feat(docs): add full documentation generation phase and related documentation generation phase

- Implement Phase 4: Full Documentation Generation with multi-layered strategy and tool fallback. - Introduce Phase 5: Related Documentation Generation for incremental updates based on git changes. - Create new utility components for displaying execution status in the terminal panel. - Add helper functions for rendering execution status icons and formatting relative time. - Establish a recent paths configuration for improved path resolution.
2026-02-14 02:42:04 +08:00 · 2026-02-13 21:46:28 +08:00
parent d750290f84
commit 25e27286b4
21 changed files with 692 additions and 498 deletions
--- a/.claude/commands/memory/prepare.md
+++ b/.claude/commands/memory/prepare.md
@@ -1,18 +1,18 @@
 ---
-name: load
+name: prepare
 description: Delegate to universal-executor agent to analyze project via Gemini/Qwen CLI and return JSON core content package for task context
 argument-hint: "[--tool gemini|qwen] \"task context description\""
 allowed-tools: Task(*), Bash(*)
 examples:
-  - /memory:load "在当前前端基础上开发用户认证功能"
-  - /memory:load --tool qwen "重构支付模块API"
+  - /memory:prepare "在当前前端基础上开发用户认证功能"
+  - /memory:prepare --tool qwen "重构支付模块API"
 ---

-# Memory Load Command (/memory:load)
+# Memory Prepare Command (/memory:prepare)

 ## 1. Overview

-The `memory:load` command **delegates to a universal-executor agent** to analyze the project and return a structured "Core Content Pack". This pack is loaded into the main thread's memory, providing essential context for subsequent agent operations while minimizing token consumption.
+The `memory:prepare` command **delegates to a universal-executor agent** to analyze the project and return a structured "Core Content Pack". This pack is loaded into the main thread's memory, providing essential context for subsequent agent operations while minimizing token consumption.

 **Core Philosophy**:
 - **Agent-Driven**: Fully delegates execution to universal-executor agent
@@ -95,11 +95,11 @@ The command fully delegates to **universal-executor agent**, which autonomously:
 ```javascript
 Task(
  subagent_type="universal-executor",
-  description="Load project memory: ${task_description}",
+  description="Prepare project memory: ${task_description}",
  prompt=`
-## Mission: Load Project Memory Context
+## Mission: Prepare Project Memory Context

-**Task**: Load project memory context for: "${task_description}"
+**Task**: Prepare project memory context for: "${task_description}"
 **Mode**: analysis
 **Tool Preference**: ${tool || 'gemini'}

@@ -186,7 +186,7 @@ Before returning:
 ### Example 1: Load Context for New Feature

 ```bash
-/memory:load "在当前前端基础上开发用户认证功能"
+/memory:prepare "在当前前端基础上开发用户认证功能"
 ```

 **Agent Execution**:
@@ -212,7 +212,7 @@ Before returning:
 ### Example 2: Using Qwen Tool

 ```bash
-/memory:load --tool qwen "重构支付模块API"
+/memory:prepare --tool qwen "重构支付模块API"
 ```

 Agent uses Qwen CLI for analysis, returns same structured package.
@@ -220,7 +220,7 @@ Agent uses Qwen CLI for analysis, returns same structured package.
 ### Example 3: Bug Fix Context

 ```bash
-/memory:load "修复登录验证错误"
+/memory:prepare "修复登录验证错误"
 ```

 Returns core context related to login validation, including test files and validation logic.
@@ -229,7 +229,7 @@ Returns core context related to login validation, including test files and valid

 - **Session-Scoped**: Content package valid for current session
 - **Subsequent Reference**: All subsequent agents/commands can access
- **Reload Required**: New sessions need to re-execute /memory:load
+- **Reload Required**: New sessions need to re-execute /memory:prepare

 ## 8. Notes

--- a/.claude/skills/memory-manage/SKILL.md
+++ b/.claude/skills/memory-manage/SKILL.md
@@ -0,0 +1,229 @@
+---
+name: memory-manage
+description: Unified memory management - CLAUDE.md updates and documentation generation with interactive routing. Triggers on "memory manage", "update claude", "update memory", "generate docs", "更新记忆", "生成文档".
+allowed-tools: Task(*), Bash(*), AskUserQuestion(*), Read(*)
+---
+
+# Memory Management Skill
+
+Unified entry point for project memory (CLAUDE.md) updates and documentation (API.md/README.md) generation. Routes to specialized sub-commands via arguments or interactive needs assessment.
+
+## Architecture Overview
+
+```
+┌──────────────────────────────────────────────────────┐
+│  Memory Manage (Router)                               │
+│  → Parse input → Detect mode → Route to phase         │
+└───────────────┬──────────────────────────────────────┘
+                │
+        ┌───────┴───────┐
+        ↓               ↓
+  ┌───────────┐   ┌───────────┐
+  │ CLAUDE.md │   │  Docs     │
+  │  管理     │   │  生成     │
+  └─────┬─────┘   └─────┬─────┘
+        │               │
+   ┌────┼────┐     ┌────┴────┐
+   ↓    ↓    ↓     ↓         ↓
+ Full Related Single Full  Related
+ 全量  增量   单模块  全量    增量
+```
+
+## Execution Flow
+
+### Step 1: Parse Input & Route
+
+Detect execution mode from user input:
+
+**Auto-Route Rules** (priority order):
+
+| Signal | Route | Examples |
+|--------|-------|---------|
+| Explicit sub-command token | → Direct dispatch | `update-full --tool qwen` |
+| Keyword: full, 全量, all, entire + update/claude | → update-full | "全量更新claude" |
+| Keyword: related, changed, 增量, diff + update/claude | → update-related | "更新变更模块" |
+| Keyword: single, module, 单模块 + path-like token | → update-single | "更新 src/auth 模块" |
+| Keyword: docs, documentation, 文档 + full/all/全量 | → docs-full | "全量生成文档" |
+| Keyword: docs, documentation, 文档 + related/changed/增量 | → docs-related | "增量生成文档" |
+| Ambiguous or no arguments | → **AskUserQuestion** | `/memory-manage` |
+
+Valid sub-command tokens: `update-full`, `update-related`, `update-single`, `docs-full`, `docs-related`
+
+**Direct dispatch examples**:
+```
+input = "update-full --tool qwen"
+  → subcommand = "update-full", remainingArgs = "--tool qwen"
+  → Read phases/01-update-full.md, execute
+
+input = "update-single src/auth --tool gemini"
+  → subcommand = "update-single", remainingArgs = "src/auth --tool gemini"
+  → Read phases/03-update-single.md, execute
+```
+
+**When ambiguous or no arguments**, proceed to Step 2.
+
+### Step 2: Interactive Needs Assessment
+
+**Q1 — 确定内容类别**:
+
+```
+AskUserQuestion({
+  questions: [{
+    question: "你要管理哪类内容？",
+    header: "类别",
+    multiSelect: false,
+    options: [
+      {
+        label: "CLAUDE.md (项目记忆)",
+        description: "更新模块指令文件，帮助Claude理解代码库结构和约定"
+      },
+      {
+        label: "项目文档 (API+README)",
+        description: "生成API.md和README.md到.workflow/docs/目录"
+      }
+    ]
+  }]
+})
+```
+
+**Q2a — CLAUDE.md 管理范围** (用户选了 CLAUDE.md):
+
+```
+AskUserQuestion({
+  questions: [{
+    question: "CLAUDE.md 更新范围？",
+    header: "范围",
+    multiSelect: false,
+    options: [
+      {
+        label: "增量更新 (Recommended)",
+        description: "仅更新git变更模块及其父级，日常开发首选"
+      },
+      {
+        label: "全量更新",
+        description: "更新所有模块，3层架构bottom-up，适合重大重构后"
+      },
+      {
+        label: "单模块更新",
+        description: "指定一个模块，Explore深度分析后生成说明书式文档"
+      }
+    ]
+  }]
+})
+```
+
+Route mapping:
+- "增量更新" → `update-related` → Ref: phases/02-update-related.md
+- "全量更新" → `update-full` → Ref: phases/01-update-full.md
+- "单模块更新" → `update-single` → continue to Q3
+
+**Q2b — 文档生成范围** (用户选了项目文档):
+
+```
+AskUserQuestion({
+  questions: [{
+    question: "文档生成范围？",
+    header: "范围",
+    multiSelect: false,
+    options: [
+      {
+        label: "增量生成 (Recommended)",
+        description: "仅为git变更模块生成/更新文档，日常开发首选"
+      },
+      {
+        label: "全量生成",
+        description: "所有模块+项目级文档(ARCHITECTURE.md等)，适合初始化"
+      }
+    ]
+  }]
+})
+```
+
+Route mapping:
+- "增量生成" → `docs-related` → Ref: phases/05-docs-related.md
+- "全量生成" → `docs-full` → Ref: phases/04-docs-full.md
+
+**Q3 — 补充路径** (仅 update-single 需要，且无路径参数时):
+
+```
+AskUserQuestion({
+  questions: [{
+    question: "请指定要更新的模块路径：",
+    header: "路径",
+    multiSelect: false,
+    options: [
+      {
+        label: "src",
+        description: "src 目录"
+      },
+      {
+        label: ".",
+        description: "项目根目录"
+      }
+    ]
+  }]
+})
+```
+
+用户可选 "Other" 输入自定义路径。
+
+### Step 3: Execute Selected Phase
+
+Based on routing result, read and execute the corresponding phase:
+
+**Phase Reference Documents** (read on-demand when phase executes):
+
+| Mode | Document | Purpose |
+|------|----------|---------|
+| update-full | [phases/01-update-full.md](phases/01-update-full.md) | Full CLAUDE.md update, 3-layer architecture, batched agents |
+| update-related | [phases/02-update-related.md](phases/02-update-related.md) | Git-changed CLAUDE.md update, depth-first |
+| update-single | [phases/03-update-single.md](phases/03-update-single.md) | Single module CLAUDE.md, Explore + handbook style |
+| docs-full | [phases/04-docs-full.md](phases/04-docs-full.md) | Full API.md + README.md generation |
+| docs-related | [phases/05-docs-related.md](phases/05-docs-related.md) | Git-changed docs generation, incremental |
+
+## Shared Configuration
+
+### Tool Fallback Hierarchy
+
+All sub-commands share the same fallback:
+
+```
+--tool gemini  →  [gemini, qwen, codex]  // default
+--tool qwen    →  [qwen, gemini, codex]
+--tool codex   →  [codex, gemini, qwen]
+```
+
+### Common Parameters
+
+| Parameter | Description | Default | Used By |
+|-----------|-------------|---------|---------|
+| `--tool <gemini\|qwen\|codex>` | Primary CLI tool | gemini | All |
+| `--path <dir>` | Target directory | . | update-full, docs-full |
+| `<path>` (positional) | Module path | _(required)_ | update-single |
+
+### Batching Thresholds
+
+| Sub-Command | Direct Execution | Agent Batch | Batch Size |
+|-------------|-----------------|-------------|------------|
+| update-full | <20 modules | ≥20 modules | 4/agent |
+| update-related | <15 modules | ≥15 modules | 4/agent |
+| update-single | always single | N/A | 1 |
+| docs-full | <20 modules | ≥20 modules | 4/agent |
+| docs-related | <15 modules | ≥15 modules | 4/agent |
+
+## Core Rules
+
+1. **Route before execute**: Determine sub-command before any execution
+2. **Phase doc is truth**: All execution logic lives in phase docs, router only dispatches
+3. **Read on-demand**: Only read the selected phase doc, never load all
+4. **Pass arguments through**: Forward remaining args unchanged to sub-command
+5. **User confirmation**: Each sub-command handles its own plan presentation and y/n confirmation
+
+## Error Handling
+
+| Error | Resolution |
+|-------|------------|
+| Unknown sub-command | Fall through to interactive flow |
+| Phase doc not found | Abort with file path error |
+| Missing path for update-single | Prompt via Q3 |
+| Sub-command execution fails | Follow phase doc's own error handling |
--- a/.claude/skills/memory-manage/phases/01-update-full.md
+++ b/.claude/skills/memory-manage/phases/01-update-full.md
@@ -1,24 +1,26 @@
---
-name: update-full
-description: Update all CLAUDE.md files using layer-based execution (Layer 3→1) with batched agents (4 modules/agent) and gemini→qwen→codex fallback, <20 modules uses direct parallel
-argument-hint: "[--tool gemini|qwen|codex] [--path <directory>]"
---
+# Phase 1: Full CLAUDE.md Update (update-full)

-# Full Documentation Update (/memory:update-full)
+全项目 CLAUDE.md 更新，使用 3 层架构 (Layer 3→1)，批量 agent 执行，自动 tool fallback。

-## Overview
+## Objective

-Orchestrates project-wide CLAUDE.md updates using batched agent execution with automatic tool fallback and 3-layer architecture support.
+- 发现所有项目模块并按深度分层
+- 按 Layer 3→2→1 顺序 bottom-up 更新所有模块的 CLAUDE.md
+- <20 模块直接并行，≥20 模块使用 agent 批处理 (4 modules/agent)
+- 自动 tool fallback (gemini→qwen→codex)
+
+## Parameters

-**Parameters**:
 - `--tool <gemini|qwen|codex>`: Primary tool (default: gemini)
 - `--path <directory>`: Target specific directory (default: entire project)

+## Execution
+
 **Execution Flow**: Discovery → Plan Presentation → Execution → Safety Verification

-## 3-Layer Architecture & Auto-Strategy Selection
+### 3-Layer Architecture & Auto-Strategy Selection

-### Layer Definition & Strategy Assignment
+#### Layer Definition & Strategy Assignment

 | Layer | Depth | Strategy | Purpose | Context Pattern |
 |-------|-------|----------|---------|----------------|
@@ -30,20 +32,17 @@ Orchestrates project-wide CLAUDE.md updates using batched agent execution with a

 **Strategy Auto-Selection**: Strategies are automatically determined by directory depth - no user configuration needed.

-### Strategy Details
-
 #### Multi-Layer Strategy (Layer 3 Only)
 - **Use Case**: Deepest directories with unstructured file layouts
 - **Behavior**: Generates CLAUDE.md for current directory AND each subdirectory containing files
 - **Context**: All files in current directory tree (`@**/*`)

-
 #### Single-Layer Strategy (Layers 1-2)
 - **Use Case**: Upper layers that aggregate from existing documentation
 - **Behavior**: Generates CLAUDE.md only for current directory
 - **Context**: Direct children CLAUDE.md files + current directory code files

-### Example Flow
+#### Example Flow
 ```
 src/auth/handlers/ (depth 3) → MULTI-LAYER STRATEGY
  CONTEXT: @**/* (all files in handlers/ and subdirs)
@@ -62,7 +61,7 @@ src/ (depth 1) → SINGLE-LAYER STRATEGY
  GENERATES: ./CLAUDE.md only
 ```

-## Core Execution Rules
+### Core Execution Rules

 1. **Analyze First**: Git cache + module discovery before updates
 2. **Wait for Approval**: Present plan, no execution without user confirmation
@@ -74,7 +73,7 @@ src/ (depth 1) → SINGLE-LAYER STRATEGY
 6. **Safety Check**: Verify only CLAUDE.md files modified
 7. **Layer-based Grouping**: Group modules by LAYER (not depth) for execution

-## Tool Fallback Hierarchy
+### Tool Fallback Hierarchy

