan --supplement additional-plan.md
+```
+
+### from-brainstorm
+
+**Function**: Convert brainstorm session ideas to issues and generate executable solutions.
+
+**Syntax**:
+```
+/issue:from-brainstorm SESSION="session-id" [--idea=<index>] [--auto] [-y|--yes]
+```
+
+**Options**:
+- `--idea=index`: Specific idea index
+- `--auto`: Auto mode
+
+**Examples**:
+```bash
+# Convert all ideas
+/issue:from-brainstorm SESSION="WFS-brainstorm-2024-01-15"
+
+# Convert specific idea
+/issue:from-brainstorm SESSION="WFS-brainstorm-2024-01-15" --idea=3
+
+# Auto mode
+/issue:from-brainstorm --auto SESSION="WFS-brainstorm-2024-01-15"
+```
+
+## Issue Workflow
+
+```mermaid
+graph TD
+    A[Discover Issue] --> B[Create Issue]
+    B --> C[Plan Solution]
+    C --> D[Form Execution Queue]
+    D --> E[Execute Queue]
+    E --> F{Success?}
+    F -->|Yes| G[Complete]
+    F -->|No| H[Feedback Learning]
+    H --> C
+```
+
+## Related Documentation
+
+- [Workflow Commands](./workflow.md)
+- [Core Orchestration](./core-orchestration.md)
+- [Team System](../../features/)

docs/commands/claude/memory.md

@@ -0,0 +1,262 @@
+# Memory Commands
+
+## One-Liner
+
+**Memory commands are the cross-session knowledge persistence system** — capturing context, updating memory, generating documentation, making AI remember the project.
+
+## Core Concepts
+
+| Concept | Description | Location |
+|---------|-------------|----------|
+| **Memory Package** | Structured project context | MCP core_memory |
+| **CLAUDE.md** | Module-level project guide | Each module/directory |
+| **Tips** | Quick notes | `MEMORY.md` |
+| **Project Documentation** | Generated documentation | `docs/` directory |
+
+## Command List
+
+| Command | Function | Syntax |
+|---------|----------|--------|
+| [`compact`](#compact) | Compress current session memory to structured text | `/memory:compact [optional: session description]` |
+| [`tips`](#tips) | Quick note-taking | `/memory:tips <note content> [--tag tags] [--context context]` |
+| [`load`](#load) | Load task context via CLI project analysis | `/memory:load [--tool gemini\|qwen] "task context description"` |
+| [`update-full`](#update-full) | Update all CLAUDE.md files | `/memory:update-full [--tool gemini\|qwen\|codex] [--path directory]` |
+| [`update-related`](#update-related) | Update CLAUDE.md for git-changed modules | `/memory:update-related [--tool gemini\|qwen\|codex]` |
+| [`docs-full-cli`](#docs-full-cli) | Generate full project documentation using CLI | `/memory:docs-full-cli [path] [--tool tool]` |
+| [`docs-related-cli`](#docs-related-cli) | Generate documentation for git-changed modules | `/memory:docs-related-cli [--tool tool]` |
+| [`style-skill-memory`](#style-skill-memory) | Generate SKILL memory package from style reference | `/memory:style-skill-memory [package-name] [--regenerate]` |
+
+## Command Details
+
+### compact
+
+**Function**: Compress current session memory to structured text, extracting objectives, plans, files, decisions, constraints, and state, saving via MCP core_memory tool.
+
+**Syntax**:
+```
+/memory:compact [optional: session description]
+```
+
+**Extracted Content**:
+- Objectives
+- Plans
+- Files
+- Decisions
+- Constraints
+- State
+
+**Examples**:
+```bash
+# Basic compression
+/memory:compact
+
+# With description
+/memory:compact "user authentication implementation session"
+```
+
+### tips
+
+**Function**: Quick note-taking command, capturing thoughts, snippets, reminders, and insights for future reference.
+
+**Syntax**:
+```
+/memory:tips <note content> [--tag <tag1,tag2>] [--context <context>]
+```
+
+**Options**:
+- `--tag=tags`: Tags (comma-separated)
+- `--context=context`: Context information
+
+**Examples**:
+```bash
+# Basic note
+/memory:tips "remember to use rate limiting for API calls"
+
+# With tags
+/memory:tips "auth middleware needs to handle token expiry" --tag auth,api
+
+# With context
+/memory:tips "use Redis to cache user sessions" --context "login optimization"
+```
+
+### load
+
+**Function**: Delegate to universal-executor agent, analyzing project via Gemini/Qwen CLI and returning JSON core content package for task context.
+
+**Syntax**:
+```
+/memory:load [--tool gemini|qwen] "task context description"
+```
+
+**Options**:
+- `--tool=tool`: CLI tool to use
+
+**Output**: JSON format project context package
+
+**Examples**:
+```bash
+# Use default tool
+/memory:load "user authentication module"
+
+# Specify tool
+/memory:load --tool gemini "payment system architecture"
+```
+
+### update-full
+
+**Function**: Update all CLAUDE.md files, using layer-based execution (Layer 3->1), batch agent processing (4 modules/agent), and gemini->qwen->codex fallback.
+
+**Syntax**:
+```
+/memory:update-full [--tool gemini|qwen|codex] [--path <directory>]
+```
+
+**Options**:
+- `--tool=tool`: CLI tool to use
+- `--path=directory`: Specific directory
+
+**Layer Structure**:
+- Layer 3: Project-level analysis
+- Layer 2: Module-level analysis
+- Layer 1: File-level analysis
+
+**Examples**:
+```bash
+# Update entire project
+/memory:update-full
+
+# Update specific directory
+/memory:update-full --path src/auth/
+
+# Specify tool
+/memory:update-full --tool qwen
+```
+
+### update-related
+
+**Function**: Update CLAUDE.md files for git-changed modules, using batch agent execution (4 modules/agent) and gemini->qwen->codex fallback.
+
+**Syntax**:
+```
+/memory:update-related [--tool gemini|qwen|codex]
+```
+
+**Options**:
+- `--tool=tool`: CLI tool to use
+
+**Examples**:
+```bash
+# Default update
+/memory:update-related
+
+# Specify tool
+/memory:update-related --tool gemini
+```
+
+### docs-full-cli
+
+**Function**: Generate full project documentation using CLI (Layer 3->1), batch agent processing (4 modules/agent), gemini->qwen->codex fallback, direct parallel for <20 modules.
+
+**Syntax**:
+```
+/memory:docs-full-cli [path] [--tool <gemini|qwen|codex>]
+```
+
+**Examples**:
+```bash
+# Generate entire project documentation
+/memory:docs-full-cli
+
+# Generate specific directory documentation
+/memory:docs-full-cli src/
+
+# Specify tool
+/memory:docs-full-cli --tool gemini
+```
+
+### docs-related-cli
+
+**Function**: Generate documentation for git-changed modules using CLI, batch agent processing (4 modules/agent), gemini->qwen->codex fallback, direct execution for <15 modules.
+
+**Syntax**:
+```
+/memory:docs-related-cli [--tool <gemini|qwen|codex>]
+```
+
+**Examples**:
+```bash
+# Default generation
+/memory:docs-related-cli
+
+# Specify tool
+/memory:docs-related-cli --tool qwen
+```
+
+### style-skill-memory
+
+**Function**: Generate SKILL memory package from style reference, facilitating loading and consistent design system usage.
+
+**Syntax**:
+```
+/memory:style-skill-memory [package-name] [--regenerate]
+```
+
+**Options**:
+- `--regenerate`: Regenerate
+
+**Examples**:
+```bash
+# Generate style memory package
+/memory:style-skill-memory my-design-system
+
+# Regenerate
+/memory:style-skill-memory my-design-system --regenerate
+```
+
+## Memory System Workflow
+
+```mermaid
+graph TD
+    A[In Session] --> B[Capture Context]
+    B --> C{Session Complete?}
+    C -->|Yes| D[Compress Memory]
+    C -->|No| E[Continue Work]
+    D --> F[Save to core_memory]
+    F --> G[Update CLAUDE.md]
+    G --> H[Generate Documentation]
+    H --> I[New Session Starts]
+    I --> J[Load Memory Package]
+    J --> K[Restore Context]
+    K --> A
+```
+
+## CLAUDE.md Structure
+
+```markdown
+# Module Name
+
+## One-Liner
+Core value description of the module
+
+## Tech Stack
+- Framework/library
+- Main dependencies
+
+## Key Files
+- File path: Description
+
+## Code Conventions
+- Naming conventions
+- Architecture patterns
+- Best practices
+
+## TODO
+- Planned features
+- Known issues
+```
+
+## Related Documentation
+
+- [Memory System](../../features/memory.md)
+- [Core Orchestration](./core-orchestration.md)
+- [Core Concepts Guide](../../guide/ch03-core-concepts.md)

docs/commands/claude/session.md

@@ -0,0 +1,256 @@
+# Session Management Commands
+
+## One-Liner
+
+**Session management commands are the workflow state managers** — creating, tracking, resuming, and completing workflow sessions.
+
+## Core Concepts
+
+| Concept | Description | Location |
+|---------|-------------|----------|
+| **Session ID** | Unique identifier (WFS-YYYY-MM-DD) | `.workflow/active/WFS-xxx/` |
+| **Session Type** | workflow, review, tdd, test, docs | Session metadata |
+| **Session State** | active, paused, completed | workflow-session.json |
+| **Artifacts** | Plans, tasks, TODOs, etc. | Session directory |
+
+## Command List
+
+| Command | Function | Syntax |
+|---------|----------|--------|
+| [`start`](#start) | Discover existing sessions or start new workflow session | `/workflow:session:start [--type type] [--auto\|--new] [description]` |
+| [`list`](#list) | List all workflow sessions | `/workflow:session:list` |
+| [`resume`](#resume) | Resume most recently paused workflow session | `/workflow:session:resume` |
+| [`complete`](#complete) | Mark active workflow session as completed | `/workflow:session:complete [-y] [--detailed]` |
+| [`solidify`](#solidify) | Crystallize session learnings into project guidelines | `/workflow:session:solidify [-y] [--type type] [--category category] "rule"` |
+
+## Command Details
+
+### start
+
+**Function**: Discover existing sessions or start new workflow session, supporting intelligent session management and conflict detection.
+
+**Syntax**:
+```
+/workflow:session:start [--type <workflow|review|tdd|test|docs>] [--auto|--new] [optional: task description]
+```
+
+**Options**:
+- `--type=type`: Session type
+  - `workflow`: Standard implementation (default)
+  - `review`: Code review
+  - `tdd`: TDD development
+  - `test`: Test generation/fix
+  - `docs`: Documentation session
+- `--auto`: Smart mode (auto detect/create)
+- `--new`: Force create new session
+
+**Session Types**:
+
+| Type | Description | Default Source |
+|------|-------------|----------------|
+| `workflow` | Standard implementation | workflow-plan skill |
+| `review` | Code review | review-cycle skill |
+| `tdd` | TDD development | workflow-tdd skill |
+| `test` | Test generation/fix | workflow-test-fix skill |
+| `docs` | Documentation session | memory-manage skill |
+
+**Workflow**:
+
+```mermaid
+graph TD
+    A[Start] --> B{Project State Exists?}
+    B -->|No| C[Call workflow:init]
+    C --> D
+    B -->|Yes| D{Mode}
+    D -->|Default| E[List Active Sessions]
+    D -->|auto| F{Active Sessions Count?}
+    D -->|new| G[Create New Session]
+    F -->|0| G
+    F -->|1| H[Use Existing Session]
+    F -->|>1| I[User Selects]
+    E --> J{User Selects}
+    J -->|Existing| K[Return Session ID]
+    J -->|New| G
+    G --> L[Generate Session ID]
+    L --> M[Create Directory Structure]
+    M --> N[Initialize Metadata]
+    N --> O[Return Session ID]
+```
+
+**Examples**:
+
+```bash
+# Discovery mode - list active sessions
+/workflow:session:start
+
+# Auto mode - smart select/create
+/workflow:session:start --auto "implement user authentication"
+
+# New mode - force create new session
+/workflow:session:start --new "refactor payment module"
+
+# Specify type
+/workflow:session:start --type review "review auth code"
+/workflow:session:start --type tdd --auto "implement login feature"
+```
+
+### list
+
+**Function**: List all workflow sessions, supporting state filtering, displaying session metadata and progress information.
+
+**Syntax**:
+```
+/workflow:session:list
+```
+
+**Output Format**:
+
+| Session ID | Type | State | Description | Progress |
+|------------|------|-------|-------------|----------|
+| WFS-2024-01-15 | workflow | active | User authentication | 5/10 |
+| WFS-2024-01-14 | review | paused | Code review | 8/8 |
+| WFS-2024-01-13 | tdd | completed | TDD development | 12/12 |
+
+**Examples**:
+```bash
+# List all sessions
+/workflow:session:list
+```
+
+### resume
+
+**Function**: Resume most recently paused workflow session, supporting automatic session discovery and state update.
+
+**Syntax**:
+```
+/workflow:session:resume
+```
+
+**Workflow**:
+
+```mermaid
+graph TD
+    A[Start] --> B[Find Paused Sessions]
+    B --> C{Found Paused Session?}
+    C -->|Yes| D[Load Session]
+    C -->|No| E[Error Message]
+    D --> F[Update State to Active]
+    F --> G[Return Session ID]
+```
+
+**Examples**:
+```bash
+# Resume most recently paused session
+/workflow:session:resume
+```
+
+### complete
+
+**Function**: Mark active workflow session as completed, archive and learn from experience, update checklist and remove active flag.
+
+**Syntax**:
+```
+/workflow:session:complete [-y|--yes] [--detailed]
+```
+
+**Options**:
+- `--detailed`: Detailed mode, collect more learnings
+
+**Workflow**:
+
+```mermaid
+graph TD
+    A[Start] --> B[Confirm Completion]
+    B --> C{Detailed Mode?}
+    C -->|Yes| D[Collect Detailed Feedback]
+    C -->|No| E[Collect Basic Feedback]
+    D --> F[Generate Learning Document]
+    E --> F
+    F --> G[Archive Session]
+    G --> H[Update Checklist]
+    H --> I[Remove Active Flag]
+    I --> J[Complete]
+```
+
+**Examples**:
+```bash
+# Standard completion
+/workflow:session:complete
+
+# Detailed completion
+/workflow:session:complete --detailed
+
+# Auto mode
+/workflow:session:complete -y
+```
+
+### solidify
+
+**Function**: Crystallize session learnings and user-defined constraints into permanent project guidelines.
+
+**Syntax**:
+```
+/workflow:session:solidify [-y|--yes] [--type <convention|constraint|learning>] [--category <category>] "rule or insight"
+```
+
+**Options**:
+- `--type=type`:
+  - `convention`: Code convention
+  - `constraint`: Constraint condition
+  - `learning`: Experience learning
+- `--category=category`: Category name (e.g., `authentication`, `testing`)
+
+**Output Locations**:
+- Conventions: `.workflow/specs/conventions/<category>.md`
+- Constraints: `.workflow/specs/constraints/<category>.md`
+- Learnings: `.workflow/specs/learnings/<category>.md`
+
+**Examples**:
+```bash
+# Add code convention
+/workflow:session:solidify --type=convention --category=auth "all auth functions must use rate limiting"
+
+# Add constraint
+/workflow:session:solidify --type=constraint --category=database "no N+1 queries"
+
+# Add learning
+/workflow:session:solidify --type=learning --category=api "REST API design lessons learned"
+```
+
+## Session Directory Structure
+
+```
+.workflow/
+├── active/                        # Active sessions
+│   └── WFS-2024-01-15/           # Session directory
+│       ├── workflow-session.json  # Session metadata
+│       ├── tasks/                 # Task definitions
+│       ├── artifacts/             # Artifact files
+│       └── context/               # Context files
+└── archived/                      # Archived sessions
+    └── WFS-2024-01-14/
+```
+
+## Session Metadata
+
+```json
+{
+  "session_id": "WFS-2024-01-15",
+  "type": "workflow",
+  "status": "active",
+  "created_at": "2024-01-15T10:00:00Z",
+  "updated_at": "2024-01-15T14:30:00Z",
+  "description": "User authentication feature implementation",
+  "progress": {
+    "total": 10,
+    "completed": 5,
+    "percentage": 50
+  }
+}
+```
+
+## Related Documentation
+
+- [Workflow Commands](./workflow.md)
+- [Core Orchestration](./core-orchestration.md)
+- [Workflow Basics](../../guide/ch04-workflow-basics.md)

