fix: use ANTHROPIC_AUTH_TOKEN for Claude CLI env injection

- Change env var from ANTHROPIC_API_KEY to ANTHROPIC_AUTH_TOKEN - Add Backend field propagation in taskSpec (cli.go) - Add stderr logging for injected env vars with API key masking - Add comprehensive tests for env injection flow Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>
test: use prefix match for version flag tests
2026-02-07 02:44:08 +08:00 · 2026-01-24 15:20:29 +08:00 · 2026-01-24 14:27:43 +08:00 · 2026-01-24 14:27:41 +08:00 · 2026-01-24 14:27:39 +08:00 · 2026-01-24 14:27:37 +08:00
122 changed files with 9467 additions and 4716 deletions
--- a/.claude-plugin/marketplace.json
+++ b/.claude-plugin/marketplace.json
@@ -42,6 +42,13 @@
      "version": "5.6.1",
      "source": "./development-essentials",
      "category": "productivity"
+    },
+    {
+      "name": "sparv",
+      "description": "Minimal SPARV workflow (Specify→Plan→Act→Review→Vault) with 10-point spec gate, unified journal, 2-action saves, 3-failure protocol, and EHRB risk detection",
+      "version": "1.1.0",
+      "source": "./skills/sparv",
+      "category": "development"
    }
  ]
 }
--- a/.github/workflows/release.yml
+++ b/.github/workflows/release.yml
@@ -74,7 +74,7 @@ jobs:
          if [ "${{ matrix.goos }}" = "windows" ]; then
            OUTPUT_NAME="${OUTPUT_NAME}.exe"
          fi
-          go build -ldflags="-s -w -X main.version=${VERSION}" -o ${OUTPUT_NAME} .
+          go build -ldflags="-s -w -X main.version=${VERSION}" -o ${OUTPUT_NAME} ./cmd/codeagent
          chmod +x ${OUTPUT_NAME}
          echo "artifact_path=codeagent-wrapper/${OUTPUT_NAME}" >> $GITHUB_OUTPUT

--- a/.gitignore
+++ b/.gitignore
@@ -7,3 +7,4 @@
 __pycache__
 .coverage
 coverage.out
+references
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -2,6 +2,66 @@

 All notable changes to this project will be documented in this file.

+## [5.6.4] - 2026-01-15
+
+### 🚀 Features
+
+- add reasoning effort config for codex backend
+- default to skip-permissions and bypass-sandbox
+- add multi-agent support with yolo mode
+- add omo module for multi-agent orchestration
+- add intelligent backend selection based on task complexity (#61)
+- v5.4.0 structured execution report (#94)
+- add millisecond-precision timestamps to all log entries (#91)
+- skill-install install script and security scan
+- add uninstall scripts with selective module removal
+
+### 🐛 Bug Fixes
+
+- filter codex stderr noise logs
+- use config override for codex reasoning effort
+- propagate SkipPermissions to parallel tasks (#113)
+- add timeout for Windows process termination
+- reject dash as workdir parameter (#118)
+- add sleep in fake script to prevent CI race condition
+- fix gemini env load
+- fix omo
+- fix codeagent skill TaskOutput
+- 修复 Gemini init 事件 session_id 未提取的问题 (#111)
+- Windows 后端退出：taskkill 结束进程树 + turn.completed 支持 (#108)
+- support model parameter for all backends, auto-inject from settings (#105)
+- replace setx with reg add to avoid 1024-char PATH truncation (#101)
+- 移除未知事件格式的日志噪声 (#96)
+- prevent duplicate PATH entries on reinstall (#95)
+- Minor issues #12 and #13 - ASCII mode and performance optimization
+- correct settings.json filename and bump version to v5.2.8
+- allow claude backend to read env from setting.json while preventing recursion (#92)
+- comprehensive security and quality improvements for PR #85 & #87 (#90)
+- Improve backend termination after message and extend timeout (#86)
+- Parser重复解析优化 + 严重bug修复 + PR #86兼容性 (#88)
+- filter noisy stderr output from gemini backend (#83)
+- 修復 wsl install.sh 格式問題 (#78)
+- 修复多 backend 并行日志 PID 混乱并移除包装格式 (#74) (#76)
+
+### 🚜 Refactor
+
+- remove sisyphus agent and unused code
+- streamline agent documentation and remove sisyphus
+
+### 📚 Documentation
+
+- add OmO workflow to README and fix plugin marketplace structure
+- update FAQ for default bypass/skip-permissions behavior
+- 添加 FAQ 常见问题章节
+- update troubleshooting with idempotent PATH commands (#95)
+
+### 💼 Other
+
+- add test-cases skill
+- add browser skill
+- BMADh和Requirements-Driven支持根据语义生成对应的文档 (#82)
+- update all readme
+
 ## [5.2.4] - 2025-12-16