 ```javascript
 --tool gemini  →  [gemini, qwen, codex]  // default
@@ -90,9 +89,7 @@ src/ (depth 1) → SINGLE-LAYER STRATEGY
 | qwen   | Architecture, system design    | gemini → codex |
 | codex  | Implementation, code quality   | gemini → qwen  |

-## Execution Phases
-
-### Phase 1: Discovery & Analysis
+### Step 1.1: Discovery & Analysis

 ```javascript
 // Cache git changes
@@ -109,7 +106,7 @@ Bash({command: "cd <target-path> && ccw tool exec get_modules_by_depth '{\"forma

 **Smart filter**: Auto-detect and skip tests/build/config/docs based on project tech stack.

-### Phase 2: Plan Presentation
+### Step 1.2: Plan Presentation

 **For <20 modules**:
 ```
@@ -168,7 +165,7 @@ Update Plan:
  Confirm execution? (y/n)
 ```

-### Phase 3A: Direct Execution (<20 modules)
+### Step 1.3A: Direct Execution (<20 modules)

 **Strategy**: Parallel execution within layer (max 4 concurrent), no agent overhead.

@@ -202,7 +199,7 @@ for (let layer of [3, 2, 1]) {
 }
 ```

-### Phase 3B: Agent Batch Execution (≥20 modules)
+### Step 1.3B: Agent Batch Execution (≥20 modules)

 **Strategy**: Batch modules into groups of 4, spawn memory-bridge agents per batch.

@@ -284,7 +281,8 @@ REPORTING FORMAT:
    ⚠️  path/to/module failed with {tool}, trying next...
    ❌ FAILED: path/to/module - all tools exhausted
 ```
-### Phase 4: Safety Verification
+
+### Step 1.4: Safety Verification

 ```javascript
 // Check only CLAUDE.md files modified
@@ -308,25 +306,11 @@ Update Summary:
 **Coordinator**: Invalid path abort, user decline handling, safety check with auto-revert
 **Fallback Triggers**: Non-zero exit code, script timeout, unexpected output

-## Usage Examples
+## Output

-```bash
-# Full project update (auto-strategy selection)
-/memory:update-full
+- **Files**: Updated CLAUDE.md files across all project modules
+- **Report**: Summary with success/failure counts and tool usage statistics

-# Target specific directory
-/memory:update-full --path .claude
-/memory:update-full --path src/features/auth
+## Next Phase

-# Use specific tool
-/memory:update-full --tool qwen
-/memory:update-full --path .claude --tool qwen
-```
-
-## Key Advantages
-
- **Efficiency**: 30 modules → 8 agents (73% reduction from sequential)
- **Resilience**: 3-tier tool fallback per module
- **Performance**: Parallel batches, no concurrency limits
- **Observability**: Per-module tool usage, batch-level metrics
- **Automation**: Zero configuration - strategy auto-selected by directory depth
+Return to [manage.md](../manage.md) router.
--- a/.claude/skills/memory-manage/phases/02-update-related.md
+++ b/.claude/skills/memory-manage/phases/02-update-related.md
@@ -1,22 +1,23 @@
---
-name: update-related
-description: Update CLAUDE.md for git-changed modules using batched agent execution (4 modules/agent) with gemini→qwen→codex fallback, <15 modules uses direct execution
-argument-hint: "[--tool gemini|qwen|codex]"
---
+# Phase 2: Related CLAUDE.md Update (update-related)

-# Related Documentation Update (/memory:update-related)
+仅更新 git 变更模块的 CLAUDE.md，使用深度优先执行和自动 tool fallback。

-## Overview
+## Objective

-Orchestrates context-aware CLAUDE.md updates for changed modules using batched agent execution with automatic tool fallback (gemini→qwen→codex).
+- 检测 git 变更，识别受影响模块
+- 按深度 N→0 顺序更新变更模块及其父级上下文
+- <15 模块直接并行，≥15 模块使用 agent 批处理
+- 自动 tool fallback (gemini→qwen→codex)
+
+## Parameters

-**Parameters**:
 - `--tool <gemini|qwen|codex>`: Primary tool (default: gemini)

-**Execution Flow**:
-1. Change Detection → 2. Plan Presentation → 3. Batched Agent Execution → 4. Safety Verification
+## Execution