docs/commands/claude/ui-design.md

@@ -0,0 +1,307 @@
+# UI Design Commands
+
+## One-Liner
+
+**UI design commands are the interface prototype generation system** — from style extraction, layout analysis to prototype assembly, fully covering the UI design workflow.
+
+## Core Concepts
+
+| Concept | Description | Storage Location |
+|----------|-------------|------------------|
+| **Design Run** | Design session directory | `.workflow/ui-design-runs/<run-id>/` |
+| **Design Tokens** | Style variables | `design-tokens.json` |
+| **Layout Templates** | Structure definitions | `layouts/` |
+| **Prototypes** | Generated components | `prototypes/` |
+
+## Command List
+
+### Discovery and Extraction
+
+| Command | Function | Syntax |
+|---------|----------|--------|
+| [`explore-auto`](#explore-auto) | Interactive exploratory UI design workflow | `/workflow:ui-design:explore-auto [--input "value"] [--targets "list"]` |
+| [`imitate-auto`](#imitate-auto) | Direct code/image input UI design | `/workflow:ui-design:imitate-auto [--input "value"] [--session id]` |
+| [`style-extract`](#style-extract) | Extract design styles from reference images or prompts | `/workflow:ui-design:style-extract [-y] [--design-id id]` |
+| [`layout-extract`](#layout-extract) | Extract layout information from reference images | `/workflow:ui-design:layout-extract [-y] [--design-id id]` |
+| [`animation-extract`](#animation-extract) | Extract animation and transition patterns | `/workflow:ui-design:animation-extract [-y] [--design-id id]` |
+
+### Import and Export
+
+| Command | Function | Syntax |
+|---------|----------|--------|
+| [`import-from-code`](#import-from-code) | Import design system from code files | `/workflow:ui-design:import-from-code [--design-id id] [--session id] [--source path]` |
+| [`codify-style`](#codify-style) | Extract styles from code and generate shareable reference package | `/workflow:ui-design:codify-style <path> [--package-name name]` |
+| [`reference-page-generator`](#reference-page-generator) | Generate multi-component reference page from design run | `/workflow:ui-design:reference-page-generator [--design-run path]` |
+
+### Generation and Sync
+
+| Command | Function | Syntax |
+|---------|----------|--------|
+| [`generate`](#generate) | Combine layout templates with design tokens to generate prototypes | `/workflow:ui-design:generate [--design-id id] [--session id]` |
+| [`design-sync`](#design-sync) | Sync final design system reference to brainstorm artifacts | `/workflow:ui-design:design-sync --session <session_id>` |
+
+## Command Details
+
+### explore-auto
+
+**Function**: Interactive exploratory UI design workflow, style-centric batch generation, creating design variants from prompts/images, supporting parallel execution and user selection.
+
+**Syntax**:
+```
+/workflow:ui-design:explore-auto [--input "<value>"] [--targets "<list>"] [--target-type "page|component"] [--session <id>] [--style-variants <count>] [--layout-variants <count>]
+```
+
+**Options**:
+- `--input=value`: Input prompt or image path
+- `--targets=list`: Target component list (comma-separated)
+- `--target-type=type`: page or component
+- `--session=id`: Session ID
+- `--style-variants=N`: Number of style variants
+- `--layout-variants=N`: Number of layout variants
+
+**Examples**:
+```bash
+# Page design exploration
+/workflow:ui-design:explore-auto --input "modern e-commerce homepage" --target-type page --style-variants 3
+
+# Component design exploration
+/workflow:ui-design:explore-auto --input "user card component" --target-type component --layout-variants 5
+
+# Multi-target design
+/workflow:ui-design:explore-auto --targets "header,sidebar,footer" --style-variants 2
+```
+
+### imitate-auto
+
+**Function**: UI design workflow supporting direct code/image input for design token extraction and prototype generation.
+
+**Syntax**:
+```
+/workflow:ui-design:imitate-auto [--input "<value>"] [--session <id>]
+```
+
+**Options**:
+- `--input=value`: Code file path or image path
+- `--session=id`: Session ID
+
+**Examples**:
+```bash
+# Imitate from code
+/workflow:ui-design:imitate-auto --input "./src/components/Button.tsx"
+
+# Imitate from image
+/workflow:ui-design:imitate-auto --input "./designs/mockup.png"
+```
+
+### style-extract
+
+**Function**: Extract design styles using Claude analysis from reference images or text prompts, supporting variant generation or refine mode.
+
+**Syntax**:
+```
+/workflow:ui-design:style-extract [-y|--yes] [--design-id <id>] [--session <id>] [--images "<glob>"] [--prompt "<description>"] [--variants <count>] [--interactive] [--refine]
+```
+
+**Options**:
+- `--images=glob`: Image glob pattern
+- `--prompt=description`: Text description
+- `--variants=N`: Number of variants
+- `--interactive`: Interactive mode
+- `--refine`: Refine mode
+
+**Examples**:
+```bash
+# Extract styles from images
+/workflow:ui-design:style-extract --images "./designs/*.png" --variants 3
+
+# Extract from prompt
+/workflow:ui-design:style-extract --prompt "dark theme, blue primary, rounded corners"
+
+# Interactive refine
+/workflow:ui-design:style-extract --images "reference.png" --refine --interactive
+```
+
+### layout-extract
+
+**Function**: Extract structural layout information using Claude analysis from reference images or text prompts, supporting variant generation or refine mode.
+
+**Syntax**:
+```
+/workflow:ui-design:layout-extract [-y|--yes] [--design-id <id>] [--session <id>] [--images "<glob>"] [--prompt "<description>"] [--targets "<list>"] [--variants <count>] [--device-type <desktop|mobile|tablet|responsive>] [--interactive] [--refine]
+```
+
+**Options**:
+- `--device-type=type`: desktop, mobile, tablet or responsive
+- Other options same as style-extract
+
+**Examples**:
+```bash
+# Extract desktop layout
+/workflow:ui-design:layout-extract --images "desktop-mockup.png" --device-type desktop
+
+# Extract responsive layout
+/workflow:ui-design:layout-extract --prompt "three-column layout, responsive design" --device-type responsive
+
+# Multiple variants
+/workflow:ui-design:layout-extract --images "layout.png" --variants 5
+```
+
+### animation-extract
+
+**Function**: Extract animation and transition patterns from prompts inference and image references for design system documentation.
+
+**Syntax**:
+```
+/workflow:ui-design:animation-extract [-y|--yes] [--design-id <id>] [--session <id>] [--images "<glob>"] [--focus "<type>"] [--interactive] [--refine]
+```
+
+**Options**:
+- `--focus=type`: Specific animation type (e.g., fade, slide, scale)
+
+**Examples**:
+```bash
+# Extract all animations
+/workflow:ui-design:animation-extract --images "./animations/*.gif"
+
+# Extract specific type
+/workflow:ui-design:animation-extract --focus "fade,slide" --interactive
+```
+
+### import-from-code
+
+**Function**: Import design system from code files (CSS/JS/HTML/SCSS), using automatic file discovery and parallel agent analysis.
+
+**Syntax**:
+```
+/workflow:ui-design:import-from-code [--design-id <id>] [--session <id>] [--source <path>]
+```
+
+**Options**:
+- `--source=path`: Source code directory
+
+**Examples**:
+```bash
+# Import from project
+/workflow:ui-design:import-from-code --source "./src/styles/"
+
+# Specify design ID
+/workflow:ui-design:import-from-code --design-id my-design --source "./theme/"
+```
+
+### codify-style
+
+**Function**: Orchestrator extracts styles from code and generates shareable reference package, supporting preview (automatic file discovery).
+
+**Syntax**:
+```
+/workflow:ui-design:codify-style <path> [--package-name <name>] [--output-dir <path>] [--overwrite]
+```
+
+**Options**:
+- `--package-name=name`: Package name
+- `--output-dir=path`: Output directory
+- `--overwrite`: Overwrite existing files
+
+**Examples**:
+```bash
+# Generate style package
+/workflow:ui-design:codify-style ./src/styles/ --package-name my-design-system
+
+# Specify output directory
+/workflow:ui-design:codify-style ./theme/ --output-dir ./design-packages/
+```
+
+### reference-page-generator
+
+**Function**: Extract and generate multi-component reference pages and documentation from design run.
+
+**Syntax**:
+```
+/workflow:ui-design:reference-page-generator [--design-run <path>] [--package-name <name>] [--output-dir <path>]
+```
+
+**Examples**:
+```bash
+# Generate reference page
+/workflow:ui-design:reference-page-generator --design-run .workflow/ui-design-runs/latest/
+
+# Specify package name
+/workflow:ui-design:reference-page-generator --package-name component-library
+```
+
+### generate
+
+**Function**: Assemble UI prototypes, combining layout templates with design tokens (default animation support), pure assembler with no new content generation.
+
+**Syntax**:
+```
+/workflow:ui-design:generate [--design-id <id>] [--session <id>]
+```
+
+**Examples**:
+```bash
+# Generate prototypes
+/workflow:ui-design:generate
+
+# Use specific design
+/workflow:ui-design:generate --design-id my-design
+```
+
+### design-sync
+
+**Function**: Sync final design system reference to brainstorm artifacts, preparing for consumption by `/workflow:plan`.
+
+**Syntax**:
+```
+/workflow:ui-design:design-sync --session <session_id> [--selected-prototypes "<list>"]
+```
+
+**Options**:
+- `--selected-prototypes=list`: List of selected prototypes
+
+**Examples**:
+```bash
+# Sync all prototypes
+/workflow:ui-design:design-sync --session WFS-design-2024-01-15
+
+# Sync selected prototypes
+/workflow:ui-design:design-sync --session WFS-design-2024-01-15 --selected-prototypes "header,button,card"
+```
+
+## UI Design Workflow
+
+```mermaid
+graph TD
+    A[Input] --> B{Input Type}
+    B -->|Image/Prompt| C[Extract Styles]
+    B -->|Code| D[Import Styles]
+    C --> E[Extract Layouts]
+    D --> E
+    E --> F[Extract Animations]
+    F --> G[Design Run]
+    G --> H[Generate Prototypes]
+    H --> I[Sync to Brainstorm]
+    I --> J[Planning Workflow]
+```
+
+## Design Run Structure
+
+```
+.workflow/ui-design-runs/<run-id>/
+├── design-tokens.json       # Design tokens
+├── layouts/                 # Layout templates
+│   ├── header.json
+│   ├── footer.json
+│   └── ...
+├── prototypes/              # Generated prototypes
+│   ├── header/
+│   ├── button/
+│   └── ...
+└── reference-pages/         # Reference pages
+```
+
+## Related Documentation
+
+- [Core Orchestration](./core-orchestration.md)
+- [Workflow Commands](./workflow.md)
+- [Brainstorming](../../features/)

docs/commands/claude/workflow.md

@@ -0,0 +1,338 @@
+# Workflow Commands
+
+## One-Liner
+
+**Workflow commands are the execution engine of Claude_dms3** — providing complete workflow support from lightweight tasks to complex projects.
+
+## Command List
+
+### Lightweight Workflows
+
+| Command | Function | Syntax |
+|---------|----------|--------|
+| [`lite-lite-lite`](#lite-lite-lite) | Ultra-lightweight multi-tool analysis and direct execution | `/workflow:lite-lite-lite [-y] <task>` |
+| [`lite-plan`](#lite-plan) | Lightweight interactive planning workflow | `/workflow:lite-plan [-y] [-e] "task"` |
+| [`lite-execute`](#lite-execute) | Execute tasks based on in-memory plan | `/workflow:lite-execute [-y] [--in-memory] [task]` |
+| [`lite-fix`](#lite-fix) | Lightweight bug diagnosis and fix | `/workflow:lite-fix [-y] [--hotfix] "bug description"` |
+
+### Standard Workflows
+
+| Command | Function | Syntax |
+|---------|----------|--------|
+| [`plan`](#plan) | 5-phase planning workflow | `/workflow:plan [-y] "description"\|file.md` |
+| [`execute`](#execute) | Coordinate agent execution of workflow tasks | `/workflow:execute [-y] [--resume-session=ID]` |
+| [`replan`](#replan) | Interactive workflow replanning | `/workflow:replan [-y] [--session ID] [task-id] "requirement"` |
+
+### Collaborative Workflows
+
+| Command | Function | Syntax |
+|---------|----------|--------|
+| [`multi-cli-plan`](#multi-cli-plan) | Multi-CLI collaborative planning | `/workflow:multi-cli-plan [-y] <task> [--max-rounds=N]` |
+| [`brainstorm-with-file`](#brainstorm-with-file) | Interactive brainstorming | `/workflow:brainstorm-with-file [-y] [-c] "idea"` |
+| [`analyze-with-file`](#analyze-with-file) | Interactive collaborative analysis | `/workflow:analyze-with-file [-y] [-c] "topic"` |
+| [`debug-with-file`](#debug-with-file) | Interactive hypothesis-driven debugging | `/workflow:debug-with-file [-y] "bug description"` |
+| [`unified-execute-with-file`](#unified-execute-with-file) | Universal execution engine | `/workflow:unified-execute-with-file [-y] [-p path] [context]` |
+
+### TDD Workflows
+
+| Command | Function | Syntax |
+|---------|----------|--------|
+| [`tdd-plan`](#tdd-plan) | TDD planning workflow | `/workflow:tdd-plan "feature description"` |
+| [`tdd-verify`](#tdd-verify) | Verify TDD workflow compliance | `/workflow:tdd-verify [--session ID]` |
+
+### Test Workflows
+
+| Command | Function | Syntax |
+|---------|----------|--------|
+| [`test-fix-gen`](#test-fix-gen) | Create test-fix workflow session | `/workflow:test-fix-gen (session-id\|"description"\|file.md)` |
+| [`test-gen`](#test-gen) | Create test session from implementation session | `/workflow:test-gen source-session-id` |
+| [`test-cycle-execute`](#test-cycle-execute) | Execute test-fix workflow | `/workflow:test-cycle-execute [--resume-session=ID]` |
+
+### Review Workflows
+
+| Command | Function | Syntax |
+|---------|----------|--------|
+| [`review`](#review) | Post-implementation review | `/workflow:review [--type=type] [--archived] [session-id]` |
+| [`review-module-cycle`](#review-module-cycle) | Standalone multi-dimensional code review | `/workflow:review-module-cycle <path> [--dimensions=dimensions]` |
+| [`review-session-cycle`](#review-session-cycle) | Session-based review | `/workflow:review-session-cycle [session-id] [--dimensions=dimensions]` |
+| [`review-cycle-fix`](#review-cycle-fix) | Auto-fix review findings | `/workflow:review-cycle-fix <export-file\|review-dir>` |
+
+### Specialized Workflows
+
+| Command | Function | Syntax |
+|---------|----------|--------|
+| [`clean`](#clean) | Smart code cleanup | `/workflow:clean [-y] [--dry-run] ["focus area"]` |
+| [`init`](#init) | Initialize project state | `/workflow:init [--regenerate]` |
+| [`plan-verify`](#plan-verify) | Verify planning consistency | `/workflow:plan-verify [--session session-id]` |
+
+## Command Details
+
+### lite-lite-lite
+
+**Function**: Ultra-lightweight multi-tool analysis and direct execution. Simple tasks have no artifacts, complex tasks automatically create planning documents in `.workflow/.scratchpad/`.
+
+**Syntax**:
+```
+/workflow:lite-lite-lite [-y|--yes] <task description>
+```
+
+**Use Cases**:
+- Ultra-simple quick tasks
+- Code modifications not needing planning documents
+- Automatic tool selection
+
+**Examples**:
+```bash
+# Ultra-simple task
+/workflow:lite-lite-lite "fix header styles"
+
+# Auto mode
+/workflow:lite-lite-lite -y "update README links"
+```
+
+### lite-plan
+
+**Function**: Lightweight interactive planning workflow, supporting in-memory planning, code exploration, and execution to lite-execute.
+
+**Syntax**:
+```
+/workflow:lite-plan [-y|--yes] [-e|--explore] "task description" | file.md
+```
+
+**Options**:
+- `-e, --explore`: Execute code exploration first
+
+**Examples**:
+```bash
+# Basic planning
+/workflow:lite-plan "add user avatar feature"
+
+# With exploration
+/workflow:lite-plan -e "refactor authentication module"
+```
+
+### lite-execute
+
+**Function**: Execute tasks based on in-memory plan, prompt description, or file content.
+
+**Syntax**:
+```
+/workflow:lite-execute [-y|--yes] [--in-memory] ["task description" | file-path]
+```
+
+**Options**:
+- `--in-memory`: Use in-memory plan
+
+**Examples**:
+```bash
+# Execute task
+/workflow:lite-execute "implement avatar upload API"
+
+# Use in-memory plan
+/workflow:lite-execute --in-memory
+```
+
+### lite-fix
+
+**Function**: Lightweight bug diagnosis and fix workflow, supporting intelligent severity assessment and optional hotfix mode.
+
+**Syntax**:
+```
+/workflow:lite-fix [-y|--yes] [--hotfix] "bug description or issue reference"
+```
+
+**Options**:
+- `--hotfix`: Hotfix mode (quick fix for production incidents)
+
+**Examples**:
+```bash
+# Bug fix
+/workflow:lite-fix "login returns 500 error"
+
+# Hotfix
+/workflow:lite-fix --hotfix "payment gateway timeout"
+```
+
+### plan
+
+**Function**: 5-phase planning workflow, outputting IMPL_PLAN.md and task JSON.
+
+**Syntax**:
+```
+/workflow:plan [-y|--yes] "text description" | file.md
+```
+
+**Phases**:
+1. Session initialization
+2. Context collection
+3. Specification loading
+4. Task generation
+5. Verification/replanning
+
+**Examples**:
+```bash
+# Plan from description
+/workflow:plan "implement user notification system"
+
+# Plan from file
+/workflow:plan requirements.md
+```
+
+### execute
+
+**Function**: Coordinate agent execution of workflow tasks, supporting automatic session discovery, parallel task processing, and state tracking.
+
+**Syntax**:
+```
+/workflow:execute [-y|--yes] [--resume-session="session-id"]
+```
+
+**Examples**:
+```bash
+# Execute current session
+/workflow:execute
+
+# Resume and execute session
+/workflow:execute --resume-session=WFS-2024-01-15
+```
+
+### replan
+
+**Function**: Interactive workflow replanning, supporting session-level artifact updates and scope clarification.
+
+**Syntax**:
+```
+/workflow:replan [-y|--yes] [--session session-id] [task-id] "requirement" | file.md [--interactive]
+```
+
+**Examples**:
+```bash
+# Replan entire session
+/workflow:replan --session=WFS-xxx "add user permission checks"
+
+# Replan specific task
+/workflow:replan TASK-001 "change to use RBAC"
+```
+
+### multi-cli-plan
+
+**Function**: Multi-CLI collaborative planning workflow, using ACE context collection and iterative cross-validation.
+
+**Syntax**:
+```
+/workflow:multi-cli-plan [-y|--yes] <task description> [--max-rounds=3] [--tools=gemini,codex] [--mode=parallel|serial]
+```
+
+**Options**:
+- `--max-rounds=N`: Maximum discussion rounds
+- `--tools=tools`: CLI tools to use
+- `--mode=mode`: Parallel or serial mode
+
+**Examples**:
+```bash
+# Multi-CLI planning
+/workflow:multi-cli-plan "design microservice architecture"
+
+# Specify tools and rounds
+/workflow:multi-cli-plan --tools=gemini,codex --max-rounds=5 "database migration plan"
+```
+
+### brainstorm-with-file
+
+**Function**: Interactive brainstorming, multi-CLI collaboration, idea expansion, and documented thinking evolution.
+
+**Syntax**:
+```
+/workflow:brainstorm-with-file [-y|--yes] [-c|--continue] [-m|--mode creative|structured] "idea or topic"
+```
+
+**Options**:
+- `-c, --continue`: Continue existing session
+- `-m, --mode=mode`: creative or structured
+
+**Examples**:
+```bash
+# Creative brainstorming
+/workflow:brainstorm-with-file --mode creative "user growth strategies"
+
+# Structured brainstorming
+/workflow:brainstorm-with-file --mode structured "API versioning approach"
+```
+
+### analyze-with-file
+
+**Function**: Interactive collaborative analysis with documented discussion, CLI-assisted exploration, and evolving understanding.
+
+**Syntax**:
+```
+/workflow:analyze-with-file [-y|--yes] [-c|--continue] "topic or question"
+```
+
+**Examples**:
+```bash
+# Analyze topic
+/workflow:analyze-with-file "authentication architecture design"
+
+# Continue discussion
+/workflow:analyze-with-file -c
+```
+
+### debug-with-file
+
+**Function**: Interactive hypothesis-driven debugging with documented exploration, understanding evolution, and Gemini-assisted correction.
+
+**Syntax**:
+```
+/workflow:debug-with-file [-y|--yes] "bug description or error message"
+```
+
+**Examples**:
+```bash
+# Debug bug
+/workflow:debug-with-file "WebSocket connection randomly disconnects"
+
+# Debug error
+/workflow:debug-with-file "TypeError: Cannot read property 'id'"
+```
+
+### unified-execute-with-file
+
+**Function**: Universal execution engine consuming any planning/brainstorming/analysis output, supporting minimal progress tracking, multi-agent coordination, and incremental execution.
+
+**Syntax**:
+```
+/workflow:unified-execute-with-file [-y|--yes] [-p|--plan <path>] [-m|--mode sequential|parallel] ["execution context"]
+```
+
+**Examples**:
+```bash
+# Execute plan
+/workflow:unified-execute-with-file -p plan.md
+
+# Parallel execution
+/workflow:unified-execute-with-file -m parallel
+```
+
+## Workflow Diagram
+
+```mermaid
+graph TD
+    A[Task Input] --> B{Task Complexity}
+    B -->|Simple| C[Lite Workflow]
+    B -->|Standard| D[Plan Workflow]
+    B -->|Complex| E[Multi-CLI Workflow]
+
+    C --> F[Direct Execution]
+    D --> G[Plan] --> H[Execute]
+    E --> I[Multi-CLI Discussion] --> G
+
+    F --> J[Complete]
+    H --> J
+    I --> J
+```
+
+## Related Documentation
+
+- [Session Management](./session.md)
+- [Core Orchestration](./core-orchestration.md)
+- [Workflow Guide](../../guide/ch04-workflow-basics.md)