--- a/README.md
+++ b/README.md
@@ -3,29 +3,13 @@
 # Claude Code Multi-Agent Workflow System

 [![Run in Smithery](https://smithery.ai/badge/skills/cexll)](https://smithery.ai/skills?ns=cexll&utm_source=github&utm_medium=badge)
-
-
 [![License: AGPL-3.0](https://img.shields.io/badge/License-AGPL_v3-blue.svg)](https://www.gnu.org/licenses/agpl-3.0)
 [![Claude Code](https://img.shields.io/badge/Claude-Code-blue)](https://claude.ai/code)
-[![Version](https://img.shields.io/badge/Version-5.2-green)](https://github.com/cexll/myclaude)
+[![Version](https://img.shields.io/badge/Version-6.x-green)](https://github.com/cexll/myclaude)

-> AI-powered development automation with multi-backend execution (Codex/Claude/Gemini)
+> AI-powered development automation with multi-backend execution (Codex/Claude/Gemini/OpenCode)

-## Core Concept: Multi-Backend Architecture
-
-This system leverages a **dual-agent architecture** with pluggable AI backends:
-
-| Role | Agent | Responsibility |
-|------|-------|----------------|
-| **Orchestrator** | Claude Code | Planning, context gathering, verification, user interaction |
-| **Executor** | codeagent-wrapper | Code editing, test execution (Codex/Claude/Gemini backends) |
-
-**Why this separation?**
- Claude Code excels at understanding context and orchestrating complex workflows
- Specialized backends (Codex for code, Claude for reasoning, Gemini for prototyping) excel at focused execution
- Backend selection via `--backend codex|claude|gemini` matches the model to the task
-
-## Quick Start(Please execute in Powershell on Windows)
+## Quick Start

 ```bash
 git clone https://github.com/cexll/myclaude.git
@@ -33,199 +17,23 @@ cd myclaude
 python3 install.py --install-dir ~/.claude
 ```