-## Core Rules
+**Execution Flow**: Change Detection → Plan Presentation → Batched Execution → Safety Verification
+
+### Core Rules

 1. **Detect Changes First**: Use git diff to identify affected modules
 2. **Wait for Approval**: Present plan, no execution without user confirmation
@@ -27,7 +28,7 @@ Orchestrates context-aware CLAUDE.md updates for changed modules using batched a
 5. **Depth Sequential**: Process depths N→0, parallel batches within depth (both modes)
 6. **Related Mode**: Update only changed modules and their parent contexts

-## Tool Fallback Hierarchy
+### Tool Fallback Hierarchy

 ```javascript
 --tool gemini  →  [gemini, qwen, codex]  // default
@@ -37,7 +38,7 @@ Orchestrates context-aware CLAUDE.md updates for changed modules using batched a

 **Trigger**: Non-zero exit code from update script

-## Phase 1: Change Detection & Analysis
+### Step 2.1: Change Detection & Analysis

 ```javascript
 // Detect changed modules
@@ -53,7 +54,7 @@ Bash({command: "git add -A 2>/dev/null || true", run_in_background: false});

 **Fallback**: If no changes detected, use recent modules (first 10 by depth).

-## Phase 2: Plan Presentation
+### Step 2.2: Plan Presentation

 **Present filtered plan**:
 ```
@@ -87,7 +88,7 @@ Related Update Plan:
 - <15 modules: Direct execution
 - ≥15 modules: Agent batch execution

-## Phase 3A: Direct Execution (<15 modules)
+### Step 2.3A: Direct Execution (<15 modules)

 **Strategy**: Parallel execution within depth (max 4 concurrent), no agent overhead.

@@ -119,11 +120,9 @@ for (let depth of sorted_depths.reverse()) {  // N → 0
 }
 ```

---
+### Step 2.3B: Agent Batch Execution (≥15 modules)

-## Phase 3B: Agent Batch Execution (≥15 modules)
-
-### Batching Strategy
+#### Batching Strategy

 ```javascript
 // Batch modules into groups of 4
@@ -137,7 +136,7 @@ function batch_modules(modules, batch_size = 4) {
 // Examples: 10→[4,4,2] | 8→[4,4] | 3→[3]
 ```

-### Coordinator Orchestration
+#### Coordinator Orchestration

 ```javascript
 let modules_by_depth = group_by_depth(changed_modules);
@@ -161,7 +160,7 @@ for (let depth of sorted_depths.reverse()) {  // N → 0
 }
 ```

-### Batch Worker Prompt Template
+#### Batch Worker Prompt Template

 ```
 PURPOSE: Update CLAUDE.md for assigned modules with tool fallback (related mode)
@@ -212,7 +211,7 @@ Report final summary with:
 - Tool usage: {{tool_1}}:X, {{tool_2}}:Y, {{tool_3}}:Z
 ```

-## Phase 4: Safety Verification
+### Step 2.4: Safety Verification

 ```javascript
 // Check only CLAUDE.md modified
@@ -240,16 +239,6 @@ Update Summary:
  4 files changed, 82 insertions(+), 6 deletions(-)
 ```

-## Execution Summary
-
-**Module Count Threshold**:
- **<15 modules**: Coordinator executes Phase 3A (Direct Execution)
- **≥15 modules**: Coordinator executes Phase 3B (Agent Batch Execution)
-
-**Agent Hierarchy** (for ≥15 modules):
- **Coordinator**: Handles batch division, spawns worker agents per depth
- **Worker Agents**: Each processes 4 modules with tool fallback (related mode)
-
 ## Error Handling

 **Batch Worker**:
@@ -268,65 +257,11 @@ Update Summary:
 - Script timeout
 - Unexpected output

-## Tool Reference
+## Output

-| Tool   | Best For                       | Fallback To    |
-|--------|--------------------------------|----------------|
-| gemini | Documentation, patterns        | qwen → codex   |
-| qwen   | Architecture, system design    | gemini → codex |
-| codex  | Implementation, code quality   | gemini → qwen  |
+- **Files**: Updated CLAUDE.md files for changed modules and their parents
+- **Report**: Summary with success/failure counts, tool usage, and git diff statistics

-## Usage Examples
+## Next Phase

-```bash
-# Daily development update
-/memory:update-related
-
-# After feature work with specific tool
-/memory:update-related --tool qwen
-
-# Code quality review after implementation
-/memory:update-related --tool codex
-```
-
-## Key Advantages
-
-**Efficiency**: 30 modules → 8 agents (73% reduction)
-**Resilience**: 3-tier fallback per module
-**Performance**: Parallel batches, no concurrency limits
-**Context-aware**: Updates based on actual git changes
-**Fast**: Only affected modules, not entire project
-
-## Coordinator Checklist
-
- Parse `--tool` (default: gemini)
- Refresh code index for accurate change detection
- Detect changed modules via detect_changed_modules.sh
- **Smart filter modules** (auto-detect tech stack, skip tests/build/config/docs)
- Cache git changes
- Apply fallback if no changes (recent 10 modules)
- Construct tool fallback order
- **Present filtered plan** with skip reasons and change types
- **Wait for y/n confirmation**
- Determine execution mode:
-  - **<15 modules**: Direct execution (Phase 3A)
-    - For each depth (N→0): Sequential module updates with tool fallback
-  - **≥15 modules**: Agent batch execution (Phase 3B)
-    - For each depth (N→0): Batch modules (4 per batch), spawn batch workers in parallel
- Wait for depth/batch completion
- Aggregate results
- Safety check (only CLAUDE.md modified)
- Display git diff statistics + summary
-
-## Comparison with Full Update
-
-| Aspect | Related Update | Full Update |
-|--------|----------------|-------------|
-| **Scope** | Changed modules only | All project modules |
-| **Speed** | Fast (minutes) | Slower (10-30 min) |
-| **Use case** | Daily development | Major refactoring |
-| **Mode** | `"related"` | `"full"` |
-| **Trigger** | After commits | After major changes |
-| **Batching** | 4 modules/agent | 4 modules/agent |
-| **Fallback** | gemini→qwen→codex | gemini→qwen→codex |
-| **Complexity threshold** | ≤15 modules | ≤20 modules |
+Return to [manage.md](../manage.md) router.
--- a/.claude/skills/memory-manage/phases/03-update-single.md
+++ b/.claude/skills/memory-manage/phases/03-update-single.md
@@ -1,81 +1,47 @@
---
-name: update-single
-description: Update single module CLAUDE.md using Explore agent for deep codebase understanding, producing manual-style documentation (handbook, not API reference)
-argument-hint: "<path> [--tool gemini|qwen|codex]"
-allowed-tools: Task(*), Bash(*), AskUserQuestion(*)
---
+# Phase 3: Single Module CLAUDE.md Update (update-single)

-# Single Module Documentation Update (/memory:update-single)
+使用 Explore agent 深度分析后，为单个目标模块生成手册式 (说明书) CLAUDE.md。

-## Overview
+## Objective

-Generates a manual-style CLAUDE.md for a single target directory using Explore agent for deep semantic codebase understanding. The output reads like a module handbook — explaining what it does, how to use it, and how it integrates — rather than dry API documentation.
-
-**Core capabilities:**
- Explore agent for semantic codebase exploration (not just file scanning)
- Manual/handbook-style output (usage guide, not reference docs)
- Interactive confirmation with exploration summary preview
+- 验证目标路径并扫描结构
+- 使用 Explore agent 进行 7 维度深度探索
+- 交互确认后通过 CLI tool 生成手册式 CLAUDE.md
 - Tool fallback (gemini→qwen→codex)

-## Usage
+## Parameters

-```bash
-/memory:update-single <path> [--tool gemini|qwen|codex]
+- `<path>`: Target directory path (required)
+- `--tool <gemini|qwen|codex>`: Primary CLI tool (default: gemini)

-# Arguments
-<path>                         Target directory path (required)
-
-# Options
--tool <gemini|qwen|codex>     Primary CLI tool (default: gemini)
-
-# Examples
-/memory:update-single src/auth
-/memory:update-single .claude/commands --tool qwen
-/memory:update-single ccw/frontend/src/components/issue
-```
-
-## Output Artifacts
-
-| Artifact | Description |
-|----------|-------------|
-| `<path>/CLAUDE.md` | Manual-style module handbook |
-
-**CLAUDE.md Style** — "说明书" not "文档":
- What this module does (purpose & responsibility)
- How to use it (patterns, conventions, examples)
- How it integrates (dependencies, exports, data flow)
- Important constraints (gotchas, rules, limitations)
-
-## Execution Process
+## Execution

 ```
-Phase 1: Target Validation & Scan
+Step 3.1: Target Validation & Scan
   ├─ Parse arguments (path, --tool)
   ├─ Validate target directory exists
   └─ Quick structure scan (file count, types, depth)

-Phase 2: Deep Exploration (Explore Agent)
+Step 3.2: Deep Exploration (Explore Agent)
   ├─ Launch Explore agent with "very thorough" level
   ├─ Analyze purpose, structure, patterns, exports, dependencies
   └─ Build comprehensive module understanding

-Phase 3: Confirmation
+Step 3.3: Confirmation
   ├─ Display exploration summary (key findings)
   └─ AskUserQuestion: Generate / Cancel

-Phase 4: Generate CLAUDE.md (CLI Tool)
+Step 3.4: Generate CLAUDE.md (CLI Tool)
   ├─ Construct manual-style prompt from exploration results
   ├─ Execute ccw cli with --mode write
   ├─ Tool fallback on failure
   └─ Write to <path>/CLAUDE.md

-Phase 5: Verification
+Step 3.5: Verification
   └─ Display generated CLAUDE.md preview + stats
 ```