docs/commands/codex/index.md

@@ -0,0 +1,56 @@
+# Codex Prompts
+
+## One-Liner
+
+**Codex Prompts is the prompt template system used by Codex CLI** — standardized prompt formats ensure consistent code quality and review effectiveness.
+
+## Core Concepts
+
+| Concept | Description | Use Cases |
+|----------|-------------|-----------|
+| **Prep Prompts** | Project context preparation prompts | Analyze project structure, extract relevant files |
+| **Review Prompts** | Code review prompts | Multi-dimensional code quality checks |
+
+## Prompt List
+
+### Prep Series
+
+| Prompt | Function | Use Cases |
+|--------|----------|-----------|
+| [`memory:prepare`](./prep.md#memory-prepare) | Project context preparation | Prepare structured project context for tasks |
+
+### Review Series
+
+| Prompt | Function | Use Cases |
+|--------|----------|-----------|
+| [`codex-review`](./review.md#codex-review) | Interactive code review | Code review using Codex CLI |
+
+## Prompt Template Format
+
+All Codex Prompts follow the standard CCW CLI prompt template:
+
+```
+PURPOSE: [objective] + [reason] + [success criteria] + [constraints/scope]
+TASK: • [step 1] • [step 2] • [step 3]
+MODE: review
+CONTEXT: [review target description] | Memory: [relevant context]
+EXPECTED: [deliverable format] + [quality criteria]
+CONSTRAINTS: [focus constraints]
+```
+
+## Field Descriptions
+
+| Field | Description | Example |
+|-------|-------------|---------|
+| **PURPOSE** | Objective and reason | "Identify security vulnerabilities to ensure code safety" |
+| **TASK** | Specific steps | "• Scan for injection vulnerabilities • Check authentication logic" |
+| **MODE** | Execution mode | analysis, write, review |
+| **CONTEXT** | Context information | "@CLAUDE.md @src/auth/**" |
+| **EXPECTED** | Output format | "Structured report with severity levels" |
+| **CONSTRAINTS** | Constraint conditions | "Focus on actionable suggestions" |
+
+## Related Documentation
+
+- [Claude Commands](../claude/)
+- [CLI Invocation System](../../features/cli.md)
+- [Code Review](../../features/)

docs/commands/codex/prep.md

@@ -0,0 +1,168 @@
+# Prep Prompts
+
+## One-Liner
+
+**Prep prompts are standardized templates for project context preparation** — generating structured project core content packages through agent-driven analysis.
+
+## Core Content Package Structure
+
+```json
+{
+  "task_context": "Task context description",
+  "keywords": ["keyword1", "keyword2"],
+  "project_summary": {
+    "architecture": "Architecture description",
+    "tech_stack": ["tech1", "tech2"],
+    "key_patterns": ["pattern1", "pattern2"]
+  },
+  "relevant_files": [
+    {
+      "path": "file path",
+      "relevance": "relevance description",
+      "priority": "high|medium|low"
+    }
+  ],
+  "integration_points": [
+    "integration point 1",
+    "integration point 2"
+  ],
+  "constraints": [
+    "constraint 1",
+    "constraint 2"
+  ]
+}
+```
+
+## memory:prepare
+
+**Function**: Delegate to universal-executor agent, analyzing project via Gemini/Qwen CLI and returning JSON core content package for task context.
+
+**Syntax**:
+```
+/memory:prepare [--tool gemini|qwen] "task context description"
+```
+
+**Options**:
+- `--tool=tool`: Specify CLI tool (default: gemini)
+  - `gemini`: Large context window, suitable for complex project analysis
+  - `qwen`: Gemini alternative with similar capabilities
+
+**Execution Flow**:
+
+```mermaid
+graph TD
+    A[Start] --> B[Analyze Project Structure]
+    B --> C[Load Documentation]
+    C --> D[Extract Keywords]
+    D --> E[Discover Files]
+    E --> F[CLI Deep Analysis]
+    F --> G[Generate Content Package]
+    G --> H[Load to Main Thread Memory]
+```
+
+**Agent Call Prompt**:
+```
+## Mission: Prepare Project Memory Context
+
+**Task**: Prepare project memory context for: "{task_description}"
+**Mode**: analysis
+**Tool Preference**: {tool}
+
+### Step 1: Foundation Analysis
+1. Project Structure: get_modules_by_depth.sh
+2. Core Documentation: CLAUDE.md, README.md
+
+### Step 2: Keyword Extraction & File Discovery
+1. Extract core keywords from task description
+2. Discover relevant files using ripgrep and find
+
+### Step 3: Deep Analysis via CLI
+Execute Gemini/Qwen CLI for deep analysis
+
+### Step 4: Generate Core Content Package
+Return structured JSON with required fields
+
+### Step 5: Return Content Package
+Load JSON into main thread memory
+```
+
+**Examples**:
+
+```bash
+# Basic usage
+/memory:prepare "develop user authentication on current frontend"
+
+# Specify tool
+/memory:prepare --tool qwen "refactor payment module API"
+
+# Bug fix context
+/memory:prepare "fix login validation error"
+```
+
+**Returned Content Package**:
+
+```json
+{
+  "task_context": "develop user authentication on current frontend",
+  "keywords": ["frontend", "user", "authentication", "auth", "login"],
+  "project_summary": {
+    "architecture": "TypeScript + React frontend, Vite build system",
+    "tech_stack": ["React", "TypeScript", "Vite", "TailwindCSS"],
+    "key_patterns": [
+      "State management via Context API",
+      "Functional components with Hooks pattern",
+      "API calls wrapped in custom hooks"
+    ]
+  },
+  "relevant_files": [
+    {
+      "path": "src/components/Auth/LoginForm.tsx",
+      "relevance": "Existing login form component",
+      "priority": "high"
+    },
+    {
+      "path": "src/contexts/AuthContext.tsx",
+      "relevance": "Authentication state management context",
+      "priority": "high"
+    },
+    {
+      "path": "CLAUDE.md",
+      "relevance": "Project development standards",
+      "priority": "high"
+    }
+  ],
+  "integration_points": [
+    "Must integrate with existing AuthContext",
+    "Follow component organization pattern: src/components/[Feature]/",
+    "API calls should use src/hooks/useApi.ts wrapper"
+  ],
+  "constraints": [
+    "Maintain backward compatibility",
+    "Follow TypeScript strict mode",
+    "Use existing UI component library"
+  ]
+}
+```
+
+## Quality Checklist
+
+Before generating content package, verify:
+- [ ] Valid JSON format
+- [ ] All required fields complete
+- [ ] relevant_files contains minimum 3-10 files
+- [ ] project_summary accurately reflects architecture
+- [ ] integration_points clearly specify integration paths
+- [ ] keywords accurately extracted (3-8 keywords)
+- [ ] Content is concise, avoid redundancy (< 5KB total)
+
+## Memory Persistence
+
+- **Session Scope**: Content package valid for current session
+- **Subsequent References**: All subsequent agents/commands can access
+- **Reload Required**: New sessions need to re-execute `/memory:prepare`
+
+## Related Documentation
+
+- [Memory Commands](../claude/memory.md)
+- [Review Prompts](./review.md)
+- [CLI Invocation System](../../features/cli.md)

docs/commands/codex/review.md

@@ -0,0 +1,197 @@
+# Review Prompts
+
+## One-Liner
+
+**Review prompts are standardized templates for code review** — multi-dimensional code quality checks ensuring code meets best practices.
+
+## Review Dimensions
+
+| Dimension | Check Items | Severity |
+|-----------|-------------|----------|
+| **Correctness** | Logic errors, boundary conditions, type safety | Critical |
+| **Security** | Injection vulnerabilities, authentication, input validation | Critical |
+| **Performance** | Algorithm complexity, N+1 queries, caching opportunities | High |
+| **Maintainability** | SOLID principles, code duplication, naming conventions | Medium |
+| **Documentation** | Comment completeness, README updates | Low |
+
+## codex-review
+
+**Function**: Interactive code review using Codex CLI via ccw endpoint, supporting configurable review targets, models, and custom instructions.
+
+**Syntax**:
+```
+/cli:codex-review [--uncommitted|--base <branch>|--commit <sha>] [--model <model>] [--title <title>] [prompt]
+```
+
+**Parameters**:
+- `--uncommitted`: Review staged, unstaged, and untracked changes
+- `--base <branch>`: Compare changes with base branch
+- `--commit <sha>`: Review changes introduced by specific commit
+- `--model <model>`: Override default model (gpt-5.2, o3, gpt-4.1, o4-mini)
+- `--title <title>`: Optional commit title for review summary
+
+**Note**: Target flags and prompt are mutually exclusive (see constraints section)
+
+### Review Focus Selection
+
+| Focus | Template | Key Checks |
+|-------|----------|------------|
+| **Comprehensive Review** | Universal template | Correctness, style, bugs, documentation |
+| **Security Focus** | Security template | Injection, authentication, validation, exposure |
+| **Performance Focus** | Performance template | Complexity, memory, queries, caching |
+| **Code Quality** | Quality template | SOLID, duplication, naming, tests |
+
+### Prompt Templates
+
+#### Comprehensive Review Template
+
+```
+PURPOSE: Comprehensive code review to identify issues, improve quality, and ensure best practices; success = actionable feedback and clear priorities
+TASK: • Review code correctness and logic errors • Check coding standards and consistency • Identify potential bugs and edge cases • Evaluate documentation completeness
+MODE: review
+CONTEXT: {target description} | Memory: Project conventions from CLAUDE.md
+EXPECTED: Structured review report with: severity levels (Critical/High/Medium/Low), file:line references, specific improvement suggestions, priority rankings
+CONSTRAINTS: Focus on actionable feedback
+```
+
+#### Security Focus Template
+
+```
+PURPOSE: Security-focused code review to identify vulnerabilities and security risks; success = all security issues documented with fixes
+TASK: • Scan for injection vulnerabilities (SQL, XSS, command) • Check authentication and authorization logic • Evaluate input validation and sanitization • Identify sensitive data exposure risks
+MODE: review
+CONTEXT: {target description} | Memory: Security best practices, OWASP Top 10
+EXPECTED: Security report with: vulnerability classification, applicable CVE references, fix code snippets, risk severity matrix
+CONSTRAINTS: Security-first analysis | Flag all potential vulnerabilities
+```
+
+#### Performance Focus Template
+
+```
+PURPOSE: Performance-focused code review to identify bottlenecks and optimization opportunities; success = measurable improvement suggestions
+TASK: • Analyze algorithm complexity (Big-O) • Identify memory allocation issues • Check N+1 queries and blocking operations • Evaluate caching opportunities
+MODE: review
+CONTEXT: {target description} | Memory: Performance patterns and anti-patterns
+EXPECTED: Performance report with: complexity analysis, bottleneck identification, optimization suggestions with expected impact, benchmark recommendations
+CONSTRAINTS: Performance optimization focus
+```
+
+#### Code Quality Template
+
+```
+PURPOSE: Code quality review to improve maintainability and readability; success = cleaner, more maintainable code
+TASK: • Evaluate SOLID principles compliance • Identify code duplication and abstraction opportunities • Review naming conventions and clarity • Evaluate test coverage impact
+MODE: review
+CONTEXT: {target description} | Memory: Project coding standards
+EXPECTED: Quality report with: principle violations, refactoring suggestions, naming improvements, maintainability score
+CONSTRAINTS: Code quality and maintainability focus
+```
+
+### Usage Examples
+
+#### Direct Execution (No Interaction)
+
+```bash
+# Review uncommitted changes with default settings
+/cli:codex-review --uncommitted
+
+# Compare with main branch
+/cli:codex-review --base main
+
+# Review specific commit
+/cli:codex-review --commit abc123
+
+# Use custom model
+/cli:codex-review --uncommitted --model o3
+
+# Security focus review
+/cli:codex-review --uncommitted security
+
+# Full options
+/cli:codex-review --base main --model o3 --title "Authentication feature" security
+```
+
+#### Interactive Mode
+
+```bash
+# Start interactive selection (guided flow)
+/cli:codex-review
+```
+
+### Constraints and Validation
+
+**Important**: Target flags and prompt are mutually exclusive
+
+Codex CLI has a constraint that target flags (`--uncommitted`, `--base`, `--commit`) cannot be used with the `[PROMPT]` positional parameter:
+
+```
+error: the argument '--uncommitted' cannot be used with '[PROMPT]'
+error: the argument '--base <BRANCH>' cannot be used with '[PROMPT]'
+error: the argument '--commit <SHA>' cannot be used with '[PROMPT]'
+```
+
+**Valid Combinations**:
+
+| Command | Result |
+|---------|--------|
+| `codex review "focus on security"` | ✓ Custom prompt, reviews uncommitted (default) |
+| `codex review --uncommitted` | ✓ No prompt, uses default review |
+| `codex review --base main` | ✓ No prompt, uses default review |
+| `codex review --commit abc123` | ✓ No prompt, uses default review |
+| `codex review --uncommitted "prompt"` | ✗ Invalid - mutually exclusive |
+| `codex review --base main "prompt"` | ✗ Invalid - mutually exclusive |
+| `codex review --commit abc123 "prompt"` | ✗ Invalid - mutually exclusive |
+
+**Valid Examples**:
+```bash
+# ✓ Valid: Prompt only (defaults to reviewing uncommitted)
+ccw cli -p "focus on security" --tool codex --mode review
+
+# ✓ Valid: Target flags only (no prompt)
+ccw cli --tool codex --mode review --uncommitted
+ccw cli --tool codex --mode review --base main
+ccw cli --tool codex --mode review --commit abc123
+
+# ✗ Invalid: Target flags with prompt (will fail)
+ccw cli -p "review this" --tool codex --mode review --uncommitted
+```
+
+## Focus Area Mapping
+
+| User Selection | Prompt Focus | Key Checks |
+|----------------|--------------|------------|
+| Comprehensive Review | Comprehensive | Correctness, style, bugs, documentation |
+| Security Focus | Security-first | Injection, authentication, validation, exposure |
+| Performance Focus | Optimization | Complexity, memory, queries, caching |
+| Code Quality | Maintainability | SOLID, duplication, naming, tests |
+
+## Error Handling
+
+### No Changes to Review
+
+```
+No changes found for review target. Suggestions:
+- For --uncommitted: Make some code changes first
+- For --base: Ensure branch exists and has diverged
+- For --commit: Verify commit SHA exists
+```
+
+### Invalid Branch
+
+```bash
+# Show available branches
+git branch -a --list | head -20
+```
+
+### Invalid Commit
+
+```bash
+# Show recent commits
+git log --oneline -10
+```
+
+## Related Documentation
+
+- [Prep Prompts](./prep.md)
+- [CLI Tool Commands](../claude/cli.md)
+- [Code Review](../../features/)

docs/commands/index.md

@@ -0,0 +1,34 @@
+# Commands Overview
+
+## One-Liner
+
+**Commands is the command system of Claude_dms3** — comprising Claude Code Commands and Codex Prompts, providing a complete command-line toolchain from specification to implementation to testing to review.
+
+## Command Categories
+
+| Category | Description | Path |
+|----------|-------------|------|
+| **Claude Commands** | Claude Code extension commands | `/commands/claude/` |
+| **Codex Prompts** | Codex CLI prompts | `/commands/codex/` |
+
+## Quick Navigation
+
+### Claude Commands
+
+- [Core Orchestration](/commands/claude/) - ccw, ccw-coordinator
+- [Workflow](/commands/claude/workflow) - workflow series commands
+- [Session Management](/commands/claude/session) - session series commands
+- [Issue](/commands/claude/issue) - issue series commands
+- [Memory](/commands/claude/memory) - memory series commands
+- [CLI](/commands/claude/cli) - cli series commands
+- [UI Design](/commands/claude/ui-design) - ui-design series commands
+
+### Codex Prompts
+
+- [Prep](/commands/codex/prep) - prep-cycle, prep-plan
+- [Review](/commands/codex/review) - review prompts
+
+## Related Documentation
+
+- [Skills](/skills/) - Skill system
+- [Features](/features/spec) - Core features

docs/features/api-settings.md

@@ -0,0 +1,101 @@
+# API Settings
+
+## One-Liner
+
+**API Settings manages AI model endpoint configuration** — Centralizes API keys, base URLs, and model selection for all supported AI backends.
+
+---
+
+## Configuration Locations
+
+| Location | Scope | Priority |
+|----------|-------|----------|
+| `~/.claude/cli-tools.json` | Global | Base |
+| `.claude/settings.json` | Project | Override |
+| `.claude/settings.local.json` | Local | Highest |
+
+---
+
+## Supported Backends
+
+| Backend | Type | Models |
+|---------|------|--------|
+| **Gemini** | Builtin | gemini-2.5-flash, gemini-2.5-pro |
+| **Qwen** | Builtin | coder-model |
+| **Codex** | Builtin | gpt-4o, gpt-4o-mini |
+| **Claude** | Builtin | claude-sonnet, claude-haiku |
+| **OpenCode** | Builtin | opencode/glm-4.7-free |
+
+---
+
+## Configuration Example
+
+```json
+// ~/.claude/cli-tools.json
+{
+  "version": "3.3.0",
+  "tools": {
+    "gemini": {
+      "enabled": true,
+      "primaryModel": "gemini-2.5-flash",
+      "secondaryModel": "gemini-2.5-flash",
+      "tags": ["analysis", "debug"],
+      "type": "builtin"
+    },
+    "codex": {
+      "enabled": true,
+      "primaryModel": "gpt-4o",
+      "tags": [],
+      "type": "builtin"
+    }
+  }
+}
+```
+
+---
+
+## Environment Variables
+
+```bash
+# API Keys
+LITELLM_API_KEY=your-api-key
+LITELLM_API_BASE=https://api.example.com/v1
+LITELLM_MODEL=gpt-4o-mini
+
+# Reranker (optional)
+RERANKER_API_KEY=your-reranker-key
+RERANKER_API_BASE=https://api.siliconflow.cn
+RERANKER_PROVIDER=siliconflow
+RERANKER_MODEL=BAAI/bge-reranker-v2-m3
+```
+
+---
+
+## Model Selection
+
+### Via CLI Flag
+
+```bash
+ccw cli -p "Analyze code" --tool gemini --model gemini-2.5-pro
+```
+
+### Via Config
+
+```json
+{
+  "tools": {
+    "gemini": {
+      "primaryModel": "gemini-2.5-flash",
+      "secondaryModel": "gemini-2.5-pro"
+    }
+  }
+}
+```
+
+---
+
+## Related Links
+
+- [CLI Call](/features/cli) - Command invocation
+- [System Settings](/features/system-settings) - System configuration
+- [CodexLens](/features/codexlens) - Code indexing