-## Workflows Overview
+## Modules Overview

-### 0. OmO Multi-Agent Orchestrator (Recommended for Complex Tasks)
-
-**Intelligent multi-agent orchestration that routes tasks to specialized agents based on risk signals.**
-
-```bash
-/omo "analyze and fix this authentication bug"
-```
-
-**Agent Hierarchy:**
-| Agent | Role | Backend | Model |
-|-------|------|---------|-------|
-| `oracle` | Technical advisor | Claude | claude-opus-4-5 |
-| `librarian` | External research | Claude | claude-sonnet-4-5 |
-| `explore` | Codebase search | OpenCode | grok-code |
-| `develop` | Code implementation | Codex | gpt-5.2 |
-| `frontend-ui-ux-engineer` | UI/UX specialist | Gemini | gemini-3-pro |
-| `document-writer` | Documentation | Gemini | gemini-3-flash |
-
-**Routing Signals (Not Fixed Pipeline):**
- Code location unclear → `explore`
- External library/API → `librarian`
- Risky/multi-file change → `oracle`
- Implementation needed → `develop` / `frontend-ui-ux-engineer`
-
-**Common Recipes:**
- Explain code: `explore`
- Small fix with known location: `develop` directly
- Bug fix, location unknown: `explore → develop`
- Cross-cutting refactor: `explore → oracle → develop`
- External API integration: `explore + librarian → oracle → develop`
-
-**Best For:** Complex bug investigation, multi-file refactoring, architecture decisions
-
---
-
-### 1. Dev Workflow (Recommended)
-
-**The primary workflow for most development tasks.**
-
-```bash
-/dev "implement user authentication with JWT"
-```
-
-**6-Step Process:**
-1. **Requirements Clarification** - Interactive Q&A to clarify scope
-2. **Codex Deep Analysis** - Codebase exploration and architecture decisions
-3. **Dev Plan Generation** - Structured task breakdown with test requirements
-4. **Parallel Execution** - Codex executes tasks concurrently
-5. **Coverage Validation** - Enforce ≥90% test coverage
-6. **Completion Summary** - Report with file changes and coverage stats
-
-**Key Features:**
- Claude Code orchestrates, Codex executes all code changes
- Automatic task parallelization for speed
- Mandatory 90% test coverage gate
- Rollback on failure
-
-**Best For:** Feature development, refactoring, bug fixes with tests
-
---
-
-### 2. BMAD Agile Workflow
-
-**Full enterprise agile methodology with 6 specialized agents.**
-
-```bash
-/bmad-pilot "build e-commerce checkout system"
-```
-
-**Agents:**
-| Agent | Role |
-|-------|------|
-| Product Owner | Requirements & user stories |
-| Architect | System design & tech decisions |
-| Tech Lead | Sprint planning & task breakdown |
-| Developer | Implementation |
-| Code Reviewer | Quality assurance |
-| QA Engineer | Testing & validation |
-
-**Process:**
-```
-Requirements → Architecture → Sprint Plan → Development → Review → QA
-     ↓              ↓             ↓            ↓          ↓       ↓
-   PRD.md      DESIGN.md     SPRINT.md     Code      REVIEW.md  TEST.md
-```
-
-**Best For:** Large features, team coordination, enterprise projects
-
---
-
-### 3. Requirements-Driven Workflow
-
-**Lightweight requirements-to-code pipeline.**
-
-```bash
-/requirements-pilot "implement API rate limiting"
-```
-
-**Process:**
-1. Requirements generation with quality scoring
-2. Implementation planning
-3. Code generation
-4. Review and testing
-
-**Best For:** Quick prototypes, well-defined features
-
---
-
-### 4. Development Essentials
-
-**Direct commands for daily coding tasks.**
-
-| Command | Purpose |