-## Implementation
-
-### Phase 1: Target Validation & Scan
+### Step 3.1: Target Validation & Scan

 ```javascript
 // Parse arguments
@@ -86,7 +52,7 @@ const primaryTool = toolFlagIdx !== -1 ? parts[toolFlagIdx + 1] : 'gemini'
 const targetPath = parts.find(p => !p.startsWith('--') && p !== primaryTool)

 if (!targetPath) {
-  console.log('ERROR: <path> is required. Usage: /memory:update-single <path> [--tool gemini|qwen|codex]')
+  console.log('ERROR: <path> is required. Usage: /memory:manage update-single <path> [--tool gemini|qwen|codex]')
  return
 }

@@ -112,9 +78,9 @@ Launching deep exploration...
 `)
 ```

-### Phase 2: Deep Exploration (Explore Agent)
+### Step 3.2: Deep Exploration (Explore Agent)

-**⚠️ CRITICAL**: Use `run_in_background: false` — exploration results are REQUIRED before generation.
+**CRITICAL**: Use `run_in_background: false` — exploration results are REQUIRED before generation.

 ```javascript
 const explorationResult = Task(
@@ -173,7 +139,7 @@ Return a structured summary covering all 7 dimensions above. Include specific fi
 )
 ```

-### Phase 3: Confirmation
+### Step 3.3: Confirmation

 ```javascript
 console.log(`
@@ -203,7 +169,7 @@ AskUserQuestion({
 // Cancel → abort
 ```

-### Phase 4: Generate CLAUDE.md (CLI Tool)
+### Step 3.4: Generate CLAUDE.md (CLI Tool)

 **Tool fallback hierarchy**:
 ```javascript
@@ -285,7 +251,7 @@ CONSTRAINTS: Only write CLAUDE.md, no other files" --tool ${tool} --mode write -
 }
 ```

-### Phase 5: Verification
+### Step 3.5: Verification

 ```javascript
 // Check file was created/updated
@@ -311,8 +277,8 @@ console.log(`

 The generated CLAUDE.md is a **说明书 (handbook)**, NOT a reference doc:

-| Aspect | Handbook Style ✅ | Reference Doc Style ❌ |
-|--------|-------------------|----------------------|
+| Aspect | Handbook Style | Reference Doc Style |
+|--------|---------------|---------------------|
 | Purpose | "This module handles user auth" | "Authentication module" |
 | Content | How to work with it | What every function does |
 | Patterns | "Always use `createAuthMiddleware()`" | "List of all exports" |
@@ -330,18 +296,11 @@ The generated CLAUDE.md is a **说明书 (handbook)**, NOT a reference doc:
 | Empty directory | Abort — nothing to document |
 | Existing CLAUDE.md | Overwrite entirely (full regeneration) |

-## Usage Examples
+## Output

-```bash
-# Generate handbook for a module
-/memory:update-single src/auth
+- **File**: `<path>/CLAUDE.md` — Manual-style module handbook
+- **Preview**: First 30 lines displayed after generation

-# Use specific tool
-/memory:update-single .claude/commands --tool qwen
+## Next Phase

-# Deep nested module
-/memory:update-single ccw/frontend/src/components/issue
-
-# Root-level documentation
-/memory:update-single .
-```
+Return to [manage.md](../manage.md) router.
--- a/.claude/skills/memory-manage/phases/04-docs-full.md
+++ b/.claude/skills/memory-manage/phases/04-docs-full.md
@@ -1,24 +1,27 @@
---
-name: docs-full-cli
-description: Generate full project documentation using CLI execution (Layer 3→1) with batched agents (4 modules/agent) and gemini→qwen→codex fallback, <20 modules uses direct parallel
-argument-hint: "[path] [--tool <gemini|qwen|codex>]"
---
+# Phase 4: Full Documentation Generation (docs-full)

-# Full Documentation Generation - CLI Mode (/memory:docs-full-cli)
+全项目 API.md + README.md 文档生成，使用 CLI 执行、3 层架构和自动 tool fallback，输出到 .workflow/docs/。

-## Overview
+## Objective

-Orchestrates project-wide documentation generation using CLI-based execution with batched agents and automatic tool fallback.
+- 发现所有项目模块并按深度分层
+- 按 Layer 3→2→1 顺序生成 API.md + README.md 文档
+- <20 模块直接并行，≥20 模块使用 agent 批处理
+- 生成项目级文档 (README.md, ARCHITECTURE.md, EXAMPLES.md)
+- 自动 tool fallback (gemini→qwen→codex)
+
+## Parameters

-**Parameters**:
 - `path`: Target directory (default: current directory)
 - `--tool <gemini|qwen|codex>`: Primary tool (default: gemini)

-**Execution Flow**: Discovery → Plan Presentation → Execution → Verification
+## Execution

-## 3-Layer Architecture & Auto-Strategy Selection
+**Execution Flow**: Discovery → Plan Presentation → Execution → Project-Level Docs → Verification

-### Layer Definition & Strategy Assignment
+### 3-Layer Architecture & Auto-Strategy Selection
+
+#### Layer Definition & Strategy Assignment

 | Layer | Depth | Strategy | Purpose | Context Pattern |
 |-------|-------|----------|---------|----------------|
@@ -28,24 +31,19 @@ Orchestrates project-wide documentation generation using CLI-based execution wit

 **Generation Direction**: Layer 3 → Layer 2 → Layer 1 (bottom-up dependency flow)

-**Strategy Auto-Selection**: Strategies are automatically determined by directory depth - no user configuration needed.
-
-### Strategy Details
-
 #### Full Strategy (Layer 3 Only)
 - **Use Case**: Deepest directories with comprehensive file coverage
 - **Behavior**: Generates API.md + README.md for current directory AND subdirectories containing code
 - **Context**: All files in current directory tree (`@**/*`)
 - **Output**: `.workflow/docs/{project_name}/{path}/API.md` + `README.md`

-
 #### Single Strategy (Layers 1-2)
 - **Use Case**: Upper layers that aggregate from existing documentation
 - **Behavior**: Generates API.md + README.md only in current directory
 - **Context**: Direct children docs + current directory code files
 - **Output**: `.workflow/docs/{project_name}/{path}/API.md` + `README.md`

-### Example Flow
+#### Example Flow
 ```
 src/auth/handlers/ (depth 3) → FULL STRATEGY
  CONTEXT: @**/* (all files in handlers/ and subdirs)
@@ -64,7 +62,7 @@ src/ (depth 1) → SINGLE STRATEGY
  GENERATES: .workflow/docs/project/{API.md,README.md} only
 ```

-## Core Execution Rules
+### Core Execution Rules

 1. **Analyze First**: Module discovery + folder classification before generation
 2. **Wait for Approval**: Present plan, no execution without user confirmation
@@ -76,7 +74,7 @@ src/ (depth 1) → SINGLE STRATEGY
 6. **Safety Check**: Verify only docs files modified in .workflow/docs/
 7. **Layer-based Grouping**: Group modules by LAYER (not depth) for execution

-## Tool Fallback Hierarchy
+### Tool Fallback Hierarchy

 ```javascript
 --tool gemini  →  [gemini, qwen, codex]  // default
@@ -84,17 +82,7 @@ src/ (depth 1) → SINGLE STRATEGY
 --tool codex   →  [codex, gemini, qwen]
 ```

-**Trigger**: Non-zero exit code from generation script
-
-| Tool   | Best For                       | Fallback To    |
-|--------|--------------------------------|----------------|
-| gemini | Documentation, patterns        | qwen → codex   |
-| qwen   | Architecture, system design    | gemini → codex |
-| codex  | Implementation, code quality   | gemini → qwen  |
-
-## Execution Phases
-
-### Phase 1: Discovery & Analysis
+### Step 4.1: Discovery & Analysis

 ```javascript
 // Get project metadata
@@ -111,7 +99,7 @@ Bash({command: "cd <target-path> && ccw tool exec get_modules_by_depth '{\"forma

 **Smart filter**: Auto-detect and skip tests/build/config/vendor based on project tech stack.

-### Phase 2: Plan Presentation
+### Step 4.2: Plan Presentation

 **For <20 modules**:
 ```
@@ -139,7 +127,6 @@ Documentation Generation Plan:

  Auto-skipped: ./tests, __pycache__, node_modules (15 paths)
  Execution order: Layer 2 → Layer 1
-  Estimated time: ~5-10 minutes

  Confirm execution? (y/n)
 ```
@@ -176,12 +163,10 @@ Documentation Generation Plan:
  - Layer 2 (15 modules, depth 1-2): 4 agents [4, 4, 4, 3]
  - Layer 1 (2 modules, depth 0): 1 agent [2]

-  Estimated time: ~15-25 minutes
-
  Confirm execution? (y/n)
 ```

-### Phase 3A: Direct Execution (<20 modules)
+### Step 4.3A: Direct Execution (<20 modules)

 **Strategy**: Parallel execution within layer (max 4 concurrent), no agent overhead.

@@ -217,7 +202,7 @@ for (let layer of [3, 2, 1]) {
 }
 ```

-### Phase 3B: Agent Batch Execution (≥20 modules)
+### Step 4.3B: Agent Batch Execution (≥20 modules)

 **Strategy**: Batch modules into groups of 4, spawn memory-bridge agents per batch.

@@ -290,7 +275,6 @@ EXECUTION FLOW (for each module):
  2. Handle complete failure (all tools failed):
     if [ $exit_code -ne 0 ]; then
       report "❌ FAILED: {{module_path}} - all tools exhausted"
-       # Continue to next module (do not abort batch)
     fi

 FOLDER TYPE HANDLING:
@@ -310,7 +294,7 @@ REPORTING FORMAT:
    ❌ FAILED: path/to/module - all tools exhausted
 ```

-### Phase 4: Project-Level Documentation
+### Step 4.4: Project-Level Documentation

 **After all module documentation is generated, create project-level documentation files.**

@@ -370,7 +354,7 @@ Project-Level Documentation:
  ✅ api/README.md (HTTP API reference) [optional]
 ```