docs/features/cli.md

@@ -0,0 +1,104 @@
+# CLI Call System
+
+## One-Liner
+
+**The CLI Call System is a unified AI model invocation framework** — Provides a consistent interface for calling multiple AI models (Gemini, Qwen, Codex, Claude) with standardized prompts, modes, and templates.
+
+---
+
+## Pain Points Solved
+
+| Pain Point | Current State | CLI Call Solution |
+|------------|---------------|-------------------|
+| **Single model limitation** | Can only use one AI model | Multi-model collaboration |
+| **Inconsistent prompts** | Different prompt formats per model | Unified prompt template |
+| **No mode control** | AI can modify files unexpectedly | analysis/write/review modes |
+| **No templates** | Reinvent prompts each time | 30+ pre-built templates |
+
+---
+
+## vs Traditional Methods
+
+| Dimension | Direct API | Individual CLIs | **CCW CLI** |
+|-----------|------------|-----------------|-------------|
+| Multi-model | Manual switch | Multiple tools | **Unified interface** |
+| Prompt format | Per-model | Per-tool | **Standardized template** |
+| Permission | Unclear | Unclear | **Explicit modes** |
+| Templates | None | None | **30+ templates** |
+
+---
+
+## Core Concepts
+
+| Concept | Description | Usage |
+|---------|-------------|-------|
+| **Tool** | AI model backend | `--tool gemini/qwen/codex/claude` |
+| **Mode** | Permission level | `analysis/write/review` |
+| **Template** | Pre-built prompt | `--rule analysis-review-code` |
+| **Session** | Conversation continuity | `--resume <id>` |
+
+---
+
+## Usage
+
+### Basic Command
+
+```bash
+ccw cli -p "Analyze authentication flow" --tool gemini --mode analysis
+```
+
+### With Template
+
+```bash
+ccw cli -p "PURPOSE: Security audit
+TASK: • Check SQL injection • Verify CSRF tokens
+MODE: analysis
+CONTEXT: @src/auth/**/*
+EXPECTED: Report with severity levels
+CONSTRAINTS: Focus on authentication" --tool gemini --mode analysis --rule analysis-assess-security-risks
+```
+
+### Session Resume
+
+```bash
+ccw cli -p "Continue analysis" --resume
+```
+
+---
+
+## Configuration
+
+```json
+// ~/.claude/cli-tools.json
+{
+  "tools": {
+    "gemini": {
+      "enabled": true,
+      "primaryModel": "gemini-2.5-flash",
+      "tags": ["analysis", "debug"]
+    },
+    "qwen": {
+      "enabled": true,
+      "primaryModel": "coder-model"
+    }
+  }
+}
+```
+
+---
+
+## Available Templates
+
+| Category | Templates |
+|----------|-----------|
+| Analysis | `analysis-review-code`, `analysis-diagnose-bug`, `analysis-assess-security` |
+| Planning | `planning-plan-architecture`, `planning-breakdown-task` |
+| Development | `development-implement-feature`, `development-refactor-codebase` |
+
+---
+
+## Related Links
+
+- [Spec System](/features/spec) - Constraint injection
+- [Memory System](/features/memory) - Persistent context
+- [CodexLens](/features/codexlens) - Code indexing

docs/features/codexlens.md

@@ -0,0 +1,115 @@
+# CodexLens Code Indexing
+
+## One-Liner
+
+**CodexLens is a semantic code search engine** — Based on vector databases and LSP integration, it enables AI to understand code semantics rather than just keyword matching.
+
+---
+
+## Pain Points Solved
+
+| Pain Point | Current State | CodexLens Solution |
+|------------|---------------|-------------------|
+| **Imprecise search** | Keywords can't find semantically related code | Semantic vector search |
+| **No context** | Search results lack call chain context | LSP integration provides reference chains |
+| **No understanding** | AI doesn't understand code relationships | Static analysis + semantic indexing |
+| **Slow navigation** | Manual file traversal | Instant semantic navigation |
+
+---
+
+## vs Traditional Methods
+
+| Dimension | Text Search | IDE Search | **CodexLens** |
+|-----------|-------------|------------|---------------|
+| Search type | Keyword | Keyword + symbol | **Semantic vector** |
+| Context | None | File-level | **Call chain + imports** |
+| AI-ready | No | No | **Direct AI consumption** |
+| Multi-file | Poor | Good | **Excellent** |
+
+---
+
+## Core Concepts
+
+| Concept | Description | Location |
+|---------|-------------|----------|
+| **Index** | Vector representation of code | `.codex-lens/index/` |
+| **Chunk** | Code segment for embedding | Configurable size |
+| **Retrieval** | Hybrid search (vector + keyword) | HybridSearch engine |
+| **LSP** | Language Server Protocol integration | Built-in LSP client |
+
+---
+
+## Usage
+
+### Indexing Project
+
+```bash
+ccw index
+ccw index --watch  # Continuous indexing
+```
+
+### Searching
+
+```bash
+ccw search "authentication logic"
+ccw search "where is user validation" --top 10
+```
+
+### Via MCP Tool
+
+```typescript
+// ACE semantic search
+mcp__ace-tool__search_context({
+  project_root_path: "/path/to/project",
+  query: "authentication logic"
+})
+```
+
+---
+
+## Configuration
+
+```json
+// ~/.codexlens/settings.json
+{
+  "embedding": {
+    "backend": "litellm",
+    "model": "Qwen/Qwen3-Embedding-8B",
+    "use_gpu": false
+  },
+  "indexing": {
+    "static_graph_enabled": true,
+    "chunk_size": 512
+  }
+}
+```
+
+---
+
+## Architecture
+
+```
+┌─────────────────────────────────────────┐
+│              CodexLens                   │
+├─────────────────────────────────────────┤
+│  ┌──────────┐  ┌──────────┐  ┌────────┐ │
+│  │  Parsers │  │ Chunker  │  │  LSP   │ │
+│  │(TS/Py/..)│  │          │  │ Client │ │
+│  └────┬─────┘  └────┬─────┘  └───┬────┘ │
+│       │             │            │       │
+│       └─────────────┼────────────┘       │
+│                     │                    │
+│              ┌──────┴──────┐             │
+│              │   Hybrid    │             │
+│              │   Search    │             │
+│              └─────────────┘             │
+└─────────────────────────────────────────┘
+```
+
+---
+
+## Related Links
+
+- [CLI Call](/features/cli) - AI model invocation
+- [Memory System](/features/memory) - Persistent context
+- [MCP Tools](/mcp/tools) - MCP integration

docs/features/dashboard.md

@@ -0,0 +1,79 @@
+# Dashboard Panel
+
+## One-Liner
+
+**The Dashboard is a VS Code webview-based management interface** — Provides visual access to project configuration, specs, memory, and settings through an intuitive GUI.
+
+---
+
+## Pain Points Solved
+
+| Pain Point | Current State | Dashboard Solution |
+|------------|---------------|-------------------|
+| **Config complexity** | JSON files hard to edit | Visual form-based editing |
+| **No overview** | Can't see project state at a glance | Unified dashboard view |
+| **Scattered settings** | Settings in multiple files | Centralized management |
+| **No visual feedback** | CLI only, no UI | Interactive webview |
+
+---
+
+## Core Features
+
+| Feature | Description |
+|---------|-------------|
+| **Project Overview** | Tech stack, dependencies, status |
+| **Spec Manager** | View and edit specification files |
+| **Memory Viewer** | Browse persistent memories |
+| **API Settings** | Configure AI model endpoints |
+| **System Settings** | Global and project settings |
+
+---
+
+## Access
+
+Open via VS Code command palette:
+```
+CCW: Open Dashboard
+```
+
+Or via CLI:
+```bash
+ccw view
+```
+
+---
+
+## Dashboard Sections
+
+### 1. Project Overview
+- Technology stack detection
+- Dependency status
+- Workflow session status
+
+### 2. Spec Manager
+- List all specs
+- Edit spec content
+- Enable/disable specs
+
+### 3. Memory Viewer
+- Browse memory entries
+- Search memories
+- Export/import
+
+### 4. API Settings
+- Configure API keys
+- Set model endpoints
+- Manage rate limits
+
+### 5. System Settings
+- Global configuration
+- Project overrides
+- Hook management
+
+---
+
+## Related Links
+
+- [API Settings](/features/api-settings) - API configuration
+- [System Settings](/features/system-settings) - System configuration
+- [Spec System](/features/spec) - Specification management

docs/features/memory.md

@@ -0,0 +1,80 @@
+# Memory System
+
+## One-Liner
+
+**The Memory System is a cross-session knowledge persistence mechanism** — Stores project context, decisions, and learnings so AI remembers across sessions without re-explanation.
+
+---
+
+## Pain Points Solved
+
+| Pain Point | Current State | Memory System Solution |
+|------------|---------------|------------------------|
+| **Cross-session amnesia** | New session requires re-explaining project | Persistent memory across sessions |
+| **Lost decisions** | Architecture decisions forgotten | Decision log persists |
+| **Repeated explanations** | Same context explained multiple times | Memory auto-injection |
+| **Knowledge silos** | Each developer maintains own context | Shared team memory |
+
+---
+
+## vs Traditional Methods
+
+| Dimension | CLAUDE.md | Notes | **Memory System** |
+|-----------|-----------|-------|-------------------|
+| Persistence | Static file | Manual | **Auto-extracted from sessions** |
+| Search | Text search | Folder search | **Semantic vector search** |
+| Updates | Manual edit | Manual note | **Auto-capture from conversations** |
+| Sharing | Git | Manual | **Auto-sync via workflow** |
+
+---
+
+## Core Concepts
+
+| Concept | Description | Location |
+|---------|-------------|----------|
+| **Core Memory** | Persistent project knowledge | `~/.claude/memory/` |
+| **Session Memory** | Current session context | `.workflow/.memory/` |
+| **Memory Entry** | Individual knowledge item | JSONL format |
+| **Memory Index** | Searchable index | Embedding-based |
+
+---
+
+## Usage
+
+### Viewing Memory
+
+```bash
+ccw memory list
+ccw memory search "authentication"
+```
+
+### Memory Categories
+
+- **Project Context**: Architecture, tech stack, patterns
+- **Decisions**: ADRs, design choices
+- **Learnings**: What worked, what didn't
+- **Conventions**: Coding standards, naming
+
+---
+
+## Configuration
+
+```json
+// ~/.claude/cli-settings.json
+{
+  "memory": {
+    "enabled": true,
+    "maxEntries": 1000,
+    "autoCapture": true,
+    "embeddingModel": "text-embedding-3-small"
+  }
+}
+```
+
+---
+
+## Related Links
+
+- [Spec System](/features/spec) - Constraint injection
+- [CLI Call](/features/cli) - Command line invocation
+- [CodexLens](/features/codexlens) - Code indexing

docs/features/spec.md

@@ -0,0 +1,96 @@
+# Spec System
+
+## One-Liner
+
+**The Spec System is an automatic constraint injection mechanism** — Through specification documents defined in YAML frontmatter, relevant constraints are automatically loaded at the start of AI sessions, ensuring AI follows project coding standards and architectural requirements.
+
+---
+
+## Pain Points Solved
+
+| Pain Point | Current State | Spec System Solution |
+|------------|---------------|---------------------|
+| **AI ignores standards** | CLAUDE.md written but AI ignores it after 5 turns | Hook auto-injection, every session carries specs |
+| **Standards scattered** | Coding conventions in different places, hard to maintain | Unified in `.workflow/specs/*.md` |
+| **Context loss** | New session requires re-explaining constraints | Spec auto-loads based on task context |
+| **Inconsistent code** | Different developers write different styles | Shared Spec ensures consistency |
+
+---
+
+## vs Traditional Methods
+
+| Dimension | CLAUDE.md | `.cursorrules` | **Spec System** |
+|-----------|-----------|----------------|-----------------|
+| Injection | Auto-load but easily truncated | Manual load | **Hook auto-injection, task-precise loading** |
+| Granularity | One large file | One large file | **Per-module files, combined by task** |
+| Cross-session memory | None | None | **Workflow journal persistence** |
+| Team sharing | Single person | Single person | **Git versioned Spec library** |
+
+---
+
+## Core Concepts
+
+| Concept | Description | Location |
+|---------|-------------|----------|
+| **Spec File** | Markdown document with YAML frontmatter | `.workflow/specs/*.md` |
+| **Hook** | Script that auto-injects specs into AI context | `.claude/hooks/` |
+| **Spec Index** | Registry of all available specs | `.workflow/specs/index.yaml` |
+| **Spec Selector** | Logic that chooses relevant specs for a task | Built into CCW |
+
+---
+
+## Usage
+
+### Creating a Spec
+
+```markdown
+---
+name: coding-standards
+description: Project coding standards
+triggers:
+  - pattern: "**/*.ts"
+  - command: "/implement"
+  - skill: "code-developer"
+applyTo:
+  - "src/**"
+priority: high
+---
+
+# Coding Standards
+
+## TypeScript Guidelines
+- Use strict mode
+- Prefer interfaces over types
+- ...
+```
+
+### Spec Loading
+
+Specs are automatically loaded based on:
+1. File patterns being edited
+2. Commands being executed
+3. Skills being invoked
+
+---
+
+## Configuration
+
+```yaml
+# .workflow/specs/index.yaml
+specs:
+  - name: coding-standards
+    path: ./coding-standards.md
+    enabled: true
+    
+  - name: api-conventions
+    path: ./api-conventions.md
+    enabled: true
+```
+
+---
+
+## Related Links
+
+- [Memory System](/features/memory) - Persistent context
+- [CLI Call](/features/cli) - Command line invocation
+- [Dashboard](/features/dashboard) - Visual management

docs/features/system-settings.md

@@ -0,0 +1,131 @@
+# System Settings
+
+## One-Liner
+
+**System Settings manages global and project-level configuration** — Controls hooks, agents, skills, and core system behavior.
+
+---
+
+## Configuration Files
+
+| File | Scope | Purpose |
+|------|-------|---------|
+| `~/.claude/CLAUDE.md` | Global | Global instructions |
+| `.claude/CLAUDE.md` | Project | Project instructions |
+| `~/.claude/cli-tools.json` | Global | CLI tool config |
+| `.claude/settings.json` | Project | Project settings |
+| `.claude/settings.local.json` | Local | Local overrides |
+
+---
+
+## Settings Schema
+
+```json
+{
+  "permissions": {
+    "allow": ["Bash(npm install)", "Bash(git status)"],
+    "deny": ["Bash(rm -rf)"]
+  },
+  "env": {
+    "ANTHROPIC_API_KEY": "your-key"
+  },
+  "enableAll": false,
+  "autoCheck": true
+}
+```
+
+---
+
+## Key Settings
+
+### Permissions
+
+```json
+{
+  "permissions": {
+    "allow": [
+      "Bash(npm run*)",
+      "Read(**)",
+      "Edit(**/*.ts)"
+    ],
+    "deny": [
+      "Bash(rm -rf /*)"
+    ]
+  }
+}
+```
+
+### Hooks
+
+```json
+{
+  "hooks": {
+    "PreToolUse": [
+      {
+        "matcher": "Bash",
+        "hooks": [".claude/hooks/pre-bash.sh"]
+      }
+    ]
+  }
+}
+```
+
+### MCP Servers
+
+```json
+{
+  "mcpServers": {
+    "ccw-tools": {
+      "command": "node",
+      "args": ["./mcp-server/dist/index.js"]
+    }
+  }
+}
+```
+
+---
+
+## Hook Configuration
+
+Hooks are scripts that run at specific events:
+
+| Event | When | Use Case |
+|-------|------|----------|
+| `PreToolUse` | Before tool execution | Validation, logging |
+| `PostToolUse` | After tool execution | Cleanup, notifications |
+| `Notification` | On notifications | Custom handlers |
+| `Stop` | On session end | Cleanup |
+
+### Hook Example
+
+```bash
+#!/bin/bash
+# .claude/hooks/pre-bash.sh
+echo "Executing: $1" >> ~/.claude/bash.log
+```
+
+---
+
+## Agent Configuration
+
+```json
+// .claude/agents/my-agent.md
+---
+description: Custom analysis agent
+model: claude-sonnet
+tools:
+  - Read
+  - Grep
+---
+
+# Agent Instructions
+...
+```
+
+---
+
+## Related Links
+
+- [API Settings](/features/api-settings) - API configuration
+- [CLI Call](/features/cli) - Command invocation
+- [Dashboard](/features/dashboard) - Visual management

docs/guide/ch01-what-is-claude-dms3.md

@@ -0,0 +1,87 @@
+# What is Claude_dms3
+
+## One-Line Positioning
+
+**Claude_dms3 is an AI-powered development workbench for VS Code** — Through semantic code indexing, multi-model CLI invocation, and team collaboration systems, it enables AI to deeply understand your project and generate high-quality code according to specifications.
+
+> AI capabilities bloom like vines — Claude_dms3 is the trellis that guides AI along your project's architecture, coding standards, and team workflows.
+
+---
+
+## 1.1 Pain Points Solved
+
+| Pain Point | Current State | Claude_dms3 Solution |
+|------------|---------------|---------------------|
+| **AI doesn't understand the project** | Every new session requires re-explaining project background, tech stack, and coding standards | Memory system persists project context, AI remembers project knowledge across sessions |
+| **Difficult code search** | Keyword search can't find semantically related code, don't know where functions are called | CodexLens semantic indexing, supports natural language search and call chain tracing |
+| **Single model limitation** | Can only call one AI model, different models excel in different scenarios | CCW unified invocation framework, supports multi-model collaboration (Gemini, Qwen, Codex, Claude) |
+| **Chaotic collaboration process** | Team members work independently, inconsistent code styles, standards hard to implement | Team workflow system (PlanEx, IterDev, Lifecycle) ensures standard execution |
+| **Standards hard to implement** | CLAUDE.md written but AI doesn't follow, project constraints ignored | Spec + Hook auto-injection, AI forced to follow project standards |
+
+---
+
+## 1.2 vs Traditional Methods
+
+| Dimension | Traditional AI Assistant | **Claude_dms3** |
+|-----------|-------------------------|-----------------|
+| **Code Search** | Text keyword search | **Semantic vector search + LSP call chain** |
+| **AI Invocation** | Single model fixed call | **Multi-model collaboration, optimal model per task** |
+| **Project Memory** | Re-explain each session | **Cross-session persistent Memory** |
+| **Standard Execution** | Relies on Prompt reminders | **Spec + Hook auto-injection** |
+| **Team Collaboration** | Each person for themselves | **Structured workflow system** |
+| **Code Quality** | Depends on AI capability | **Multi-dimensional review + auto-fix cycle** |
+
+---
+
+## 1.3 Core Concepts Overview
+
+| Concept | Description | Location/Command |
+|---------|-------------|------------------|
+| **CodexLens** | Semantic code indexing and search engine | `ccw search` |
+| **CCW** | Unified CLI tool invocation framework | `ccw cli` |
+| **Memory** | Cross-session knowledge persistence | `ccw memory` |
+| **Spec** | Project specification and constraint system | `.workflow/specs/` |
+| **Hook** | Auto-triggered context injection scripts | `.claude/hooks/` |
+| **Agent** | Specialized AI subprocess for specific roles | `.claude/agents/` |
+| **Skill** | Reusable AI capability modules | `.claude/skills/` |
+| **Workflow** | Multi-phase development orchestration | `/workflow:*` |
+
+---
+
+## 1.4 Architecture Overview
+
+```
+┌─────────────────────────────────────────────────────────────┐
+│                    Claude_dms3 Architecture                  │
+├─────────────────────────────────────────────────────────────┤
+│                                                              │
+│  ┌─────────────┐  ┌─────────────┐  ┌─────────────┐          │
+│  │  CodexLens  │  │     CCW     │  │   Memory    │          │
+│  │ (Semantic   │  │  (CLI Call  │  │ (Persistent │          │
+│  │   Index)    │  │  Framework) │  │   Context)  │          │
+│  └──────┬──────┘  └──────┬──────┘  └──────┬──────┘          │
+│         │                │                │                  │
+│         └────────────────┼────────────────┘                  │
+│                          │                                   │
+│                    ┌─────┴─────┐                             │
+│                    │   Spec    │                             │
+│                    │  System   │                             │
+│                    └─────┬─────┘                             │
+│                          │                                   │
+│         ┌────────────────┼────────────────┐                  │
+│         │                │                │                  │
+│    ┌────┴────┐     ┌─────┴─────┐    ┌────┴────┐             │
+│    │  Hooks  │     │  Skills   │    │ Agents  │             │
+│    │(Inject) │     │(Reusable) │    │(Roles)  │             │
+│    └─────────┘     └───────────┘    └─────────┘             │
+│                                                              │
+└─────────────────────────────────────────────────────────────┘
+```
+
+---
+
+## Next Steps
+
+- [Getting Started](/guide/ch02-getting-started) - Install and configure
+- [Core Concepts](/guide/ch03-core-concepts) - Understand the fundamentals
+- [Workflow Basics](/guide/ch04-workflow-basics) - Start your first workflow