-|---------|---------|
-| `/code` | Implement a feature |
-| `/debug` | Debug an issue |
-| `/test` | Write tests |
-| `/review` | Code review |
-| `/optimize` | Performance optimization |
-| `/refactor` | Code refactoring |
-| `/docs` | Documentation |
-
-**Best For:** Quick tasks, no workflow overhead needed
-
-## Enterprise Workflow Features
-
- **Multi-backend execution:** `codeagent-wrapper --backend codex|claude|gemini` (default `codex`) so you can match the model to the task without changing workflows.
- **GitHub workflow commands:** `/gh-create-issue "short need"` creates structured issues; `/gh-issue-implement 123` pulls issue #123, drives development, and prepares the PR.
- **Skills + hooks activation:** .claude/hooks run automation (tests, reviews), while `.claude/skills/skill-rules.json` auto-suggests the right skills. Keep hooks enabled in `.claude/settings.json` to activate the enterprise workflow helpers.
-
---
-
-## Version Requirements
-
-### Codex CLI
-**Minimum version:** Check compatibility with your installation
-
-The codeagent-wrapper uses these Codex CLI features:
- `codex e` - Execute commands (shorthand for `codex exec`)
- `--skip-git-repo-check` - Skip git repository validation
- `--json` - JSON stream output format
- `-C <workdir>` - Set working directory
- `resume <session_id>` - Resume previous sessions
-
-**Verify Codex CLI is installed:**
-```bash
-which codex
-codex --version
-```
-
-### Claude CLI
-**Minimum version:** Check compatibility with your installation
-
-Required features:
- `--output-format stream-json` - Streaming JSON output format
- `--setting-sources` - Control setting sources (prevents infinite recursion)
- `--dangerously-skip-permissions` - Skip permission prompts (use with caution)
- `-p` - Prompt input flag
- `-r <session_id>` - Resume sessions
-
-**Security Note:** The wrapper adds `--dangerously-skip-permissions` for Claude by default. Set `CODEAGENT_SKIP_PERMISSIONS=false` to disable if you need permission prompts.
-
-**Verify Claude CLI is installed:**
-```bash
-which claude
-claude --version
-```
-
-### Gemini CLI
-**Minimum version:** Check compatibility with your installation
-
-Required features:
- `-o stream-json` - JSON stream output format
- `-y` - Auto-approve prompts (non-interactive mode)
- `-r <session_id>` - Resume sessions
- `-p` - Prompt input flag
-
-**Verify Gemini CLI is installed:**
-```bash
-which gemini
-gemini --version
-```
-
---
+| Module | Description | Documentation |
+|--------|-------------|---------------|
+| [do](skills/do/README.md) | **Recommended** - 7-phase feature development with codeagent orchestration | `/do` command |
+| [dev](dev-workflow/README.md) | Lightweight dev workflow with Codex integration | `/dev` command |
+| [omo](skills/omo/README.md) | Multi-agent orchestration with intelligent routing | `/omo` command |
+| [bmad](bmad-agile-workflow/README.md) | BMAD agile workflow with 6 specialized agents | `/bmad-pilot` command |
+| [requirements](requirements-driven-workflow/README.md) | Lightweight requirements-to-code pipeline | `/requirements-pilot` command |
+| [essentials](development-essentials/README.md) | Core development commands and utilities | `/code`, `/debug`, etc. |
+| [sparv](skills/sparv/README.md) | SPARV workflow (Specify→Plan→Act→Review→Vault) | `/sparv` command |
+| course | Course development (combines dev + product-requirements + test-cases) | Composite module |