-### Phase 5: Verification
+### Step 4.5: Verification

 ```javascript
 // Check documentation files created
@@ -398,12 +382,6 @@ Documentation Generation Summary:
    └── README.md
 ```

-## Error Handling
-
-**Batch Worker**: Tool fallback per module, batch isolation, clear status reporting
-**Coordinator**: Invalid path abort, user decline handling, verification with cleanup
-**Fallback Triggers**: Non-zero exit code, script timeout, unexpected output
-
 ## Output Structure

 ```
@@ -426,36 +404,18 @@ Documentation Generation Summary:
 │   └── core/
 │       ├── API.md
 │       └── README.md
-├── README.md                      # ✨ Project root overview (auto-generated)
-├── ARCHITECTURE.md                # ✨ System design (auto-generated)
-├── EXAMPLES.md                    # ✨ Usage examples (auto-generated)
-└── api/                           # ✨ Optional (auto-generated if HTTP API detected)
+├── README.md                      # Project root overview (auto-generated)
+├── ARCHITECTURE.md                # System design (auto-generated)
+├── EXAMPLES.md                    # Usage examples (auto-generated)
+└── api/                           # Optional (auto-generated if HTTP API detected)
    └── README.md                  # HTTP API reference
 ```

-## Usage Examples
+## Error Handling

-```bash
-# Full project documentation generation
-/memory:docs-full-cli
-
-# Target specific directory
-/memory:docs-full-cli src/features/auth
-/memory:docs-full-cli .claude
-
-# Use specific tool
-/memory:docs-full-cli --tool qwen
-/memory:docs-full-cli src --tool qwen
-```
-
-## Key Advantages
-
- **Efficiency**: 30 modules → 8 agents (73% reduction from sequential)
- **Resilience**: 3-tier tool fallback per module
- **Performance**: Parallel batches, no concurrency limits
- **Observability**: Per-module tool usage, batch-level metrics
- **Automation**: Zero configuration - strategy auto-selected by directory depth
- **Path Mirroring**: Clear 1:1 mapping between source and documentation structure
+**Batch Worker**: Tool fallback per module, batch isolation, clear status reporting
+**Coordinator**: Invalid path abort, user decline handling, verification with cleanup
+**Fallback Triggers**: Non-zero exit code, script timeout, unexpected output

 ## Template Reference

@@ -464,8 +424,12 @@ Templates used from `~/.ccw/workflows/cli-templates/prompts/documentation/`:
 - `module-readme.txt`: Module purpose, usage, dependencies
 - `folder-navigation.txt`: Navigation README for folders with subdirectories

-## Related Commands
+## Output

- `/memory:docs` - Agent-based documentation planning workflow
- `/memory:docs-related-cli` - Update docs for changed modules only
- `/workflow:execute` - Execute documentation tasks (when using agent mode)
+- **Directory**: `.workflow/docs/{project_name}/` — Complete documentation tree
+- **Project-Level**: README.md, ARCHITECTURE.md, EXAMPLES.md, api/README.md (optional)
+- **Report**: Summary with success/failure counts and tool usage statistics
+
+## Next Phase
+
+Return to [manage.md](../manage.md) router.
--- a/.claude/skills/memory-manage/phases/05-docs-related.md
+++ b/.claude/skills/memory-manage/phases/05-docs-related.md
@@ -1,22 +1,24 @@
---
-name: docs-related-cli
-description: Generate/update documentation for git-changed modules using CLI execution with batched agents (4 modules/agent) and gemini→qwen→codex fallback, <15 modules uses direct parallel
-argument-hint: "[--tool <gemini|qwen|codex>]"
---
+# Phase 5: Related Documentation Generation (docs-related)

-# Related Documentation Generation - CLI Mode (/memory:docs-related-cli)
+仅为 git 变更模块生成/更新 API.md + README.md 文档，使用增量策略和自动 tool fallback。

-## Overview
+## Objective

-Orchestrates context-aware documentation generation/update for changed modules using CLI-based execution with batched agents and automatic tool fallback (gemini→qwen→codex).
+- 检测 git 变更，识别受影响模块
+- 按深度 N→0 顺序为变更模块生成/更新文档
+- <15 模块直接并行，≥15 模块使用 agent 批处理
+- 使用 single 策略进行增量更新
+- 自动 tool fallback (gemini→qwen→codex)
+
+## Parameters

-**Parameters**:
 - `--tool <gemini|qwen|codex>`: Primary tool (default: gemini)

-**Execution Flow**:
-1. Change Detection → 2. Plan Presentation → 3. Batched Execution → 4. Verification
+## Execution

-## Core Rules
+**Execution Flow**: Change Detection → Plan Presentation → Batched Execution → Verification
+
+### Core Rules

 1. **Detect Changes First**: Use git diff to identify affected modules
 2. **Wait for Approval**: Present plan, no execution without user confirmation
@@ -28,7 +30,7 @@ Orchestrates context-aware documentation generation/update for changed modules u
 6. **Related Mode**: Generate/update only changed modules and their parent contexts
 7. **Single Strategy**: Always use `single` strategy (incremental update)

-## Tool Fallback Hierarchy
+### Tool Fallback Hierarchy

 ```javascript
 --tool gemini  →  [gemini, qwen, codex]  // default
@@ -36,15 +38,7 @@ Orchestrates context-aware documentation generation/update for changed modules u
 --tool codex   →  [codex, gemini, qwen]
 ```

-**Trigger**: Non-zero exit code from generation script
-
-| Tool   | Best For                       | Fallback To    |
-|--------|--------------------------------|----------------|
-| gemini | Documentation, patterns        | qwen → codex   |
-| qwen   | Architecture, system design    | gemini → codex |
-| codex  | Implementation, code quality   | gemini → qwen  |
-
-## Phase 1: Change Detection & Analysis
+### Step 5.1: Change Detection & Analysis

 ```javascript
 // Get project metadata
@@ -63,7 +57,7 @@ Bash({command: "git add -A 2>/dev/null || true", run_in_background: false});

 **Fallback**: If no changes detected, use recent modules (first 10 by depth).

-## Phase 2: Plan Presentation
+### Step 5.2: Plan Presentation

 **Present filtered plan**:
 ```
@@ -95,8 +89,6 @@ Related Documentation Generation Plan:
  - Depth 1 (1 module): 1 agent [1]
  - Depth 0 (1 module): 1 agent [1]

-  Estimated time: ~5-10 minutes
-
  Confirm execution? (y/n)
 ```

@@ -106,7 +98,7 @@ Related Documentation Generation Plan:
 - <15 modules: Direct execution
 - ≥15 modules: Agent batch execution

-## Phase 3A: Direct Execution (<15 modules)
+### Step 5.3A: Direct Execution (<15 modules)

 **Strategy**: Parallel execution within depth (max 4 concurrent), no agent overhead.

@@ -140,9 +132,9 @@ for (let depth of sorted_depths.reverse()) {  // N → 0
 }
 ```

-## Phase 3B: Agent Batch Execution (≥15 modules)
+### Step 5.3B: Agent Batch Execution (≥15 modules)

-### Batching Strategy
+#### Batching Strategy

 ```javascript
 // Batch modules into groups of 4
@@ -156,7 +148,7 @@ function batch_modules(modules, batch_size = 4) {
 // Examples: 10→[4,4,2] | 8→[4,4] | 3→[3]
 ```

-### Coordinator Orchestration
+#### Coordinator Orchestration

 ```javascript
 let modules_by_depth = group_by_depth(changed_modules);
@@ -181,7 +173,7 @@ for (let depth of sorted_depths.reverse()) {  // N → 0
 }
 ```

-### Batch Worker Prompt Template
+#### Batch Worker Prompt Template

 ```
 PURPOSE: Generate/update documentation for assigned modules with tool fallback (related mode)
@@ -239,7 +231,7 @@ Report final summary with:
 - Tool usage: {{tool_1}}:X, {{tool_2}}:Y, {{tool_3}}:Z
 ```

-## Phase 4: Verification
+### Step 5.4: Verification

 ```javascript
 // Check documentation files created/updated
@@ -270,15 +262,23 @@ Documentation Generation Summary:
  .workflow/docs/myproject/README.md                (updated)
 ```

-## Execution Summary
+## Output Structure

-**Module Count Threshold**:
- **<15 modules**: Coordinator executes Phase 3A (Direct Execution)
- **≥15 modules**: Coordinator executes Phase 3B (Agent Batch Execution)
-
-**Agent Hierarchy** (for ≥15 modules):
- **Coordinator**: Handles batch division, spawns worker agents per depth
- **Worker Agents**: Each processes 4 modules with tool fallback (related mode)
+```
+.workflow/docs/{project_name}/
+├── src/                           # Mirrors source structure
+│   ├── modules/
+│   │   ├── README.md
+│   │   ├── auth/
+│   │   │   ├── API.md             # Updated based on code changes
+│   │   │   └── README.md          # Updated based on code changes
+│   │   └── api/
+│   │       ├── API.md
+│   │       └── README.md
+│   └── utils/
+│       └── README.md
+└── README.md
+```

 ## Error Handling

@@ -298,80 +298,6 @@ Documentation Generation Summary:
 - Script timeout
 - Unexpected output