docs/guide/ch02-getting-started.md

@@ -0,0 +1,295 @@
+# Getting Started
+
+## One-Line Positioning
+
+**Getting Started is a 5-minute quick-start guide** — Installation, first command, first workflow, quickly experience Claude_dms3's core features.
+
+---
+
+## 2.1 Installation
+
+### 2.1.1 Prerequisites
+
+| Requirement | Version | Description |
+| --- | --- | --- |
+| **Node.js** | 18+ | Required for CCW modules |
+| **Python** | 3.10+ | Required for CodexLens modules |
+| **VS Code** | Latest | Extension runtime environment |
+| **Git** | Latest | Version control |
+
+### 2.1.2 Clone Project
+
+```bash
+# Clone repository
+git clone https://github.com/your-repo/claude-dms3.git
+cd claude-dms3
+
+# Install dependencies
+npm install
+```
+
+### 2.1.3 Configure API Keys
+
+Configure API Keys in `~/.claude/settings.json`:
+
+```json
+{
+  "openai": {
+    "apiKey": "sk-xxx"
+  },
+  "anthropic": {
+    "apiKey": "sk-ant-xxx"
+  },
+  "google": {
+    "apiKey": "AIza-xxx"
+  }
+}
+```
+
+::: tip Tip
+API Keys can also be configured at the project level in `.claude/settings.json`. Project-level configuration takes priority over global configuration.
+:::
+
+---
+
+## 2.2 Initialize Project
+
+### 2.2.1 Start Workflow Session
+
+Open your project in VS Code, then run:
+
+```
+/workflow:session:start
+```
+
+This creates a new workflow session. All subsequent operations will be performed within this session context.
+
+### 2.2.2 Initialize Project Specs
+
+```
+/workflow:init
+```
+
+This creates the `project-tech.json` file, recording your project's technology stack information.
+
+### 2.2.3 Populate Project Specs
+
+```
+/workflow:init-guidelines
+```
+
+Interactively populate project specifications, including coding style, architectural decisions, and other information.
+
+---
+
+## 2.3 First Command
+
+### 2.3.1 Code Analysis
+
+Use CCW CLI tool to analyze code:
+
+```bash
+ccw cli -p "Analyze the code structure and design patterns of this file" --tool gemini --mode analysis
+```
+
+**Parameter Description**:
+- `-p`: Prompt (task description)
+- `--tool gemini`: Use Gemini model
+- `--mode analysis`: Analysis mode (read-only, no file modifications)
+
+### 2.3.2 Code Generation
+
+Use CCW CLI tool to generate code:
+
+```bash
+ccw cli -p "Create a React component implementing user login form" --tool qwen --mode write
+```
+
+**Parameter Description**:
+- `--mode write`: Write mode (can create/modify files)
+
+::: danger Warning
+`--mode write` will modify files. Ensure your code is committed or backed up.
+:::
+
+---
+
+## 2.4 First Workflow
+
+### 2.4.1 Start Planning Workflow
+
+```
+/workflow:plan
+```
+
+This launches the PlanEx workflow, including the following steps:
+
+1. **Analyze Requirements** - Understand user intent
+2. **Explore Code** - Search related code and patterns
+3. **Generate Plan** - Create structured task list
+4. **Execute Tasks** - Execute development according to plan
+
+### 2.4.2 Brainstorming
+
+```
+/brainstorm
+```
+
+Multi-perspective brainstorming for diverse viewpoints:
+
+| Perspective | Role | Focus |
+| --- | --- | --- |
+| Product | Product Manager | Market fit, user value |
+| Technical | Tech Lead | Feasibility, technical debt |
+| Quality | QA Lead | Completeness, testability |
+| Risk | Risk Analyst | Risk identification, dependencies |
+
+---
+
+## 2.5 Using Memory
+
+### 2.5.1 View Project Memory
+
+```bash
+ccw memory list
+```
+
+Display all project memories, including learnings, decisions, conventions, and issues.
+
+### 2.5.2 Search Related Memory
+
+```bash
+ccw memory search "authentication"
+```
+
+Semantic search for memories related to "authentication".
+
+### 2.5.3 Add Memory
+
+```
+/memory-capture
+```
+
+Interactively capture important knowledge points from the current session.
+
+---
+
+## 2.6 Code Search
+
+### 2.6.1 Semantic Search
+
+Use CodexLens search in VS Code:
+
+```bash
+# Search via CodexLens MCP endpoint
+ccw search "user login logic"
+```
+
+### 2.6.2 Call Chain Tracing
+
+Search function definitions and all call locations:
+
+```bash
+ccw search --trace "authenticateUser"
+```
+
+---
+
+## 2.7 Dashboard Panel
+
+### 2.7.1 Open Dashboard
+
+Run in VS Code:
+
+```
+ccw-dashboard.open
+```
+
+Or use Command Palette (Ctrl+Shift+P) and search "CCW Dashboard".
+
+### 2.7.2 Panel Features
+
+| Feature | Description |
+| --- | --- |
+| **Tech Stack** | Display frameworks and libraries used |
+| **Specs Docs** | Quick view of project specifications |
+| **Memory** | Browse and search project memory |
+| **Code Search** | Integrated CodexLens semantic search |
+
+---
+
+## 2.8 FAQ
+
+### 2.8.1 API Key Configuration
+
+**Q: Where to configure API Keys?**
+
+A: Can be configured in two locations:
+- Global configuration: `~/.claude/settings.json`
+- Project configuration: `.claude/settings.json`
+
+Project configuration takes priority over global configuration.
+
+### 2.8.2 Model Selection
+
+**Q: How to choose the right model?**
+
+A: Select based on task type:
+- Code analysis, architecture design → Gemini
+- General development → Qwen
+- Code review → Codex (GPT)
+- Long text understanding → Claude
+
+### 2.8.3 Workflow Selection
+
+**Q: When to use which workflow?**
+
+A: Select based on task objective:
+- New feature development → `/workflow:plan`
+- Problem diagnosis → `/debug-with-file`
+- Code review → `/review-code`
+- Refactoring planning → `/refactor-cycle`
+- UI development → `/workflow:ui-design`
+
+---
+
+## 2.9 Quick Reference
+
+### Installation Steps
+
+```bash
+# 1. Clone project
+git clone https://github.com/your-repo/claude-dms3.git
+cd claude-dms3
+
+# 2. Install dependencies
+npm install
+
+# 3. Configure API Keys
+# Edit ~/.claude/settings.json
+
+# 4. Start workflow session
+/workflow:session:start
+
+# 5. Initialize project
+/workflow:init
+```
+
+### Common Commands
+
+| Command | Function |
+| --- | --- |
+| `/workflow:session:start` | Start session |
+| `/workflow:plan` | Planning workflow |
+| `/brainstorm` | Brainstorming |
+| `/review-code` | Code review |
+| `ccw memory list` | View Memory |
+| `ccw cli -p "..."` | CLI invocation |
+
+---
+
+## Next Steps
+
+- [Core Concepts](ch03-core-concepts.md) — Deep dive into Commands, Skills, Prompts
+- [Workflow Basics](ch04-workflow-basics.md) — Learn various workflow commands
+- [Advanced Tips](ch05-advanced-tips.md) — CLI toolchain, multi-model collaboration, memory management optimization

docs/guide/ch03-core-concepts.md

@@ -0,0 +1,264 @@
+# Core Concepts
+
+## One-Line Positioning
+
+**Core Concepts are the foundation for understanding Claude_dms3** — Three-layer abstraction of Commands, Skills, Prompts, Workflow session management, and team collaboration patterns.
+
+---
+
+## 3.1 Three-Layer Abstraction
+
+Claude_dms3's command system is divided into three layers of abstraction:
+
+### 3.1.1 Commands - Built-in Commands
+
+**Commands are the atomic operations of Claude_dms3** — Predefined reusable commands that complete specific tasks.
+
+| Category | Count | Description |
+| --- | --- | --- |
+| **Core Orchestration** | 2 | ccw, ccw-coordinator |
+| **CLI Tools** | 2 | cli-init, codex-review |
+| **Issue Workflow** | 8 | discover, plan, execute, queue, etc. |
+| **Memory** | 2 | prepare, style-skill-memory |
+| **Workflow Session** | 6 | start, resume, list, complete, etc. |
+| **Workflow Analysis** | 10+ | analyze, brainstorm, debug, refactor, etc. |
+| **Workflow UI Design** | 9 | generate, import-from-code, style-extract, etc. |
+
+::: tip Tip
+Commands are defined in the `.claude/commands/` directory. Each command is a Markdown file.
+:::
+
+### 3.1.2 Skills - Composite Skills
+
+**Skills are combined encapsulations of Commands** — Reusable skills for specific scenarios, containing multiple steps and best practices.
+
+| Skill | Function | Trigger |
+| --- | --- | --- |
+| **brainstorm** | Multi-perspective brainstorming | `/brainstorm` |
+| **ccw-help** | CCW command help | `/ccw-help` |
+| **command-generator** | Generate Claude commands | `/command-generator` |
+| **issue-manage** | Issue management | `/issue-manage` |
+| **memory-capture** | Memory compression and capture | `/memory-capture` |
+| **memory-manage** | Memory updates | `/memory-manage` |
+| **review-code** | Multi-dimensional code review | `/review-code` |
+| **review-cycle** | Code review and fix cycle | `/review-cycle` |
+| **skill-generator** | Generate Claude skills | `/skill-generator` |
+| **skill-tuning** | Skill diagnosis and tuning | `/skill-tuning` |
+
+::: tip Tip
+Skills are defined in the `.claude/skills/` directory, containing SKILL.md specification files and reference documentation.
+:::
+
+### 3.1.3 Prompts - Codex Prompts
+
+**Prompts are instruction templates for the Codex model** — Prompt templates optimized for the Codex (GPT) model.
+
+| Prompt | Function |
+| --- | --- |
+| **prep-cycle** | Prep cycle prompt |
+| **prep-plan** | Prep planning prompt |
+
+::: tip Tip
+Codex Prompts are defined in the `.codex/prompts/` directory, optimized specifically for the Codex model.
+:::
+
+---
+
+## 3.2 Three-Layer Relationship
+
+```
+┌─────────────────────────────────────────────────────┐
+│                    User Request                      │
+└────────────────────┬────────────────────────────────┘
+                     │
+                     ▼
+┌─────────────────────────────────────────────────────┐
+│                  ccw (Orchestrator)                  │
+│  Intent Analysis → Workflow Selection → Execution    │
+└────────────────────┬────────────────────────────────┘
+                     │
+         ┌───────────┼───────────┐
+         ▼           ▼           ▼
+    ┌────────┐  ┌────────┐  ┌────────┐
+    │ Command│  │ Skill  │  │ Prompt │
+    │ (Atom) │  │(Composite)│ │(Template)│
+    └────────┘  └────────┘  └────────┘
+         │           │           │
+         └───────────┼───────────┘
+                     ▼
+            ┌────────────────┐
+            │ AI Model Call   │
+            └────────────────┘
+```
+
+### 3.2.1 Call Path
+
+1. **User initiates request** → Enter command or describe requirements in VS Code
+2. **ccw orchestration** → Intent analysis, select appropriate workflow
+3. **Execute Command** → Execute atomic command operations
+4. **Call Skill** → For complex logic, call composite skills
+5. **Use Prompt** → For specific models, use optimized prompts
+6. **AI model execution** → Call configured AI model
+7. **Return result** → Format output to user
+
+---
+
+## 3.3 Workflow Session Management
+
+### 3.3.1 Session Lifecycle
+
+```
+┌─────────┐     ┌─────────┐     ┌─────────┐     ┌─────────┐
+│  Start  │────▶│ Resume  │────▶│ Execute │────▶│Complete │
+│  Launch │     │  Resume │     │ Execute │     │ Complete│
+└─────────┘     └─────────┘     └─────────┘     └─────────┘
+     │                                │
+     ▼                                ▼
+┌─────────┐                     ┌─────────┐
+│  List   │                     │ Solidify│
+│  List   │                     │ Solidify│
+└─────────┘                     └─────────┘
+```
+
+### 3.3.2 Session Commands
+
+| Command | Function | Example |
+| --- | --- | --- |
+| **start** | Start new session | `/workflow:session:start` |
+| **resume** | Resume existing session | `/workflow:session:resume <session-id>` |
+| **list** | List all sessions | `/workflow:session:list` |
+| **sync** | Sync session state | `/workflow:session:sync` |
+| **complete** | Complete current session | `/workflow:session:complete` |
+| **solidify** | Solidify session results | `/workflow:session:solidify` |
+
+### 3.3.3 Session Directory Structure
+
+```
+.workflow/
+├── .team/
+│   └── TC-<project>-<date>/      # Session directory
+│       ├── spec/                  # Session specifications
+│       │   ├── discovery-context.json
+│       │   └── requirements.md
+│       ├── artifacts/             # Session artifacts
+│       ├── wisdom/                # Session wisdom
+│       │   ├── learnings.md
+│       │   ├── decisions.md
+│       │   ├── conventions.md
+│       │   └── issues.md
+│       └── .team-msg/             # Message bus
+```
+
+---
+
+## 3.4 Team Collaboration Patterns
+
+### 3.4.1 Role System
+
+Claude_dms3 supports 8 team workflows, each defining different roles:
+
+| Workflow | Roles | Description |
+| --- | --- | --- |
+| **PlanEx** | planner, executor | Planning-execution separation |
+| **IterDev** | developer, reviewer | Iterative development |
+| **Lifecycle** | analyzer, developer, tester, reviewer | Lifecycle coverage |
+| **Issue** | discoverer, planner, executor | Issue-driven |
+| **Testing** | tester, developer | Test-driven |
+| **QA** | qa, developer | Quality assurance |
+| **Brainstorm** | facilitator, perspectives | Multi-perspective analysis |
+| **UIDesign** | designer, developer | UI design generation |
+
+### 3.4.2 Message Bus
+
+Team members communicate via the message bus:
+
+```
+┌────────────┐                   ┌────────────┐
+│  Planner   │                   │ Executor   │
+└─────┬──────┘                   └──────┬─────┘
+      │                                 │
+      │  [plan_ready]                   │
+      ├────────────────────────────────▶
+      │                                 │
+      │                          [task_complete]
+      │◀────────────────────────────────┤
+      │                                 │
+      │   [plan_approved]               │
+      ├────────────────────────────────▶
+      │                                 │
+```
+
+### 3.4.3 Workflow Selection Guide
+
+| Task Objective | Recommended Workflow | Command |
+| --- | --- | --- |
+| New feature development | PlanEx | `/workflow:plan` |
+| Bug fix | Lifecycle | `/debug-with-file` |
+| Code refactoring | IterDev | `/refactor-cycle` |
+| Technical decision | Brainstorm | `/brainstorm-with-file` |
+| UI development | UIDesign | `/workflow:ui-design` |
+| Integration testing | Testing | `/integration-test-cycle` |
+| Code review | QA | `/review-cycle` |
+| Issue management | Issue | `/issue` series |
+
+---
+
+## 3.5 Core Concepts Overview
+
+| Concept | Description | Location/Command |
+| --- | --- | --- |
+| **Command** | Atomic operation commands | `.claude/commands/` |
+| **Skill** | Composite skill encapsulation | `.claude/skills/` |
+| **Prompt** | Codex prompt templates | `.codex/prompts/` |
+| **Workflow** | Team collaboration process | `/workflow:*` |
+| **Session** | Session context management | `/workflow:session:*` |
+| **Memory** | Cross-session knowledge persistence | `ccw memory` |
+| **Spec** | Project specification constraints | `.workflow/specs/` |
+| **CodexLens** | Semantic code indexing | `.codex-lens/` |
+| **CCW** | CLI invocation framework | `ccw` directory |
+
+---
+
+## 3.6 Data Flow
+
+```
+User Request
+   │
+   ▼
+┌──────────────┐
+│  CCW Orchestrator│ ──▶ Intent Analysis
+└──────────────┘
+   │
+   ├─▶ Workflow Selection
+   │      │
+   │      ├─▶ PlanEx
+   │      ├─▶ IterDev
+   │      ├─▶ Lifecycle
+   │      └─▶ ...
+   │
+   ├─▶ Command Execution
+   │      │
+   │      ├─▶ Built-in commands
+   │      └─▶ Skill calls
+   │
+   ├─▶ AI Model Invocation
+   │      │
+   │      ├─▶ Gemini
+   │      ├─▶ Qwen
+   │      ├─▶ Codex
+   │      └─▶ Claude
+   │
+   └─▶ Result Return
+          │
+          ├─▶ File modification
+          ├─▶ Memory update
+          └─▶ Dashboard update
+```
+
+---
+
+## Next Steps
+
+- [Workflow Basics](ch04-workflow-basics.md) — Learn various workflow commands
+- [Advanced Tips](ch05-advanced-tips.md) — CLI toolchain, multi-model collaboration
+- [Best Practices](ch06-best-practices.md) — Team collaboration standards, code review process