 ## Installation

-### Modular Installation (Recommended)
-
 ```bash
-# Install all enabled modules (dev + essentials by default)
+# Install all enabled modules
 python3 install.py --install-dir ~/.claude

 # Install specific module
@@ -234,173 +42,72 @@ python3 install.py --module dev
 # List available modules
 python3 install.py --list-modules

-# Force overwrite existing files
+# Force overwrite
 python3 install.py --force
 ```

-### Available Modules
+### Module Configuration

-| Module | Default | Description |
-|--------|---------|-------------|
-| `dev` | ✓ Enabled | Dev workflow + Codex integration |
-| `essentials` | ✓ Enabled | Core development commands |
-| `bmad` | Disabled | Full BMAD agile workflow |
-| `requirements` | Disabled | Requirements-driven workflow |
-
-### What Gets Installed
-
-```
-~/.claude/
-├── bin/
-│   └── codeagent-wrapper    # Main executable
-├── CLAUDE.md                # Core instructions and role definition
-├── commands/                # Slash commands (/dev, /code, etc.)
-├── agents/                  # Agent definitions
-├── skills/
-│   └── codex/
-│       └── SKILL.md         # Codex integration skill
-├── config.json              # Configuration
-└── installed_modules.json   # Installation status
-```
-
-### Customizing Installation Directory
-
-By default, myclaude installs to `~/.claude`. You can customize this using the `INSTALL_DIR` environment variable:
-
-```bash
-# Install to custom directory
-INSTALL_DIR=/opt/myclaude bash install.sh
-
-# Update your PATH accordingly
-export PATH="/opt/myclaude/bin:$PATH"
-```
-
-**Directory Structure:**
- `$INSTALL_DIR/bin/` - codeagent-wrapper binary
- `$INSTALL_DIR/skills/` - Skill definitions
- `$INSTALL_DIR/config.json` - Configuration file
- `$INSTALL_DIR/commands/` - Slash command definitions
- `$INSTALL_DIR/agents/` - Agent definitions
-
-**Note:** When using a custom installation directory, ensure that `$INSTALL_DIR/bin` is added to your `PATH` environment variable.
-
-### Configuration
-
-Edit `config.json` to customize:
+Edit `config.json` to enable/disable modules:

 ```json
 {
-  "version": "1.0",
-  "install_dir": "~/.claude",
  "modules": {
-    "dev": {
-      "enabled": true,
-      "operations": [
-        {"type": "merge_dir", "source": "dev-workflow"},
-        {"type": "copy_file", "source": "memorys/CLAUDE.md", "target": "CLAUDE.md"},
-        {"type": "copy_file", "source": "skills/codex/SKILL.md", "target": "skills/codex/SKILL.md"},
-        {"type": "run_command", "command": "bash install.sh"}
-      ]
-    }
+    "dev": { "enabled": true },
+    "bmad": { "enabled": false },
+    "requirements": { "enabled": false },
+    "essentials": { "enabled": false },
+    "omo": { "enabled": false },
+    "sparv": { "enabled": false },
+    "do": { "enabled": false },
+    "course": { "enabled": false }
  }
 }
 ```

-**Operation Types:**
-| Type | Description |
-|------|-------------|
-| `merge_dir` | Merge subdirs (commands/, agents/) into install dir |
-| `copy_dir` | Copy entire directory |
-| `copy_file` | Copy single file to target path |
-| `run_command` | Execute shell command |
-
---
-
-## Codex Integration
-
-The `codex` skill enables Claude Code to delegate code execution to Codex CLI.
-
-### Usage in Workflows
-
-```bash
-# Codex is invoked via the skill
-codeagent-wrapper - <<'EOF'
-implement @src/auth.ts with JWT validation
-EOF
-```
-
-### Parallel Execution
-
-```bash
-codeagent-wrapper --parallel <<'EOF'
---TASK---
-id: backend_api
-workdir: /project/backend
---CONTENT---
-implement REST endpoints for /api/users
-
---TASK---
-id: frontend_ui
-workdir: /project/frontend
-dependencies: backend_api
---CONTENT---
-create React components consuming the API
-EOF
-```
-
-### Install Codex Wrapper
-
-```bash
-# Automatic (via dev module)
-python3 install.py --module dev
-
-# Manual
-bash install.sh
-```
-
-#### Windows
-
-Windows installs place `codeagent-wrapper.exe` in `%USERPROFILE%\bin`.
-
-```powershell
-# PowerShell (recommended)
-powershell -ExecutionPolicy Bypass -File install.ps1
-
-# Batch (cmd)
-install.bat
-```
-
-**Add to PATH** (if installer doesn't detect it):
-
-```powershell
-# PowerShell - persistent for current user
-[Environment]::SetEnvironmentVariable('PATH', "$HOME\bin;" + [Environment]::GetEnvironmentVariable('PATH','User'), 'User')
-
-# PowerShell - current session only
-$Env:PATH = "$HOME\bin;$Env:PATH"
-```
-
-```batch
-REM cmd.exe - persistent for current user (use PowerShell method above instead)
-REM WARNING: This expands %PATH% which includes system PATH, causing duplication
-REM Note: Using reg add instead of setx to avoid 1024-character truncation limit
-reg add "HKCU\Environment" /v Path /t REG_EXPAND_SZ /d "%USERPROFILE%\bin;%PATH%" /f
-```
-
---
-
 ## Workflow Selection Guide