-## Output Structure
-
-```
-.workflow/docs/{project_name}/
-├── src/                           # Mirrors source structure
-│   ├── modules/
-│   │   ├── README.md
-│   │   ├── auth/
-│   │   │   ├── API.md             # Updated based on code changes
-│   │   │   └── README.md          # Updated based on code changes
-│   │   └── api/
-│   │       ├── API.md
-│   │       └── README.md
-│   └── utils/
-│       └── README.md
-└── README.md
-```
-
-## Usage Examples
-
-```bash
-# Daily development documentation update
-/memory:docs-related-cli
-
-# After feature work with specific tool
-/memory:docs-related-cli --tool qwen
-
-# Code quality documentation review after implementation
-/memory:docs-related-cli --tool codex
-```
-
-## Key Advantages
-
-**Efficiency**: 30 modules → 8 agents (73% reduction)
-**Resilience**: 3-tier fallback per module
-**Performance**: Parallel batches, no concurrency limits
-**Context-aware**: Updates based on actual git changes
-**Fast**: Only affected modules, not entire project
-**Incremental**: Single strategy for focused updates
-
-## Coordinator Checklist
-
- Parse `--tool` (default: gemini)
- Get project metadata (name, root)
- Detect changed modules via detect_changed_modules.sh
- **Smart filter modules** (auto-detect tech stack, skip tests/build/config/vendor)
- Cache git changes
- Apply fallback if no changes (recent 10 modules)
- Construct tool fallback order
- **Present filtered plan** with skip reasons and change types
- **Wait for y/n confirmation**
- Determine execution mode:
-  - **<15 modules**: Direct execution (Phase 3A)
-    - For each depth (N→0): Sequential module updates with tool fallback
-  - **≥15 modules**: Agent batch execution (Phase 3B)
-    - For each depth (N→0): Batch modules (4 per batch), spawn batch workers in parallel
- Wait for depth/batch completion
- Aggregate results
- Verification check (documentation files created/updated)
- Display summary + recent changes
-
-## Comparison with Full Documentation Generation
-
-| Aspect | Related Generation | Full Generation |
-|--------|-------------------|-----------------|
-| **Scope** | Changed modules only | All project modules |
-| **Speed** | Fast (minutes) | Slower (10-30 min) |
-| **Use case** | Daily development | Initial setup, major refactoring |
-| **Strategy** | `single` (all) | `full` (L3) + `single` (L1-2) |
-| **Trigger** | After commits | After setup or major changes |
-| **Batching** | 4 modules/agent | 4 modules/agent |
-| **Fallback** | gemini→qwen→codex | gemini→qwen→codex |
-| **Complexity threshold** | ≤15 modules | ≤20 modules |
-
 ## Template Reference

 Templates used from `~/.ccw/workflows/cli-templates/prompts/documentation/`:
@@ -379,8 +305,11 @@ Templates used from `~/.ccw/workflows/cli-templates/prompts/documentation/`:
 - `module-readme.txt`: Module purpose, usage, dependencies
 - `folder-navigation.txt`: Navigation README for folders

-## Related Commands
+## Output

- `/memory:docs-full-cli` - Full project documentation generation
- `/memory:docs` - Agent-based documentation planning workflow
- `/memory:update-related` - Update CLAUDE.md for changed modules
+- **Directory**: `.workflow/docs/{project_name}/` — Updated documentation for changed modules
+- **Report**: Summary with success/failure counts, tool usage, and file change list
+
+## Next Phase
+
+Return to [manage.md](../manage.md) router.
--- a/ccw/.tmp-ccw-path-resolver-home/config/recent-paths.json
+++ b/ccw/.tmp-ccw-path-resolver-home/config/recent-paths.json
@@ -0,0 +1,5 @@
+{
+  "paths": [
+    "C:/project/demo"
+  ]
+}
--- a/ccw/frontend/src/components/issue/hub/ExecutionPanel.tsx
+++ b/ccw/frontend/src/components/issue/hub/ExecutionPanel.tsx
@@ -11,9 +11,7 @@ import {
  Play,
  CheckCircle,
  XCircle,
-  Clock,
  Terminal,
-  Loader2,
 } from 'lucide-react';
 import { Card } from '@/components/ui/Card';
 import { Badge } from '@/components/ui/Badge';
@@ -23,50 +21,9 @@ import {
  selectExecutionStats,
  useTerminalPanelStore,
 } from '@/stores';
-import type { QueueExecution, QueueExecutionStatus } from '@/stores/queueExecutionStore';
+import type { QueueExecution } from '@/stores/queueExecutionStore';
 import { cn } from '@/lib/utils';
-
-// ========== Helpers ==========
-
-function statusBadgeVariant(status: QueueExecutionStatus): 'info' | 'success' | 'destructive' | 'secondary' {
-  switch (status) {
-    case 'running':
-      return 'info';
-    case 'completed':
-      return 'success';
-    case 'failed':
-      return 'destructive';
-    case 'pending':
-    default:
-      return 'secondary';
-  }
-}
-
-function statusIcon(status: QueueExecutionStatus) {
-  switch (status) {
-    case 'running':
-      return <Loader2 className="w-3.5 h-3.5 animate-spin" />;
-    case 'completed':
-      return <CheckCircle className="w-3.5 h-3.5" />;
-    case 'failed':
-      return <XCircle className="w-3.5 h-3.5" />;
-    case 'pending':
-    default:
-      return <Clock className="w-3.5 h-3.5" />;
-  }
-}
-
-function formatRelativeTime(isoString: string): string {
-  const diff = Date.now() - new Date(isoString).getTime();
-  const seconds = Math.floor(diff / 1000);
-  if (seconds < 60) return `${seconds}s ago`;
-  const minutes = Math.floor(seconds / 60);
-  if (minutes < 60) return `${minutes}m ago`;
-  const hours = Math.floor(minutes / 60);
-  if (hours < 24) return `${hours}h ago`;
-  const days = Math.floor(hours / 24);
-  return `${days}d ago`;
-}
+import { statusIcon, statusBadgeVariant, formatRelativeTime } from '@/lib/execution-display-utils';

 // ========== Empty State ==========

--- a/ccw/frontend/src/components/issue/queue/QueueItemExecutor.tsx
+++ b/ccw/frontend/src/components/issue/queue/QueueItemExecutor.tsx
@@ -38,6 +38,7 @@ import {
 } from '@/components/shared/CliExecutionSettings';
 import { buildQueueItemContext } from '@/lib/queue-prompt';
 import { useQueueExecutionStore, type QueueExecution } from '@/stores/queueExecutionStore';
+import type { Flow } from '@/types/flow';

 // ---------------------------------------------------------------------------
 // Types
@@ -70,7 +71,7 @@ export function QueueItemExecutor({ item, className }: QueueItemExecutorProps) {
  // Resolve the parent issue for context building
  const { issues } = useIssues();
  const issue = useMemo(
-    () => issues.find((i) => i.id === item.issue_id) as any,
+    () => issues.find((i) => i.id === item.issue_id),
    [issues, item.issue_id]
  );

@@ -203,13 +204,13 @@ export function QueueItemExecutor({ item, className }: QueueItemExecutorProps) {
        throw new Error('Failed to create flow');
      }

-      // Hydrate Orchestrator stores
-      const flowDto = created.data as any;
+      // Hydrate Orchestrator stores -- convert OrchestratorFlowDto to Flow
+      const flowDto = created.data;
      const parsedVersion = parseInt(String(flowDto.version ?? '1'), 10);
-      const flowForStore = {
+      const flowForStore: Flow = {
        ...flowDto,
        version: Number.isFinite(parsedVersion) ? parsedVersion : 1,
-      } as any;
+      } as Flow;
      useFlowStore.getState().setCurrentFlow(flowForStore);

      // Execute the flow
--- a/ccw/frontend/src/components/mcp/CcwToolsMcpCard.tsx
+++ b/ccw/frontend/src/components/mcp/CcwToolsMcpCard.tsx
@@ -59,7 +59,7 @@ export interface CcwConfig {
  enabledTools: string[];
  projectRoot?: string;
  allowedDirs?: string;
-  disableSandbox?: boolean;
+  enableSandbox?: boolean;
 }

 /**
@@ -75,7 +75,7 @@ export interface CcwToolsMcpCardProps {
  /** Comma-separated list of allowed directories */
  allowedDirs?: string;
  /** Whether sandbox is disabled */
-  disableSandbox?: boolean;
+  enableSandbox?: boolean;
  /** Callback when a tool is toggled */
  onToggleTool: (tool: string, enabled: boolean) => void;
  /** Callback when configuration is updated */
@@ -110,7 +110,7 @@ export function CcwToolsMcpCard({
  enabledTools,
  projectRoot,
  allowedDirs,
-  disableSandbox,
+  enableSandbox,
  onToggleTool,
  onUpdateConfig,
  onInstall,
@@ -123,7 +123,7 @@ export function CcwToolsMcpCard({
  // Local state for config inputs
  const [projectRootInput, setProjectRootInput] = useState(projectRoot || '');
  const [allowedDirsInput, setAllowedDirsInput] = useState(allowedDirs || '');
-  const [disableSandboxInput, setDisableSandboxInput] = useState(disableSandbox || false);
+  const [enableSandboxInput, setEnableSandboxInput] = useState(enableSandbox || false);
  const [isExpanded, setIsExpanded] = useState(false);
  const [installScope, setInstallScope] = useState<'global' | 'project'>('global');

@@ -193,7 +193,7 @@ export function CcwToolsMcpCard({
    updateConfigMutation.mutate({
      projectRoot: projectRootInput || undefined,
      allowedDirs: allowedDirsInput || undefined,
-      disableSandbox: disableSandboxInput,
+      enableSandbox: enableSandboxInput,
    });
  };

@@ -387,22 +387,22 @@ export function CcwToolsMcpCard({
              </p>
            </div>