docs/guide/ch04-workflow-basics.md

@@ -0,0 +1,328 @@
+# Workflow Basics
+
+## One-Line Positioning
+
+**Workflows are the core of team collaboration** — 8 workflows covering the full development lifecycle, from planning to execution, from analysis to testing.
+
+---
+
+## 4.1 Workflow Overview
+
+| Workflow | Core Command | Use Case | Roles |
+| --- | --- | --- | --- |
+| **PlanEx** | `/workflow:plan` | New feature development, requirement implementation | planner, executor |
+| **IterDev** | `/refactor-cycle` | Code refactoring, technical debt handling | developer, reviewer |
+| **Lifecycle** | `/unified-execute-with-file` | Complete development cycle | analyzer, developer, tester, reviewer |
+| **Issue** | `/issue:*` | Issue-driven development | discoverer, planner, executor |
+| **Testing** | `/integration-test-cycle` | Integration testing, test generation | tester, developer |
+| **QA** | `/review-cycle` | Code review and quality assurance | qa, developer |
+| **Brainstorm** | `/brainstorm-with-file` | Multi-perspective analysis, technical decisions | facilitator, perspectives |
+| **UIDesign** | `/workflow:ui-design` | UI design and code generation | designer, developer |
+
+---
+
+## 4.2 PlanEx - Planning-Execution Workflow
+
+### 4.2.1 One-Line Positioning
+
+**PlanEx is a planning-execution separation workflow** — Plan first, then execute, ensuring tasks are clear before starting work.
+
+### 4.2.2 Launch Method
+
+```
+/workflow:plan
+```
+
+Or describe requirements directly:
+
+```
+Implement user login functionality, supporting email and phone number login
+```
+
+### 4.2.3 Workflow Process
+
+```
+┌─────────────┐     ┌─────────────┐     ┌─────────────┐
+│  Planner    │────▶│  Executor   │────▶│  Reviewer   │
+│  Planning   │     │  Execution  │     │  Review     │
+└─────────────┘     └─────────────┘     └─────────────┘
+     │                  │                    │
+     ▼                  ▼                    ▼
+┌─────────────┐  ┌─────────────┐    ┌─────────────┐
+│ Requirements│  │ Task        │    │ Code        │
+│ Analysis    │  │ Execution   │    │ Review      │
+│ Task Breakdown│  Code Gen    │    │ Quality     │
+│ Plan Gen    │  │ Test Write  │    │ Feedback    │
+└─────────────┘  └─────────────┘    └─────────────┘
+```
+
+### 4.2.4 Output Artifacts
+
+| Artifact | Location | Description |
+| --- | --- | --- |
+| **Requirements Analysis** | `artifacts/requirements.md` | Detailed requirement analysis |
+| **Task Plan** | `artifacts/plan.md` | Structured task list |
+| **Execution Artifacts** | `artifacts/implementation/` | Code and tests |
+| **Wisdom Accumulation** | `wisdom/learnings.md` | Lessons learned |
+
+---
+
+## 4.3 IterDev - Iterative Development Workflow
+
+### 4.3.1 One-Line Positioning
+
+**IterDev is an iterative refactoring workflow** — Discover technical debt, plan refactoring, improve iteratively.
+
+### 4.3.2 Launch Method
+
+```
+/refactor-cycle
+```
+
+### 4.3.3 Workflow Process
+
+```
+┌─────────────┐     ┌─────────────┐     ┌─────────────┐
+│  Discover   │────▶│   Plan      │────▶│  Refactor   │
+│  Discovery  │     │  Planning   │     │ Refactoring │
+└─────────────┘     └─────────────┘     └─────────────┘
+     │                  │                    │
+     ▼                  ▼                    ▼
+┌─────────────┐  ┌─────────────┐    ┌─────────────┐
+│ Code        │  │ Refactor    │    │ Code        │
+│ Analysis    │  │ Strategy    │    │ Modification│
+│ Problem     │  │ Priority    │    │ Test        │
+│ ID          │  │ Task Breakdown│  │ Verification│
+│ Tech Debt   │  │             │    │ Doc Update  │
+└─────────────┘  └─────────────┘    └─────────────┘
+```
+
+### 4.3.4 Use Cases
+
+| Scenario | Example |
+| --- | --- |
+| **Code Smells** | Long functions, duplicate code |
+| **Architecture Improvement** | Decoupling, modularization |
+| **Performance Optimization** | Algorithm optimization, caching strategy |
+| **Security Hardening** | Fix security vulnerabilities |
+| **Standard Unification** | Code style consistency |
+
+---
+
+## 4.4 Lifecycle - Lifecycle Workflow
+
+### 4.4.1 One-Line Positioning
+
+**Lifecycle is a full-lifecycle coverage workflow** — From analysis to testing to review, complete closed loop.
+
+### 4.4.2 Launch Method
+
+```
+/unified-execute-with-file <file>
+```
+
+### 4.4.3 Role Responsibilities
+
+| Role | Responsibility | Output |
+| --- | --- | --- |
+| **Analyzer** | Analyze requirements, explore code | Analysis report |
+| **Developer** | Implement features, write tests | Code + tests |
+| **Tester** | Run tests, verify functionality | Test report |
+| **Reviewer** | Code review, quality check | Review report |
+
+### 4.4.4 Workflow Process
+
+```
+┌─────────┐   ┌─────────┐   ┌─────────┐   ┌─────────┐
+│Analyzer │──▶│Developer│──▶│ Tester  │──▶│Reviewer │
+│ Analysis │   │ Develop │   │ Test    │   │ Review  │
+└─────────┘   └─────────┘   └─────────┘   └─────────┘
+     │            │            │            │
+     ▼            ▼            ▼            ▼
+ Requirement   Code         Test         Quality
+ Analysis      Implementation Verification Gate
+ Code          Unit         Regression   Final
+ Exploration   Test         Test         Confirmation
+```
+
+---
+
+## 4.5 Issue - Issue Management Workflow
+
+### 4.5.1 One-Line Positioning
+
+**Issue is an issue-driven development workflow** — From issue discovery to planning to execution, complete tracking.
+
+### 4.5.2 Issue Commands
+
+| Command | Function | Example |
+| --- | --- | --- |
+| **discover** | Discover Issue | `/issue discover https://github.com/xxx/issue/1` |
+| **discover-by-prompt** | Create from Prompt | `/issue discover-by-prompt "Login failed"` |
+| **from-brainstorm** | Create from brainstorm | `/issue from-brainstorm` |
+| **plan** | Batch plan Issues | `/issue plan` |
+| **queue** | Form execution queue | `/issue queue` |
+| **execute** | Execute Issue queue | `/issue execute` |
+
+### 4.5.3 Workflow Process
+
+```
+┌─────────┐   ┌─────────┐   ┌─────────┐   ┌─────────┐
+│Discover │──▶│  Plan   │──▶│  Queue  │──▶│ Execute │
+│Discovery│   │  Plan   │   │  Queue  │   │ Execute │
+└─────────┘   └─────────┘   └─────────┘   └─────────┘
+     │            │            │            │
+     ▼            ▼            ▼            ▼
+  Identify      Analyze      Priority     Implement
+  Problems      Requirements Sort          Solution
+  Define        Plan         Dependencies Verify
+  Scope                      Results
+```
+
+---
+
+## 4.6 Testing - Testing Workflow
+
+### 4.6.1 One-Line Positioning
+
+**Testing is a self-iterating test workflow** — Auto-generate tests, iteratively improve test coverage.
+
+### 4.6.2 Launch Method
+
+```
+/integration-test-cycle
+```
+
+### 4.6.3 Workflow Process
+
+```
+┌─────────┐   ┌─────────┐   ┌─────────┐
+│Generate │──▶│ Execute │──▶│ Verify  │
+│ Generate │   │ Execute │   │ Verify  │
+└─────────┘   └─────────┘   └─────────┘
+     │            │            │
+     ▼            ▼            ▼
+ Test Cases   Run Tests    Coverage
+ Mock Data    Failure      Analysis
+              Analysis     Gap Fill
+```
+
+---
+
+## 4.7 QA - Quality Assurance Workflow
+
+### 4.7.1 One-Line Positioning
+
+**QA is a code review workflow** — 6-dimensional code review, auto-discover issues.
+
+### 4.7.2 Launch Method
+
+```
+/review-cycle
+```
+
+### 4.7.3 Review Dimensions
+
+| Dimension | Check Items |
+| --- | --- |
+| **Correctness** | Logic correct, boundary handling |
+| **Performance** | Algorithm efficiency, resource usage |
+| **Security** | Injection vulnerabilities, permission checks |
+| **Maintainability** | Code clarity, modularity |
+| **Test Coverage** | Unit tests, boundary tests |
+| **Standard Compliance** | Coding standards, project conventions |
+
+---
+
+## 4.8 Brainstorm - Brainstorming Workflow
+
+### 4.8.1 One-Line Positioning
+
+**Brainstorm is a multi-perspective analysis workflow** — Analyze problems from multiple viewpoints for comprehensive insights.
+
+### 4.8.2 Launch Method
+
+```
+/brainstorm-with-file <file>
+```
+
+### 4.8.3 Analysis Perspectives
+
+| Perspective | Role | Focus |
+| --- | --- | --- |
+| **Product** | Product Manager | Market fit, user value |
+| **Technical** | Tech Lead | Feasibility, technical debt |
+| **Quality** | QA Lead | Completeness, testability |
+| **Risk** | Risk Analyst | Risk identification, dependencies |
+
+### 4.8.4 Output Format
+
+```markdown
+## Consensus Points
+- [Consensus point 1]
+- [Consensus point 2]
+
+## Divergences
+- [Divergence 1]
+  - Perspective A: ...
+  - Perspective B: ...
+  - Recommendation: ...
+
+## Action Items
+- [ ] [Action item 1]
+- [ ] [Action item 2]
+```
+
+---
+
+## 4.9 UIDesign - UI Design Workflow
+
+### 4.9.1 One-Line Positioning
+
+**UIDesign is a UI design generation workflow** — From design to code, auto-extract styles and layouts.
+
+### 4.9.2 UI Design Commands
+
+| Command | Function |
+| --- | --- |
+| **generate** | Generate UI components |
+| **import-from-code** | Import styles from code |
+| **style-extract** | Extract style specifications |
+| **layout-extract** | Extract layout structure |
+| **imitate-auto** | Imitate reference page |
+| **codify-style** | Convert styles to code |
+| **design-sync** | Sync design changes |
+
+---
+
+## 4.10 Quick Reference
+
+### Workflow Selection Guide
+
+| Requirement | Recommended Workflow | Command |
+| --- | --- | --- |
+| New feature development | PlanEx | `/workflow:plan` |
+| Code refactoring | IterDev | `/refactor-cycle` |
+| Complete development | Lifecycle | `/unified-execute-with-file` |
+| Issue management | Issue | `/issue:*` |
+| Test generation | Testing | `/integration-test-cycle` |
+| Code review | QA | `/review-cycle` |
+| Multi-perspective analysis | Brainstorm | `/brainstorm-with-file` |
+| UI development | UIDesign | `/workflow:ui-design` |
+
+### Session Management Commands
+
+| Command | Function |
+| --- | --- |
+| `/workflow:session:start` | Start new session |
+| `/workflow:session:resume` | Resume session |
+| `/workflow:session:list` | List sessions |
+| `/workflow:session:complete` | Complete session |
+| `/workflow:session:solidify` | Solidify results |
+
+---
+
+## Next Steps
+
+- [Advanced Tips](ch05-advanced-tips.md) — CLI toolchain, multi-model collaboration
+- [Best Practices](ch06-best-practices.md) — Team collaboration standards, code review process

docs/guide/ch05-advanced-tips.md

@@ -0,0 +1,331 @@
+# Advanced Tips
+
+## One-Line Positioning
+
+**Advanced Tips are the key to efficiency improvement** — Deep CLI toolchain usage, multi-model collaboration optimization, memory management best practices.
+
+---
+
+## 5.1 CLI Toolchain Usage
+
+### 5.1.1 CLI Configuration
+
+CLI tool configuration file: `~/.claude/cli-tools.json`
+
+```json
+{
+  "version": "3.3.0",
+  "tools": {
+    "gemini": {
+      "enabled": true,
+      "primaryModel": "gemini-2.5-flash",
+      "secondaryModel": "gemini-2.5-flash",
+      "tags": ["analysis", "Debug"],
+      "type": "builtin"
+    },
+    "qwen": {
+      "enabled": true,
+      "primaryModel": "coder-model",
+      "secondaryModel": "coder-model",
+      "tags": [],
+      "type": "builtin"
+    },
+    "codex": {
+      "enabled": true,
+      "primaryModel": "gpt-5.2",
+      "secondaryModel": "gpt-5.2",
+      "tags": [],
+      "type": "builtin"
+    }
+  }
+}
+```
+
+### 5.1.2 Tag Routing
+
+Automatically select models based on task type:
+
+| Tag | Applicable Model | Task Type |
+| --- | --- | --- |
+| **analysis** | Gemini | Code analysis, architecture design |
+| **Debug** | Gemini | Root cause analysis, problem diagnosis |
+| **implementation** | Qwen | Feature development, code generation |
+| **review** | Codex | Code review, Git operations |
+
+### 5.1.3 CLI Command Templates
+
+#### Analysis Task
+
+```bash
+ccw cli -p "PURPOSE: Identify security vulnerabilities
+TASK: • Scan for SQL injection • Check XSS • Verify CSRF
+MODE: analysis
+CONTEXT: @src/auth/**/*
+EXPECTED: Security report with severity grading and fix recommendations
+CONSTRAINTS: Focus on auth module" --tool gemini --mode analysis --rule analysis-assess-security-risks
+```
+
+#### Implementation Task
+
+```bash
+ccw cli -p "PURPOSE: Implement rate limiting
+TASK: • Create middleware • Configure routes • Redis backend
+MODE: write
+CONTEXT: @src/middleware/**/* @src/config/**/*
+EXPECTED: Production code + unit tests + integration tests
+CONSTRAINTS: Follow existing middleware patterns" --tool qwen --mode write --rule development-implement-feature
+```
+
+### 5.1.4 Rule Templates
+
+| Rule | Purpose |
+| --- | --- |
+| **analysis-diagnose-bug-root-cause** | Bug root cause analysis |
+| **analysis-analyze-code-patterns** | Code pattern analysis |
+| **analysis-review-architecture** | Architecture review |
+| **analysis-assess-security-risks** | Security assessment |
+| **development-implement-feature** | Feature implementation |
+| **development-refactor-codebase** | Code refactoring |
+| **development-generate-tests** | Test generation |
+
+---
+
+## 5.2 Multi-Model Collaboration
+
+### 5.2.1 Model Selection Guide
+
+| Task | Recommended Model | Reason |
+| --- | --- | --- |
+| **Code Analysis** | Gemini | Strong at deep code understanding and pattern recognition |
+| **Bug Diagnosis** | Gemini | Powerful root cause analysis capability |
+| **Feature Development** | Qwen | High code generation efficiency |
+| **Code Review** | Codex (GPT) | Good Git integration, standard review format |
+| **Long Text** | Claude | Large context window |
+
+### 5.2.2 Collaboration Patterns
+
+#### Serial Collaboration
+
+```bash
+# Step 1: Gemini analysis
+ccw cli -p "Analyze code architecture" --tool gemini --mode analysis
+
+# Step 2: Qwen implementation
+ccw cli -p "Implement feature based on analysis" --tool qwen --mode write
+
+# Step 3: Codex review
+ccw cli -p "Review implementation code" --tool codex --mode review
+```
+
+#### Parallel Collaboration
+
+Use `--tool gemini` and `--tool qwen` to analyze the same problem simultaneously:
+
+```bash
+# Terminal 1
+ccw cli -p "Analyze from performance perspective" --tool gemini --mode analysis &
+
+# Terminal 2
+ccw cli -p "Analyze from security perspective" --tool codex --mode analysis &
+```
+
+### 5.2.3 Session Resume
+
+Cross-model session resume:
+
+```bash
+# First call
+ccw cli -p "Analyze authentication module" --tool gemini --mode analysis
+
+# Resume session to continue
+ccw cli -p "Based on previous analysis, design improvement plan" --tool qwen --mode write --resume
+```
+
+---
+
+## 5.3 Memory Management
+
+### 5.3.1 Memory Categories
+
+| Category | Purpose | Example Content |
+| --- | --- | --- |
+| **learnings** | Learning insights | New technology usage experience |
+| **decisions** | Architecture decisions | Technology selection rationale |
+| **conventions** | Coding standards | Naming conventions, patterns |
+| **issues** | Known issues | Bugs, limitations, TODOs |
+
+### 5.3.2 Memory Commands
+
+| Command | Function | Example |
+| --- | --- | --- |
+| **list** | List all memories | `ccw memory list` |
+| **search** | Search memories | `ccw memory search "authentication"` |
+| **export** | Export memory | `ccw memory export <id>` |
+| **import** | Import memory | `ccw memory import "..."` |
+| **embed** | Generate embeddings | `ccw memory embed <id>` |
+
+### 5.3.3 Memory Best Practices
+
+::: tip Tip
+- **Regular cleanup**: Organize Memory weekly, delete outdated content
+- **Structure**: Use standard format for easy search and reuse
+- **Context**: Record decision background, not just conclusions
+- **Linking**: Cross-reference related content
+:::
+
+### 5.3.4 Memory Template
+
+```markdown
+## Title
+### Background
+- **Problem**: ...
+- **Impact**: ...
+
+### Decision
+- **Solution**: ...
+- **Rationale**: ...
+
+### Result
+- **Effect**: ...
+- **Lessons Learned**: ...
+
+### Related
+- [Related Memory 1](memory-id-1)
+- [Related Documentation](link)
+```
+
+---
+
+## 5.4 CodexLens Advanced Usage
+
+### 5.4.1 Hybrid Search
+
+Combine vector search and keyword search:
+
+```bash
+# Pure vector search
+ccw search --mode vector "user authentication"
+
+# Hybrid search (default)
+ccw search --mode hybrid "user authentication"
+
+# Pure keyword search
+ccw search --mode keyword "authenticate"
+```
+
+### 5.4.2 Call Chain Tracing
+
+Trace complete call chains of functions:
+
+```bash
+# Trace up (who called me)
+ccw search --trace-up "authenticateUser"
+
+# Trace down (who I called)
+ccw search --trace-down "authenticateUser"
+
+# Full call chain
+ccw search --trace-full "authenticateUser"
+```
+
+### 5.4.3 Semantic Search Techniques
+
+| Technique | Example | Effect |
+| --- | --- | --- |
+| **Functional description** | "handle user login" | Find login-related code |
+| **Problem description** | "memory leak locations" | Find potential issues |
+| **Pattern description** | "singleton implementation" | Find design patterns |
+| **Technical description** | "using React Hooks" | Find related usage |
+
+---
+
+## 5.5 Hook Auto-Injection
+
+### 5.5.1 Hook Types
+
+| Hook Type | Trigger Time | Purpose |
+| --- | --- | --- |
+| **pre-command** | Before command execution | Inject specifications, load context |
+| **post-command** | After command execution | Save Memory, update state |
+| **pre-commit** | Before Git commit | Code review, standard checks |
+| **file-change** | On file change | Auto-format, update index |
+
+### 5.5.2 Hook Configuration
+
+Configure in `.claude/hooks.json`:
+
+```json
+{
+  "pre-command": [
+    {
+      "name": "inject-specs",
+      "description": "Inject project specifications",
+      "command": "cat .workflow/specs/project-constraints.md"
+    },
+    {
+      "name": "load-memory",
+      "description": "Load related Memory",
+      "command": "ccw memory search \"{query}\""
+    }
+  ],
+  "post-command": [
+    {
+      "name": "save-memory",
+      "description": "Save important decisions",
+      "command": "ccw memory import \"{content}\""
+    }
+  ]
+}
+```
+
+---
+
+## 5.6 Performance Optimization
+
+### 5.6.1 Indexing Optimization
+
+| Optimization | Description |
+| --- | --- |
+| **Incremental indexing** | Only index changed files |
+| **Parallel indexing** | Multi-process parallel processing |
+| **Caching strategy** | Vector embedding cache |
+
+### 5.6.2 Search Optimization
+
+| Optimization | Description |
+| --- | --- |
+| **Result caching** | Same query returns cached results |
+| **Paginated loading** | Large result sets paginated |
+| **Smart deduplication** | Auto-duplicate similar results |
+
+---
+
+## 5.7 Quick Reference
+
+### CLI Command Cheatsheet
+
+| Command | Function |
+| --- | --- |
+| `ccw cli -p "..." --tool gemini --mode analysis` | Analysis task |
+| `ccw cli -p "..." --tool qwen --mode write` | Implementation task |
+| `ccw cli -p "..." --tool codex --mode review` | Review task |
+| `ccw memory list` | List memories |
+| `ccw memory search "..."` | Search memories |
+| `ccw search "..."` | Semantic search |
+| `ccw search --trace "..."` | Call chain tracing |
+
+### Rule Template Cheatsheet
+
+| Rule | Purpose |
+| --- | --- |
+| `analysis-diagnose-bug-root-cause` | Bug analysis |
+| `analysis-assess-security-risks` | Security assessment |
+| `development-implement-feature` | Feature implementation |
+| `development-refactor-codebase` | Code refactoring |
+| `development-generate-tests` | Test generation |
+
+---
+
+## Next Steps
+
+- [Best Practices](ch06-best-practices.md) — Team collaboration standards, code review process, documentation maintenance strategy

docs/guide/ch06-best-practices.md

@@ -0,0 +1,330 @@
+# Best Practices
+
+## One-Line Positioning
+
+**Best Practices ensure efficient team collaboration** — Practical experience summary on standard formulation, code review, documentation maintenance, and team collaboration.
+
+---
+
+## 6.1 Team Collaboration Standards
+
+### 6.1.1 Role Responsibility Definitions
+
+| Role | Responsibility | Skill Requirements |
+| --- | --- | --- |
+| **Planner** | Requirement analysis, task breakdown | Architectural thinking, communication skills |
+| **Developer** | Code implementation, unit testing | Coding ability, testing awareness |
+| **Reviewer** | Code review, quality gatekeeping | Code sensitivity, standard understanding |
+| **QA** | Quality assurance, test verification | Test design, risk identification |
+| **Facilitator** | Coordination, progress tracking | Project management, conflict resolution |
+
+### 6.1.2 Workflow Selection
+
+| Scenario | Recommended Workflow | Rationale |
+| --- | --- | --- |
+| **New Feature Development** | PlanEx | Planning-execution separation, reduces risk |
+| **Bug Fix** | Lifecycle | Complete closed loop, ensures fix quality |
+| **Code Refactoring** | IterDev | Iterative improvement, continuous optimization |
+| **Technical Decision** | Brainstorm | Multi-perspective analysis, comprehensive consideration |
+| **Issue Management** | Issue | Traceable, manageable |
+| **UI Development** | UIDesign | Seamless design-to-code transition |
+
+### 6.1.3 Communication Protocols
+
+#### Message Format
+
+```
+[<Role>] <Action> <Object>: <Result>
+
+Examples:
+[Planner] Task breakdown complete: Generated 5 subtasks
+[Developer] Code implementation complete: user-auth.ts, 324 lines
+[Reviewer] Code review complete: Found 2 issues, suggested 1 optimization
+```
+
+#### Status Reporting
+
+| Status | Meaning | Next Action |
+| --- | --- | --- |
+| **pending** | Pending | Wait for dependencies to complete |
+| **in_progress** | In progress | Continue execution |
+| **completed** | Completed | Can be depended upon |
+| **blocked** | Blocked | Manual intervention required |
+
+---
+
+## 6.2 Code Review Process
+
+### 6.2.1 Review Dimensions
+
+| Dimension | Check Items | Severity |
+| --- | --- | --- |
+| **Correctness** | Logic correct, boundary handling | HIGH |
+| **Performance** | Algorithm efficiency, resource usage | MEDIUM |
+| **Security** | Injection vulnerabilities, permission checks | HIGH |
+| **Maintainability** | Code clarity, modularity | MEDIUM |
+| **Test Coverage** | Unit tests, boundary tests | MEDIUM |
+| **Standard Compliance** | Coding standards, project conventions | LOW |
+
+### 6.2.2 Review Process
+
+```
+┌─────────┐   ┌─────────┐   ┌─────────┐   ┌─────────┐
+│ Submit  │──▶│ Review  │──▶│ Feedback│──▶│  Fix    │
+│ Code    │   │ Code    │   │ Comments│   │ Issues  │
+└─────────┘   └─────────┘   └─────────┘   └─────────┘
+     │            │            │            │
+     ▼            ▼            ▼            ▼
+  Push PR      Auto Review  Manual Review  Fix Verify
+  CI Check     6 Dimensions  Code Walkthrough Re-review
+```
+
+### 6.2.3 Review Checklist
+
+#### Code Correctness
+- [ ] Logic correct, no bugs
+- [ ] Boundary condition handling
+- [ ] Complete error handling
+- [ ] Proper resource cleanup
+
+#### Performance
+- [ ] Reasonable algorithm complexity
+- [ ] No memory leaks
+- [ ] No redundant computation
+- [ ] Reasonable caching strategy
+
+#### Security
+- [ ] No SQL injection
+- [ ] No XSS vulnerabilities
+- [ ] Complete permission checks
+- [ ] Sensitive data protection
+
+#### Maintainability
+- [ ] Clear naming
+- [ ] Good modularity
+- [ ] Sufficient comments
+- [ ] Easy to test
+
+#### Test Coverage
+- [ ] Complete unit tests
+- [ ] Boundary test coverage
+- [ ] Exception case testing
+- [ ] Integration test verification
+
+#### Standard Compliance
+- [ ] Unified coding style
+- [ ] Naming standard compliance
+- [ ] Project convention adherence
+- [ ] Complete documentation
+
+### 6.2.4 Feedback Format
+
+```markdown
+## Review Results
+
+### Issues
+1. **[HIGH]** SQL injection risk
+   - Location: `src/auth/login.ts:45`
+   - Recommendation: Use parameterized queries
+
+2. **[MEDIUM]** Performance issue
+   - Location: `src/utils/cache.ts:78`
+   - Recommendation: Use LRU cache
+
+### Suggestions
+1. Naming optimization: `data` → `userData`
+2. Module separation: Consider extracting Auth logic independently
+
+### Approval Conditions
+- [ ] Fix HIGH issues
+- [ ] Consider MEDIUM suggestions
+```
+
+---
+
+## 6.3 Documentation Maintenance Strategy
+
+### 6.3.1 Documentation Classification
+
+| Type | Location | Update Frequency | Owner |
+| --- | --- | --- | --- |
+| **Spec Documents** | `.workflow/specs/` | As needed | Architect |
+| **Reference Docs** | `docs/ref/` | Every change | Developer |
+| **Guide Docs** | `docs/guide/` | Monthly | Technical Writer |
+| **API Docs** | `docs/api/` | Auto-generated | Tools |
+| **FAQ** | `docs/faq.md` | Weekly | Support Team |
+
+### 6.3.2 Documentation Update Triggers
+
+| Event | Update Content |
+| --- | --- |
+| **New Feature** | Add feature documentation and API reference |
+| **Spec Change** | Update spec documents and migration guide |
+| **Bug Fix** | Update FAQ and known issues |
+| **Architecture Change** | Update architecture docs and decision records |
+| **Code Review** | Supplement missing comments and docs |
+
+### 6.3.3 Documentation Quality Standards
+
+| Standard | Requirement |
+| --- | --- |
+| **Accuracy** | Consistent with actual code |
+| **Completeness** | Cover all public APIs |
+| **Clarity** | Easy to understand, sufficient examples |
+| **Timeliness** | Updated promptly, not lagging |
+| **Navigability** | Clear structure, easy to find |
+
+---
+
+## 6.4 Memory Management Best Practices
+
+### 6.4.1 Memory Recording Triggers
+
+| Trigger | Record Content |
+| --- | --- |
+| **Architecture Decisions** | Technology selection, design decisions |
+| **Problem Resolution** | Bug root cause, solutions |
+| **Experience Summary** | Best practices, gotchas |
+| **Standard Conventions** | Coding standards, naming conventions |
+| **Known Issues** | Bugs, limitations, TODOs |
+
+### 6.4.2 Memory Format Standards
+
+```markdown
+## [Type] Title
+
+### Background
+- **Problem**: ...
+- **Impact**: ...
+- **Context**: ...
+
+### Analysis/Decision
+- **Solution**: ...
+- **Rationale**: ...
+- **Alternatives**: ...
+
+### Result
+- **Effect**: ...
+- **Data**: ...
+
+### Related
+- [Related Memory](memory-id)
+- [Related Code](file-link)
+- [Related Documentation](doc-link)
+```
+
+### 6.4.3 Memory Maintenance
+
+| Maintenance Item | Frequency | Content |
+| --- | --- | --- |
+| **Deduplication** | Weekly | Merge duplicate memories |
+| **Update** | As needed | Update outdated content |
+| **Archive** | Monthly | Archive historical memories |
+| **Cleanup** | Quarterly | Delete useless memories |
+
+---
+
+## 6.5 Hook Usage Standards
+
+### 6.5.1 Hook Types
+
+| Hook Type | Purpose | Example |
+| --- | --- | --- |
+| **pre-command** | Inject specifications, load context | Auto-load CLAUDE.md |
+| **post-command** | Save Memory, update index | Auto-save decisions |
+| **pre-commit** | Code review, standard checks | Auto-run Lint |
+| **file-change** | Auto-format, update index | Auto-format code |
+
+### 6.5.2 Hook Configuration Principles
+
+| Principle | Description |
+| --- | --- |
+| **Minimize** | Only configure necessary Hooks |
+| **Idempotent** | Hook execution results are repeatable |
+| **Recoverable** | Hook failure doesn't affect main flow |
+| **Observable** | Hook execution has logging |
+
+---
+
+## 6.6 Team Collaboration Techniques
+
+### 6.6.1 Conflict Resolution
+
+| Conflict Type | Resolution Strategy |
+| --- | --- |
+| **Standard Conflict** | Team discussion, unify standards |
+| **Technical Disagreement** | Brainstorm, data-driven |
+| **Schedule Conflict** | Priority sorting, resource adjustment |
+| **Quality Conflict** | Set standards, automated checks |
+
+### 6.6.2 Knowledge Sharing
+
+| Method | Frequency | Content |
+| --- | --- | --- |
+| **Tech Sharing** | Weekly | New technologies, best practices |
+| **Code Walkthrough** | Every PR | Code logic, design approach |
+| **Documentation Sync** | Monthly | Documentation updates, standard changes |
+| **Incident Retrospective** | Every incident | Root cause analysis, improvements |
+
+### 6.6.3 Efficiency Improvement
+
+| Technique | Effect |
+| --- | --- |
+| **Templating** | Reuse successful patterns |
+| **Automation** | Reduce repetitive work |
+| **Tooling** | Improve development efficiency |
+| **Standardization** | Lower communication cost |
+
+---
+
+## 6.7 Quick Reference
+
+### Workflow Selection Guide
+
+| Scenario | Workflow | Command |
+| --- | --- | --- |
+| New Feature | PlanEx | `/workflow:plan` |
+| Bug Fix | Lifecycle | `/unified-execute-with-file` |
+| Refactoring | IterDev | `/refactor-cycle` |
+| Decision | Brainstorm | `/brainstorm-with-file` |
+| UI Development | UIDesign | `/workflow:ui-design` |
+
+### Code Review Checklist
+
+- [ ] Correctness check
+- [ ] Performance check
+- [ ] Security check
+- [ ] Maintainability check
+- [ ] Test coverage check
+- [ ] Standard compliance check
+
+### Memory Maintenance
+
+| Operation | Command |
+| --- | --- |
+| List memories | `ccw memory list` |
+| Search memories | `ccw memory search "..."` |
+| Import memory | `ccw memory import "..."` |
+| Export memory | `ccw memory export <id>` |
+
+---
+
+## Summary
+
+Claude_dms3 best practices can be summarized as:
+
+1. **Standards First** - Establish clear team standards
+2. **Process Assurance** - Use appropriate workflows
+3. **Quality Gatekeeping** - Strict code review
+4. **Knowledge Accumulation** - Continuously maintain Memory and documentation
+5. **Continuous Improvement** - Regular retrospectives and optimization
+
+---
+
+## Related Links
+
+- [What is Claude_dms3](ch01-what-is-claude-dms3.md)
+- [Getting Started](ch02-getting-started.md)
+- [Core Concepts](ch03-core-concepts.md)
+- [Workflow Basics](ch04-workflow-basics.md)
+- [Advanced Tips](ch05-advanced-tips.md)

docs/guide/claude-md.md

@@ -0,0 +1,225 @@
+# CLAUDE.md Guide
+
+Configure project-specific instructions for CCW using CLAUDE.md.
+
+## What is CLAUDE.md?
+
+`CLAUDE.md` is a special file that contains project-specific instructions, conventions, and preferences for CCW. It's automatically loaded when CCW operates on your project.
+
+## File Location
+
+Place `CLAUDE.md` in your project root:
+
+```
+my-project/
+├── CLAUDE.md          # Project instructions
+├── package.json
+└── src/
+```
+
+## File Structure
+
+```markdown
+# Project Name
+
+## Overview
+Brief description of the project.
+
+## Tech Stack
+- Frontend: Framework + libraries
+- Backend: Runtime + framework
+- Database: Storage solution
+
+## Coding Standards
+- Style guide
+- Linting rules
+- Formatting preferences
+
+## Architecture
+- Project structure
+- Key patterns
+- Important conventions
+
+## Development Guidelines
+- How to add features
+- Testing requirements
+- Documentation standards
+```
+
+## Example CLAUDE.md
+
+```markdown
+# E-Commerce Platform
+
+## Overview
+Multi-tenant e-commerce platform with headless architecture.
+
+## Tech Stack
+- Frontend: Vue 3 + TypeScript + Vite
+- Backend: Node.js + NestJS
+- Database: PostgreSQL + Redis
+- Queue: BullMQ
+
+## Coding Standards
+
+### TypeScript
+- Use strict mode
+- No implicit any
+- Explicit return types
+
+### Naming Conventions
+- Components: PascalCase (UserProfile.ts)
+- Utilities: camelCase (formatDate.ts)
+- Constants: UPPER_SNAKE_CASE (API_URL)
+
+### File Structure
+```
+src/
+├── components/     # Vue components
+├── composables/    # Vue composables
+├── services/       # Business logic
+├── types/          # TypeScript types
+└── utils/          # Utilities
+```
+
+## Architecture
+
+### Layered Architecture
+1. **Presentation Layer**: Vue components
+2. **Application Layer**: Composables and services
+3. **Domain Layer**: Business logic
+4. **Infrastructure Layer**: External services
+
+### Key Patterns
+- Repository pattern for data access
+- Factory pattern for complex objects
+- Strategy pattern for payments
+
+## Development Guidelines
+
+### Adding Features
+1. Create feature branch from develop
+2. Implement feature with tests
+3. Update documentation
+4. Create PR with template
+
+### Testing
+- Unit tests: Vitest
+- E2E tests: Playwright
+- Coverage: >80%
+
+### Commits
+Follow conventional commits:
+- feat: New feature
+- fix: Bug fix
+- docs: Documentation
+- refactor: Refactoring
+- test: Tests
+
+## Important Notes
+- Always use TypeScript strict mode
+- Never commit .env files
+- Run linter before commit
+- Update API docs for backend changes
+```
+
+## Sections
+
+### Required Sections
+
+| Section | Purpose |
+|---------|---------|
+| Overview | Project description |
+| Tech Stack | Technologies used |
+| Coding Standards | Style conventions |
+| Architecture | System design |
+
+### Optional Sections
+
+| Section | Purpose |
+|---------|---------|
+| Testing | Test requirements |
+| Deployment | Deploy process |
+| Troubleshooting | Common issues |
+| References | External docs |
+
+## Best Practices
+
+### 1. Keep It Current
+
+Update CLAUDE.md when:
+- Tech stack changes
+- New patterns adopted
+- Standards updated
+
+### 2. Be Specific
+
+Instead of:
+```markdown
+## Style
+Follow good practices
+```
+
+Use:
+```markdown
+## Style
+- Use ESLint with project config
+- Max line length: 100
+- Use single quotes for strings
+```
+
+### 3. Provide Examples
+
+```markdown
+## Naming
+Components use PascalCase:
+- UserProfile.vue ✓
+- userProfile.vue ✗
+```
+
+## Multiple Projects
+
+For monorepos, use multiple CLAUDE.md files:
+
+```
+monorepo/
+├── CLAUDE.md           # Root instructions
+├── packages/
+│   ├── frontend/
+│   │   └── CLAUDE.md   # Frontend specific
+│   └── backend/
+│       └── CLAUDE.md   # Backend specific
+```
+
+## Template
+
+```markdown
+# [Project Name]
+
+## Overview
+[1-2 sentence description]
+
+## Tech Stack
+- [Framework/Language]
+- [Key libraries]
+
+## Coding Standards
+- [Style guide]
+- [Linting]
+
+## Architecture
+- [Structure]
+- [Patterns]
+
+## Development
+- [How to add features]
+- [Testing approach]
+
+## Notes
+- [Important conventions]
+```
+
+::: info See Also
+- [Configuration](./cli-tools.md) - CLI tools config
+- [Workflows](../workflows/) - Development workflows
+:::