-            {/* Disable Sandbox */}
+            {/* Enable Sandbox */}
            <div className="flex items-center gap-2">
              <input
                type="checkbox"
-                id="ccw-disable-sandbox"
-                checked={disableSandboxInput}
-                onChange={(e) => setDisableSandboxInput(e.target.checked)}
+                id="ccw-enable-sandbox"
+                checked={enableSandboxInput}
+                onChange={(e) => setEnableSandboxInput(e.target.checked)}
                disabled={!isInstalled}
                className="w-4 h-4"
              />
              <label
-                htmlFor="ccw-disable-sandbox"
+                htmlFor="ccw-enable-sandbox"
                className="text-sm text-foreground flex items-center gap-1 cursor-pointer"
              >
                <Shield className="w-4 h-4" />
-                {formatMessage({ id: 'mcp.ccw.paths.disableSandbox' })}
+                {formatMessage({ id: 'mcp.ccw.paths.enableSandbox' })}
              </label>
            </div>

--- a/ccw/frontend/src/components/terminal-panel/QueueExecutionListView.tsx
+++ b/ccw/frontend/src/components/terminal-panel/QueueExecutionListView.tsx
@@ -0,0 +1,133 @@
+// ========================================
+// QueueExecutionListView Component
+// ========================================
+// Compact execution list for TerminalPanel queue view.
+// Subscribes to queueExecutionStore and renders execution entries
+// with status badges, tool/mode labels, and relative timestamps.
+// Click navigates to terminal view (session) or orchestrator page.
+
+import { useMemo } from 'react';
+import { useIntl } from 'react-intl';
+import { useNavigate } from 'react-router-dom';
+import { ClipboardList } from 'lucide-react';
+import { Badge } from '@/components/ui/Badge';
+import { cn } from '@/lib/utils';
+import { useQueueExecutionStore } from '@/stores/queueExecutionStore';
+import { useTerminalPanelStore } from '@/stores/terminalPanelStore';
+import type { QueueExecution } from '@/stores/queueExecutionStore';
+import { ROUTES } from '@/router';
+import { statusIcon, statusBadgeVariant, formatRelativeTime } from '@/lib/execution-display-utils';
+
+// ========== Execution Item ==========
+
+function QueueExecutionItem({
+  execution,
+  onClick,
+}: {
+  execution: QueueExecution;
+  onClick: () => void;
+}) {
+  const { formatMessage } = useIntl();
+
+  const typeLabel = execution.type === 'session'
+    ? formatMessage({ id: 'home.terminalPanel.queueView.session' })
+    : formatMessage({ id: 'home.terminalPanel.queueView.orchestrator' });
+
+  return (
+    <button
+      type="button"
+      className={cn(
+        'w-full text-left px-3 py-2.5 rounded-md transition-colors',
+        'hover:bg-muted/60 focus:outline-none focus:ring-1 focus:ring-primary/30'
+      )}
+      onClick={onClick}
+    >
+      <div className="flex items-center justify-between gap-2">
+        <div className="flex items-center gap-2 min-w-0">
+          {statusIcon(execution.status)}
+          <span className="text-sm font-medium font-mono text-foreground truncate">
+            {execution.queueItemId}
+          </span>
+        </div>
+        <Badge variant={statusBadgeVariant(execution.status)} className="shrink-0 text-[10px] px-1.5 py-0">
+          {formatMessage({ id: `home.terminalPanel.status.${execution.status}` })}
+        </Badge>
+      </div>
+      <div className="mt-1 flex items-center gap-2 text-xs text-muted-foreground">
+        <span className="font-mono">{execution.tool}</span>
+        <span className="text-border">|</span>
+        <span>{execution.mode}</span>
+        <span className="text-border">|</span>
+        <span>{typeLabel}</span>
+        <span className="ml-auto shrink-0">{formatRelativeTime(execution.startedAt)}</span>
+      </div>
+    </button>
+  );
+}
+
+// ========== Empty State ==========
+
+function QueueEmptyState() {
+  const { formatMessage } = useIntl();
+
+  return (
+    <div className="flex-1 flex items-center justify-center text-muted-foreground">
+      <div className="text-center">
+        <ClipboardList className="h-12 w-12 mx-auto mb-4 opacity-50" />
+        <p className="text-sm">{formatMessage({ id: 'home.terminalPanel.queueView.emptyTitle' })}</p>
+        <p className="text-xs mt-1">{formatMessage({ id: 'home.terminalPanel.queueView.emptyDesc' })}</p>
+      </div>
+    </div>
+  );
+}
+
+// ========== Main Component ==========
+
+export function QueueExecutionListView() {
+  const navigate = useNavigate();
+  const executions = useQueueExecutionStore((s) => s.executions);
+  const setPanelView = useTerminalPanelStore((s) => s.setPanelView);
+  const openTerminal = useTerminalPanelStore((s) => s.openTerminal);
+
+  // Sort: running first, then pending, then failed, then completed; within same status by startedAt desc
+  const sortedExecutions = useMemo(() => {
+    const all = Object.values(executions);
+    const statusOrder: Record<string, number> = {
+      running: 0,
+      pending: 1,
+      failed: 2,
+      completed: 3,
+    };
+    return all.sort((a, b) => {
+      const sa = statusOrder[a.status] ?? 4;
+      const sb = statusOrder[b.status] ?? 4;
+      if (sa !== sb) return sa - sb;
+      return new Date(b.startedAt).getTime() - new Date(a.startedAt).getTime();
+    });
+  }, [executions]);
+
+  const handleClick = (exec: QueueExecution) => {
+    if (exec.type === 'session' && exec.sessionKey) {
+      setPanelView('terminal');
+      openTerminal(exec.sessionKey);
+    } else {
+      navigate(ROUTES.ORCHESTRATOR);
+    }
+  };
+
+  if (sortedExecutions.length === 0) {
+    return <QueueEmptyState />;
+  }
+
+  return (
+    <div className="flex-1 flex flex-col min-h-0 overflow-y-auto p-2 space-y-0.5">
+      {sortedExecutions.map((exec) => (
+        <QueueExecutionItem
+          key={exec.id}
+          execution={exec}
+          onClick={() => handleClick(exec)}
+        />
+      ))}
+    </div>
+  );
+}
--- a/ccw/frontend/src/components/terminal-panel/TerminalMainArea.tsx
+++ b/ccw/frontend/src/components/terminal-panel/TerminalMainArea.tsx
@@ -201,8 +201,9 @@ export function TerminalMainArea({ onClose }: TerminalMainAreaProps) {
        category: 'user',
      }, projectPath || undefined);
      setPrompt('');
-    } catch {
-      // Error shown in terminal output
+    } catch (err) {
+      // Error shown in terminal output; log for DevTools debugging
+      console.error('[TerminalMainArea] executeInCliSession failed:', err);
    } finally {
      setIsExecuting(false);
    }
--- a/ccw/frontend/src/hooks/useCliSessionCore.ts
+++ b/ccw/frontend/src/hooks/useCliSessionCore.ts
@@ -22,6 +22,8 @@ export interface UseCliSessionCoreOptions {
  autoSelectLast?: boolean;
  /** Default resumeKey used when creating sessions via ensureSession/handleCreateSession. */
  resumeKey?: string;
+  /** Shell to use when creating new sessions. Defaults to 'bash'. */
+  preferredShell?: string;
  /** Additional createCliSession fields (cols, rows, tool, model). */
  createSessionDefaults?: {
    cols?: number;
@@ -60,6 +62,7 @@ export function useCliSessionCore(options: UseCliSessionCoreOptions = {}): UseCl
  const {
    autoSelectLast = true,
    resumeKey,
+    preferredShell = 'bash',
    createSessionDefaults,
  } = options;

@@ -92,7 +95,7 @@ export function useCliSessionCore(options: UseCliSessionCoreOptions = {}): UseCl
    setError(null);
    try {
      const r = await fetchCliSessions(projectPath || undefined);
-      setSessions(r.sessions as unknown as CliSession[]);
+      setSessions(r.sessions);
    } catch (e) {
      setError(e instanceof Error ? e.message : String(e));
    } finally {
@@ -120,16 +123,16 @@ export function useCliSessionCore(options: UseCliSessionCoreOptions = {}): UseCl
    const created = await createCliSession(
      {
        workingDir: projectPath,
-        preferredShell: 'bash',
+        preferredShell,
        resumeKey,
        ...createSessionDefaults,
      },
      projectPath
    );
-    upsertSession(created.session as unknown as CliSession);
+    upsertSession(created.session);
    setSelectedSessionKey(created.session.sessionKey);
    return created.session.sessionKey;
-  }, [selectedSessionKey, projectPath, resumeKey, createSessionDefaults, upsertSession]);
+  }, [selectedSessionKey, projectPath, resumeKey, preferredShell, createSessionDefaults, upsertSession]);