docs/guide/cli-tools.md

@@ -0,0 +1,272 @@
+# CLI Tools Configuration
+
+Configure and customize CCW CLI tools for your development workflow.
+
+## Configuration File
+
+CCW CLI tools are configured in `~/.claude/cli-tools.json`:
+
+```json
+{
+  "version": "3.3.0",
+  "tools": {
+    "tool-id": {
+      "enabled": true,
+      "primaryModel": "model-name",
+      "secondaryModel": "fallback-model",
+      "tags": ["tag1", "tag2"],
+      "type": "builtin | api-endpoint | cli-wrapper"
+    }
+  }
+}
+```
+
+## Tool Types
+
+### Builtin Tools
+
+Full-featured tools with all capabilities:
+
+```json
+{
+  "gemini": {
+    "enabled": true,
+    "primaryModel": "gemini-2.5-flash",
+    "secondaryModel": "gemini-2.5-pro",
+    "tags": ["analysis", "debug"],
+    "type": "builtin"
+  }
+}
+```
+
+**Capabilities**: Analysis + Write tools
+
+### API Endpoint Tools
+
+Analysis-only tools for specialized tasks:
+
+```json
+{
+  "custom-api": {
+    "enabled": true,
+    "primaryModel": "custom-model",
+    "tags": ["specialized-analysis"],
+    "type": "api-endpoint"
+  }
+}
+```
+
+**Capabilities**: Analysis only
+
+## CLI Command Format
+
+### Universal Template
+
+```bash
+ccw cli -p "PURPOSE: [goal] + [why] + [success criteria]
+TASK: • [step 1] • [step 2] • [step 3]
+MODE: [analysis|write|review]
+CONTEXT: @[file patterns] | Memory: [context]
+EXPECTED: [output format]
+CONSTRAINTS: [constraints]" --tool <tool-id> --mode <mode> --rule <template>
+```
+
+### Required Parameters
+
+| Parameter | Description | Options |
+|-----------|-------------|---------|
+| `--mode <mode>` | **REQUIRED** - Execution permission level | `analysis` (read-only) \| `write` (create/modify) \| `review` (git-aware review) |
+| `-p <prompt>` | **REQUIRED** - Task prompt with structured template | - |
+
+### Optional Parameters
+
+| Parameter | Description | Example |
+|-----------|-------------|---------|
+| `--tool <tool>` | Explicit tool selection | `--tool gemini` |
+| `--rule <template>` | Load rule template for structured prompts | `--rule analysis-review-architecture` |
+| `--resume [id]` | Resume previous session | `--resume` or `--resume session-id` |
+| `--cd <path>` | Set working directory | `--cd src/auth` |
+| `--includeDirs <dirs>` | Include additional directories (comma-separated) | `--includeDirs ../shared,../types` |
+| `--model <model>` | Override tool's primary model | `--model gemini-2.5-pro` |
+
+## Tool Selection
+
+### Tag-Based Routing
+
+Tools are selected based on task requirements:
+
+```bash
+# Task with "analysis" tag routes to gemini
+ccw cli -p "PURPOSE: Debug authentication issue
+TASK: • Trace auth flow • Identify failure point
+MODE: analysis" --tool gemini --mode analysis
+
+# No tags - uses first enabled tool
+ccw cli -p "PURPOSE: Implement feature X
+TASK: • Create component • Add tests
+MODE: write" --mode write
+```
+
+### Explicit Selection
+
+Override automatic selection:
+
+```bash
+ccw cli -p "Task description" --tool codex --mode write
+```
+
+### Rule Templates
+
+Auto-load structured prompt templates:
+
+```bash
+# Architecture review template
+ccw cli -p "Analyze system architecture" --mode analysis --rule analysis-review-architecture
+
+# Feature implementation template
+ccw cli -p "Add OAuth2 authentication" --mode write --rule development-implement-feature
+```
+
+## Model Configuration
+
+### Primary vs Secondary
+
+```json
+{
+  "codex": {
+    "primaryModel": "gpt-5.2",
+    "secondaryModel": "gpt-5.2"
+  }
+}
+```
+
+- **primaryModel**: Default model for the tool
+- **secondaryModel**: Fallback if primary fails
+
+### Available Models
+
+| Tool | Available Models |
+|------|------------------|
+| gemini | gemini-3-pro-preview, gemini-2.5-pro, gemini-2.5-flash, gemini-2.0-flash |
+| codex | gpt-5.2 |
+| claude | sonnet, haiku |
+| qwen | coder-model |
+
+## Tool Tags
+
+Tags enable automatic tool selection:
+
+| Tag | Use Case |
+|-----|----------|
+| analysis | Code review, architecture analysis |
+| debug | Bug diagnosis, troubleshooting |
+| implementation | Feature development, code generation |
+| documentation | Doc generation, technical writing |
+| testing | Test generation, coverage analysis |
+
+## Example Configurations
+
+### Development Setup
+
+```json
+{
+  "version": "3.3.0",
+  "tools": {
+    "gemini": {
+      "enabled": true,
+      "primaryModel": "gemini-2.5-flash",
+      "tags": ["development", "debug"],
+      "type": "builtin"
+    },
+    "codex": {
+      "enabled": true,
+      "primaryModel": "gpt-5.2",
+      "tags": ["implementation", "review"],
+      "type": "builtin"
+    }
+  }
+}
+```
+
+### Cost Optimization
+
+```json
+{
+  "tools": {
+    "gemini": {
+      "enabled": true,
+      "primaryModel": "gemini-2.0-flash",
+      "tags": ["analysis"],
+      "type": "builtin"
+    }
+  }
+}
+```
+
+### Quality Focus
+
+```json
+{
+  "tools": {
+    "codex": {
+      "enabled": true,
+      "primaryModel": "gpt-5.2",
+      "tags": ["review", "implementation"],
+      "type": "builtin"
+    },
+    "claude": {
+      "enabled": true,
+      "primaryModel": "sonnet",
+      "tags": ["documentation"],
+      "type": "builtin"
+    }
+  }
+}
+```
+
+## Validation
+
+To verify your configuration, check the config file directly:
+
+```bash
+cat ~/.claude/cli-tools.json
+```
+
+Or test tool availability:
+
+```bash
+ccw cli -p "PURPOSE: Test tool availability
+TASK: Verify tool is working
+MODE: analysis" --mode analysis
+```
+
+## Troubleshooting
+
+### Tool Not Available
+
+```bash
+Error: Tool 'custom-tool' not found
+```
+
+**Solution**: Check tool is enabled in config:
+
+```json
+{
+  "custom-tool": {
+    "enabled": true
+  }
+}
+```
+
+### Model Not Found
+
+```bash
+Error: Model 'invalid-model' not available
+```
+
+**Solution**: Use valid model name from available models list.
+
+::: info See Also
+- [CLI Reference](../cli/commands.md) - CLI usage
+- [Modes](#modes) - Execution modes
+:::

docs/guide/first-workflow.md

@@ -0,0 +1,93 @@
+# First Workflow: Build a Simple API
+
+Complete your first CCW workflow in 30 minutes. We'll build a simple REST API from specification to implementation.
+
+## What We'll Build
+
+A simple users API with:
+- GET /users - List all users
+- GET /users/:id - Get user by ID
+- POST /users - Create new user
+- PUT /users/:id - Update user
+- DELETE /users/:id - Delete user
+
+## Prerequisites
+
+- CCW installed ([Installation Guide](./installation.md))
+- Node.js >= 18.0.0
+- Code editor (VS Code recommended)
+
+## Step 1: Create Project (5 minutes)
+
+```bash
+# Create project directory
+mkdir user-api
+cd user-api
+
+# Initialize npm project
+npm init -y
+
+# Install dependencies
+npm install express
+npm install --save-dev typescript @types/node @types/express
+```
+
+## Step 2: Generate Specification (5 minutes)
+
+```bash
+# Use CCW to generate API specification
+ccw cli -p "Generate a REST API specification for a users resource with CRUD operations" --mode analysis
+```
+
+CCW will analyze your request and generate a specification document.
+
+## Step 3: Implement API (15 minutes)
+
+```bash
+# Implement the API
+ccw cli -p "Implement the users API following the specification with Express and TypeScript" --mode write
+```
+
+CCW will:
+1. Create the project structure
+2. Implement the routes
+3. Add type definitions
+4. Include error handling
+
+## Step 4: Review Code (5 minutes)
+
+```bash
+# Review the implementation
+ccw cli -p "Review the users API code for quality, security, and best practices" --mode analysis
+```
+
+## Step 5: Test and Run
+
+```bash
+# Compile TypeScript
+npx tsc
+
+# Run the server
+node dist/index.js
+
+# Test the API
+curl http://localhost:3000/users
+```
+
+## Expected Result
+
+You should have:
+- `src/index.ts` - Main server file
+- `src/routes/users.ts` - User routes
+- `src/types/user.ts` - User types
+- `src/middleware/error.ts` - Error handling
+
+## Next Steps
+
+- [CLI Reference](../cli/commands.md) - Learn all CLI commands
+- [Skills Library](../skills/core-skills.md) - Explore built-in skills
+- [Workflow System](../workflows/4-level.md) - Understand workflow orchestration
+
+::: tip Congratulations! 🎉
+You've completed your first CCW workflow. You can now use CCW for more complex projects.
+:::

docs/guide/getting-started.md

@@ -0,0 +1,55 @@
+# Getting Started with CCW
+
+Welcome to CCW (Claude Code Workspace) - an advanced AI-powered development environment that helps you write better code faster.
+
+## What is CCW?
+
+CCW is a comprehensive development environment that combines:
+
+- **Main Orchestrator (`/ccw`)**: Intent-aware workflow selection and automatic command routing
+- **AI-Powered CLI Tools**: Analyze, review, and implement code with multiple AI backends
+- **Specialized Agents**: Code execution, TDD development, testing, debugging, and documentation
+- **Workflow Orchestration**: 4-level workflow system from spec to implementation
+- **Extensible Skills**: 50+ built-in skills with custom skill support
+- **MCP Integration**: Model Context Protocol for enhanced tool integration
+
+## Quick Start
+
+### Installation
+
+```bash
+# Install CCW globally
+npm install -g claude-code-workflow
+
+# Or use with npx
+npx ccw --help
+```
+
+### Your First Workflow
+
+Create a simple workflow in under 5 minutes:
+
+```bash
+# Main orchestrator - automatically selects workflow based on intent
+/ccw "Create a new project"                    # Auto-selects appropriate workflow
+/ccw "Analyze the codebase structure"          # Auto-selects analysis workflow
+/ccw "Add user authentication"                 # Auto-selects implementation workflow
+
+# Auto-mode - skip confirmation
+/ccw -y "Fix the login timeout issue"          # Execute without confirmation prompts
+
+# Or use specific workflow commands
+/workflow:init                                  # Initialize project state
+/workflow:plan "Add OAuth2 authentication"     # Create implementation plan
+/workflow:execute                               # Execute planned tasks
+```
+
+## Next Steps
+
+- [Installation Guide](./installation.md) - Detailed installation instructions
+- [First Workflow](./first-workflow.md) - 30-minute quickstart tutorial
+- [CLI Tools](./cli-tools.md) - Customize your CCW setup
+
+::: tip Need Help?
+Check out our [GitHub Discussions](https://github.com/your-repo/ccw/discussions) or join our [Discord community](https://discord.gg/ccw).
+:::

docs/guide/installation.md

@@ -0,0 +1,191 @@
+# Installation
+
+Learn how to install and configure CCW on your system.
+
+## Prerequisites
+
+Before installing CCW, make sure you have:
+
+- **Node.js** >= 18.0.0
+- **npm** >= 9.0.0 or **yarn** >= 1.22.0
+- **Git** for version control features
+
+## Install CCW
+
+### Global Installation (Recommended)
+
+```bash
+npm install -g claude-code-workflow
+```
+
+### Project-Specific Installation
+
+```bash
+# In your project directory
+npm install --save-dev claude-code-workflow
+
+# Run with npx
+npx ccw [command]
+```
+
+### Using Yarn
+
+```bash
+# Global
+yarn global add claude-code-workflow
+
+# Project-specific
+yarn add -D claude-code-workflow
+```
+
+## Verify Installation
+
+```bash
+ccw --version
+# Output: CCW v1.0.0
+
+ccw --help
+# Shows all available commands
+```
+
+## Configuration
+
+### CLI Tools Configuration
+
+Create or edit `~/.claude/cli-tools.json`:
+
+```json
+{
+  "version": "3.3.0",
+  "tools": {
+    "gemini": {
+      "enabled": true,
+      "primaryModel": "gemini-2.5-flash",
+      "secondaryModel": "gemini-2.5-flash",
+      "tags": ["analysis", "debug"],
+      "type": "builtin"
+    },
+    "codex": {
+      "enabled": true,
+      "primaryModel": "gpt-5.2",
+      "secondaryModel": "gpt-5.2",
+      "tags": [],
+      "type": "builtin"
+    }
+  }
+}
+```
+
+### CLAUDE.md Instructions
+
+Create `CLAUDE.md` in your project root:
+
+```markdown
+# Project Instructions
+
+## Coding Standards
+- Use TypeScript for type safety
+- Follow ESLint configuration
+- Write tests for all new features
+
+## Architecture
+- Frontend: Vue 3 + Vite
+- Backend: Node.js + Express
+- Database: PostgreSQL
+```
+
+## Updating CCW
+
+```bash
+# Update to the latest version
+npm update -g claude-code-workflow
+
+# Or install a specific version
+npm install -g claude-code-workflow@latest
+```
+
+## Uninstallation
+
+```bash
+npm uninstall -g claude-code-workflow
+
+# Remove configuration (optional)
+rm -rf ~/.claude
+```
+
+## Troubleshooting
+
+### Permission Issues
+
+If you encounter permission errors:
+
+```bash
+# Use sudo (not recommended)
+sudo npm install -g claude-code-workflow
+
+# Or fix npm permissions (recommended)
+mkdir ~/.npm-global
+npm config set prefix '~/.npm-global'
+export PATH=~/.npm-global/bin:$PATH
+```
+
+### PATH Issues
+
+Add npm global bin to your PATH:
+
+```bash
+# For bash/zsh
+echo 'export PATH=$(npm config get prefix)/bin:$PATH' >> ~/.bashrc
+
+# For fish
+echo 'set -gx PATH (npm config get prefix)/bin $PATH' >> ~/.config/fish/config.fish
+```
+
+::: info Next Steps
+After installation, check out the [First Workflow](./first-workflow.md) guide.
+:::
+
+## Quick Start Example
+
+After installation, try these commands to verify everything works:
+
+```bash
+# 1. Initialize in your project
+cd your-project
+ccw init
+
+# 2. Try a simple analysis
+ccw cli -p "Analyze the project structure" --tool gemini --mode analysis
+
+# 3. Run the main orchestrator
+/ccw "Summarize the codebase architecture"
+
+# 4. Check available commands
+ccw --help
+```
+
+### Expected Output
+
+```
+$ ccw --version
+CCW v7.0.5
+
+$ ccw init
+✔ Created .claude/CLAUDE.md
+✔ Created .ccw/workflows/
+✔ Configuration complete
+
+$ ccw cli -p "Analyze project" --tool gemini --mode analysis
+Analyzing with Gemini...
+✔ Analysis complete
+```
+
+### Common First-Time Issues
+
+| Issue | Solution |
+|-------|----------|
+| `ccw: command not found` | Add npm global bin to PATH, or reinstall |
+| `Permission denied` | Use `sudo` or fix npm permissions |
+| `API key not found` | Configure API keys in `~/.claude/cli-tools.json` |
+| `Node version mismatch` | Update to Node.js >= 18.0.0 |
+

docs/guide/workflows.md

@@ -0,0 +1,209 @@
+# CCW Workflows
+
+Understanding and using CCW's workflow system for efficient development.
+
+## What are Workflows?
+
+CCW workflows are orchestrated sequences of tasks that guide a project from initial concept to completed implementation. They ensure consistency, quality, and proper documentation throughout the development lifecycle.
+
+## Workflow Levels
+
+CCW uses a 4-level workflow system:
+
+```
+Level 1: SPECIFICATION
+    ↓
+Level 2: PLANNING
+    ↓
+Level 3: IMPLEMENTATION
+    ↓
+Level 4: VALIDATION
+```
+
+## Using Workflows
+
+### Starting a Workflow
+
+Begin a new workflow with the team-lifecycle-v4 skill:
+
+```javascript
+Skill(skill="team-lifecycle-v4", args="Build user authentication system")
+```
+
+This creates a complete workflow:
+1. Specification phase (RESEARCH-001 through QUALITY-001)
+2. Planning phase (PLAN-001)
+3. Implementation phase (IMPL-001)
+4. Validation phase (TEST-001 and REVIEW-001)
+
+### Workflow Execution
+
+The workflow executes automatically:
+
+1. **Specification**: Analyst and writer agents research and document requirements
+2. **Checkpoint**: User reviews and approves specification
+3. **Planning**: Planner creates implementation plan with task breakdown
+4. **Implementation**: Executor writes code
+5. **Validation**: Tester and reviewer validate quality
+
+### Resume Workflow
+
+After a checkpoint, resume the workflow:
+
+```bash
+ccw workflow resume
+```
+
+## Workflow Tasks
+
+### Specification Tasks
+
+| Task | Agent | Output |
+|------|-------|--------|
+| RESEARCH-001 | analyst | Discovery context |
+| DRAFT-001 | writer | Product brief |
+| DRAFT-002 | writer | Requirements (PRD) |
+| DRAFT-003 | writer | Architecture design |
+| DRAFT-004 | writer | Epics & stories |
+| QUALITY-001 | reviewer | Readiness report |
+
+### Implementation Tasks
+
+| Task | Agent | Output |
+|------|-------|--------|
+| PLAN-001 | planner | Implementation plan |
+| IMPL-001 | executor | Source code |
+| TEST-001 | tester | Test results |
+| REVIEW-001 | reviewer | Code review |
+
+## Custom Workflows
+
+Create custom workflows for your team:
+
+```yaml
+# .ccw/workflows/feature-development.yaml
+name: Feature Development
+description: Standard workflow for new features
+
+levels:
+  - name: specification
+    tasks:
+      - type: research
+        agent: analyst
+      - type: document
+        agent: writer
+        documents: [prd, architecture]
+
+  - name: planning
+    tasks:
+      - type: plan
+        agent: planner
+
+  - name: implementation
+    tasks:
+      - type: implement
+        agent: executor
+      - type: test
+        agent: tester
+
+  - name: validation
+    tasks:
+      - type: review
+        agent: reviewer
+
+checkpoints:
+  - after: specification
+    action: await_user_approval
+  - after: validation
+    action: verify_quality_gates
+```
+
+## Workflow Configuration
+
+Configure workflow behavior in `~/.claude/workflows/config.json`:
+
+```json
+{
+  "defaults": {
+    "autoAdvance": true,
+    "checkpoints": ["specification", "implementation"],
+    "parallel": true
+  },
+  "agents": {
+    "analyst": {
+      "timeout": 300000,
+      "retries": 3
+    }
+  }
+}
+```
+
+## Best Practices
+
+### 1. Clear Requirements
+
+Start with clear, specific requirements:
+
+```javascript
+// Good: Specific
+Skill(skill="team-lifecycle-v4", args="Build JWT authentication with refresh tokens")
+
+// Bad: Vague
+Skill(skill="team-lifecycle-v4", args="Add auth")
+```
+
+### 2. Checkpoint Reviews
+
+Always review checkpoints:
+
+- Specification checkpoint: Validate requirements
+- Implementation checkpoint: Verify progress
+
+### 3. Feedback Loops
+
+Provide feedback during workflow:
+
+```bash
+# Add feedback during review
+ccw workflow feedback --task REVIEW-001 --message "Tests need more edge cases"
+```
+
+### 4. Monitor Progress
+
+Track workflow status:
+
+```bash
+# Check workflow status
+ccw workflow status
+
+# View task details
+ccw workflow task IMPL-001
+```
+
+## Troubleshooting
+
+### Stalled Workflow
+
+If a workflow stalls:
+
+```bash
+# Check for blocked tasks
+ccw workflow status --blocked
+
+# Reset stuck tasks
+ccw workflow reset --task IMPL-001
+```
+
+### Failed Tasks
+
+Retry failed tasks:
+
+```bash
+# Retry with new prompt
+ccw workflow retry --task IMPL-001 --prompt "Fix the TypeScript errors"
+```
+
+::: info See Also
+- [4-Level System](../workflows/4-level.md) - Detailed workflow explanation
+- [Best Practices](../workflows/best-practices.md) - Workflow optimization
+:::

docs/index.md

@@ -0,0 +1,7 @@
+---
+layout: page
+title: CCW Documentation
+titleTemplate: Claude Code Workspace
+---
+
+<ProfessionalHome lang="en" />