  // ------- handleCreateSession -------
  const handleCreateSession = useCallback(async () => {
@@ -139,21 +142,21 @@ export function useCliSessionCore(options: UseCliSessionCoreOptions = {}): UseCl
      const created = await createCliSession(
        {
          workingDir: projectPath,
-          preferredShell: 'bash',
+          preferredShell,
          resumeKey,
          ...createSessionDefaults,
        },
        projectPath
      );
-      upsertSession(created.session as unknown as CliSession);
+      upsertSession(created.session);
      setSelectedSessionKey(created.session.sessionKey);
      // Refresh full list so store stays consistent
      const r = await fetchCliSessions(projectPath || undefined);
-      setSessions(r.sessions as unknown as CliSession[]);
+      setSessions(r.sessions);
    } catch (e) {
      setError(e instanceof Error ? e.message : String(e));
    }
-  }, [projectPath, resumeKey, createSessionDefaults, upsertSession, setSessions]);
+  }, [projectPath, resumeKey, preferredShell, createSessionDefaults, upsertSession, setSessions]);

  return {
    sessions,
--- a/ccw/frontend/src/lib/execution-display-utils.ts
+++ b/ccw/frontend/src/lib/execution-display-utils.ts
@@ -0,0 +1,65 @@
+// ========================================
+// Execution Display Utilities
+// ========================================
+// Shared helpers for rendering queue execution status across
+// ExecutionPanel and QueueExecutionListView components.
+
+import type { ReactElement } from 'react';
+import { createElement } from 'react';
+import {
+  Loader2,
+  CheckCircle,
+  XCircle,
+  Clock,
+} from 'lucide-react';
+import type { QueueExecutionStatus } from '@/stores/queueExecutionStore';
+
+/**
+ * Map execution status to Badge variant.
+ */
+export function statusBadgeVariant(status: QueueExecutionStatus): 'info' | 'success' | 'destructive' | 'secondary' {
+  switch (status) {
+    case 'running':
+      return 'info';
+    case 'completed':
+      return 'success';
+    case 'failed':
+      return 'destructive';
+    case 'pending':
+    default:
+      return 'secondary';
+  }
+}
+
+/**
+ * Map execution status to a small icon element.
+ */
+export function statusIcon(status: QueueExecutionStatus): ReactElement {
+  const base = 'w-3.5 h-3.5';
+  switch (status) {
+    case 'running':
+      return createElement(Loader2, { className: `${base} animate-spin text-info` });
+    case 'completed':
+      return createElement(CheckCircle, { className: `${base} text-success` });
+    case 'failed':
+      return createElement(XCircle, { className: `${base} text-destructive` });
+    case 'pending':
+    default:
+      return createElement(Clock, { className: `${base} text-muted-foreground` });
+  }
+}
+
+/**
+ * Format an ISO date string as a human-readable relative time.
+ */
+export function formatRelativeTime(isoString: string): string {
+  const diff = Date.now() - new Date(isoString).getTime();
+  const seconds = Math.floor(diff / 1000);
+  if (seconds < 60) return `${seconds}s ago`;
+  const minutes = Math.floor(seconds / 60);
+  if (minutes < 60) return `${minutes}m ago`;
+  const hours = Math.floor(minutes / 60);
+  if (hours < 24) return `${hours}h ago`;
+  const days = Math.floor(hours / 24);
+  return `${days}d ago`;
+}
--- a/ccw/frontend/src/lib/queue-prompt.ts
+++ b/ccw/frontend/src/lib/queue-prompt.ts
@@ -7,6 +7,34 @@

 import type { QueueItem } from '@/lib/api';

+// ---------------------------------------------------------------------------
+// Minimal interfaces for the issue data consumed by buildQueueItemContext.
+// The backend may return solution.tasks[] which is not part of the core
+// IssueSolution interface, so we define a local superset here.
+// ---------------------------------------------------------------------------
+
+/** A lightweight task entry nested inside a solution from the backend. */
+interface SolutionTask {
+  id: string;
+  title?: string;
+  description?: string;
+}
+
+/** Solution shape as consumed by the prompt builder (superset of IssueSolution). */
+interface PromptSolution {
+  id: string;
+  description?: string;
+  approach?: string;
+  tasks?: SolutionTask[];
+}
+
+/** Minimal issue shape consumed by the prompt builder. */
+export interface QueueItemIssue {
+  title?: string;
+  context?: string;
+  solutions?: PromptSolution[];
+}
+
 /**
 * Build a context string for executing a queue item.
 *
@@ -18,7 +46,7 @@ import type { QueueItem } from '@/lib/api';
 */
 export function buildQueueItemContext(
  item: QueueItem,
-  issue: any | undefined
+  issue: QueueItemIssue | undefined
 ): string {
  const lines: string[] = [];

@@ -38,7 +66,7 @@ export function buildQueueItemContext(
    }

    const solution = Array.isArray(issue.solutions)
-      ? issue.solutions.find((s: any) => s?.id === item.solution_id)
+      ? issue.solutions.find((s) => s?.id === item.solution_id)
      : undefined;

    if (solution) {
@@ -54,7 +82,7 @@ export function buildQueueItemContext(
      // Include matched task from solution.tasks when available
      const tasks = Array.isArray(solution.tasks) ? solution.tasks : [];
      const task = item.task_id
-        ? tasks.find((t: any) => t?.id === item.task_id)
+        ? tasks.find((t) => t?.id === item.task_id)
        : undefined;
      if (task) {
        lines.push('');
--- a/ccw/frontend/src/locales/en/mcp-manager.json
+++ b/ccw/frontend/src/locales/en/mcp-manager.json
@@ -153,7 +153,7 @@
      "allowedDirs": "Allowed Directories",
      "allowedDirsPlaceholder": "dir1,dir2,dir3",
      "allowedDirsHint": "Comma-separated list of allowed directories",
-      "disableSandbox": "Disable Sandbox"
+      "enableSandbox": "Enable Sandbox (restrict to workspace files only)"
    },
    "actions": {
      "enableAll": "Enable All",
--- a/ccw/frontend/src/locales/zh/mcp-manager.json
+++ b/ccw/frontend/src/locales/zh/mcp-manager.json
@@ -153,7 +153,7 @@
      "allowedDirs": "允许的目录",
      "allowedDirsPlaceholder": "目录1,目录2,目录3",
      "allowedDirsHint": "逗号分隔的允许目录列表",
-      "disableSandbox": "禁用沙箱"
+      "enableSandbox": "启用沙箱（仅允许修改工作空间下的文件）"
    },
    "actions": {
      "enableAll": "全部启用",
--- a/ccw/frontend/src/pages/McpManagerPage.tsx
+++ b/ccw/frontend/src/pages/McpManagerPage.tsx
@@ -349,7 +349,7 @@ export function McpManagerPage() {
    enabledTools: [],
    projectRoot: undefined,
    allowedDirs: undefined,
-    disableSandbox: undefined,
+    enableSandbox: undefined,
  };

  const handleToggleCcwTool = async (tool: string, enabled: boolean) => {
@@ -407,7 +407,7 @@ export function McpManagerPage() {
    enabledTools: [],
    projectRoot: undefined,
    allowedDirs: undefined,
-    disableSandbox: undefined,
+    enableSandbox: undefined,
  };

  const handleToggleCcwToolCodex = async (tool: string, enabled: boolean) => {
@@ -721,7 +721,7 @@ export function McpManagerPage() {
          enabledTools={ccwConfig.enabledTools}
          projectRoot={ccwConfig.projectRoot}
          allowedDirs={ccwConfig.allowedDirs}
-          disableSandbox={ccwConfig.disableSandbox}
+          enableSandbox={ccwConfig.enableSandbox}
          onToggleTool={handleToggleCcwTool}
          onUpdateConfig={handleUpdateCcwConfig}
          onInstall={handleCcwInstall}
@@ -734,7 +734,7 @@ export function McpManagerPage() {
          enabledTools={ccwCodexConfig.enabledTools}
          projectRoot={ccwCodexConfig.projectRoot}
          allowedDirs={ccwCodexConfig.allowedDirs}
-          disableSandbox={ccwCodexConfig.disableSandbox}
+          enableSandbox={ccwCodexConfig.enableSandbox}
          onToggleTool={handleToggleCcwToolCodex}
          onUpdateConfig={handleUpdateCcwConfigCodex}
          onInstall={handleCcwInstallCodex}
--- a/ccw/src/core/routes/mcp-routes.ts
+++ b/ccw/src/core/routes/mcp-routes.ts
@@ -1191,8 +1191,8 @@ export async function handleMcpRoutes(ctx: RouteContext): Promise<boolean> {
        return { error: 'projectPath is required', status: 400 };
      }

-      // Check if sandbox should be disabled
-      const disableSandbox = body.disableSandbox === true;
+      // Check if sandbox should be enabled
+      const enableSandbox = body.enableSandbox === true;

      // Parse enabled tools from request body
      const enabledTools = Array.isArray(body.enabledTools) && body.enabledTools.length > 0
@@ -1207,7 +1207,7 @@ export async function handleMcpRoutes(ctx: RouteContext): Promise<boolean> {
        args: isWin ? ["/c", "npx", "-y", "ccw-mcp"] : ["-y", "ccw-mcp"],
        env: {
          CCW_ENABLED_TOOLS: enabledTools,
-          ...(disableSandbox && { CCW_DISABLE_SANDBOX: "1" })
+          ...(enableSandbox && { CCW_ENABLE_SANDBOX: "1" })
        }
      };

--- a/ccw/src/mcp-server/index.ts
+++ b/ccw/src/mcp-server/index.ts
@@ -12,7 +12,7 @@ import {
 } from '@modelcontextprotocol/sdk/types.js';
 import { getAllToolSchemas, executeTool, executeToolWithProgress } from '../tools/index.js';
 import type { ToolSchema, ToolResult } from '../types/tool.js';
-import { getProjectRoot, getAllowedDirectories, isSandboxDisabled } from '../utils/path-validator.js';
+import { getProjectRoot, getAllowedDirectories, isSandboxEnabled } from '../utils/path-validator.js';

 const SERVER_NAME = 'ccw-tools';
 const SERVER_VERSION = '6.2.0';
@@ -178,13 +178,14 @@ async function main(): Promise<void> {
  // Log server start (to stderr to not interfere with stdio protocol)
  const projectRoot = getProjectRoot();
  const allowedDirs = getAllowedDirectories();
-  const sandboxDisabled = isSandboxDisabled();
+  const sandboxEnabled = isSandboxEnabled();
  console.error(`${SERVER_NAME} v${SERVER_VERSION} started`);
  console.error(`Project root: ${projectRoot}`);
-  if (sandboxDisabled) {
-    console.error(`Sandbox: DISABLED (CCW_DISABLE_SANDBOX=true)`);
-  } else {
+  if (sandboxEnabled) {
+    console.error(`Sandbox: ENABLED (CCW_ENABLE_SANDBOX=true)`);
    console.error(`Allowed directories: ${allowedDirs.join(', ')}`);
+  } else {
+    console.error(`Sandbox: DISABLED (default)`);
  }
  if (!process.env[ENV_PROJECT_ROOT]) {
    console.error(`[Warning] ${ENV_PROJECT_ROOT} not set, using process.cwd()`);