-| Scenario | Recommended Workflow |
-|----------|---------------------|
-| New feature with tests | `/dev` |
-| Quick bug fix | `/debug` or `/code` |
-| Large multi-sprint feature | `/bmad-pilot` |
-| Prototype or POC | `/requirements-pilot` |
-| Code review | `/review` |
-| Performance issue | `/optimize` |
+| Scenario | Recommended |
+|----------|-------------|
+| Feature development (default) | `/do` |
+| Lightweight feature | `/dev` |
+| Bug investigation + fix | `/omo` |
+| Large enterprise project | `/bmad-pilot` |
+| Quick prototype | `/requirements-pilot` |
+| Simple task | `/code`, `/debug` |

---
+## Core Architecture
+
+| Role | Agent | Responsibility |
+|------|-------|----------------|
+| **Orchestrator** | Claude Code | Planning, context gathering, verification |
+| **Executor** | codeagent-wrapper | Code editing, test execution (Codex/Claude/Gemini/OpenCode) |
+
+## Backend CLI Requirements
+
+| Backend | Required Features |
+|---------|-------------------|
+| Codex | `codex e`, `--json`, `-C`, `resume` |
+| Claude | `--output-format stream-json`, `-r` |
+| Gemini | `-o stream-json`, `-y`, `-r` |
+
+## Directory Structure After Installation
+
+```
+~/.claude/
+├── bin/codeagent-wrapper
+├── CLAUDE.md
+├── commands/
+├── agents/
+├── skills/
+└── config.json
+```
+
+## Documentation
+
+- [Codeagent-Wrapper Guide](docs/CODEAGENT-WRAPPER.md)
+- [Hooks Documentation](docs/HOOKS.md)
+- [codeagent-wrapper](codeagent-wrapper/README.md)

 ## Troubleshooting

@@ -408,214 +115,38 @@ reg add "HKCU\Environment" /v Path /t REG_EXPAND_SZ /d "%USERPROFILE%\bin;%PATH%

 **Codex wrapper not found:**
 ```bash
-# Installer auto-adds PATH, check if configured
-if [[ ":$PATH:" != *":$HOME/.claude/bin:"* ]]; then
-    echo "PATH not configured. Reinstalling..."
-    bash install.sh
-fi
-
-# Or manually add (idempotent command)
-[[ ":$PATH:" != *":$HOME/.claude/bin:"* ]] && echo 'export PATH="$HOME/.claude/bin:$PATH"' >> ~/.zshrc
-```
-
-**Permission denied:**
-```bash
-python3 install.py --install-dir ~/.claude --force
+bash install.sh
 ```

 **Module not loading:**
 ```bash
-# Check installation status
 cat ~/.claude/installed_modules.json
-
-# Reinstall specific module
-python3 install.py --module dev --force
+python3 install.py --module <name> --force
 ```

-### Version Compatibility Issues
-
-**Backend CLI not found:**
+**Backend CLI errors:**
 ```bash
-# Check if backend CLIs are installed
-which codex
-which claude
-which gemini
-
-# Install missing backends
-# Codex: Follow installation instructions at https://codex.docs
-# Claude: Follow installation instructions at https://claude.ai/docs
-# Gemini: Follow installation instructions at https://ai.google.dev/docs
+which codex && codex --version
+which claude && claude --version
+which gemini && gemini --version
 ```

-**Unsupported CLI flags:**
-```bash
-# If you see errors like "unknown flag" or "invalid option"
+## FAQ

-# Check backend CLI version
-codex --version
-claude --version
-gemini --version
+| Issue | Solution |
+|-------|----------|
+| "Unknown event format" | Logging display issue, can be ignored |
+| Gemini can't read .gitignore files | Remove from .gitignore or use different backend |
+| `/dev` slow | Check logs, try faster model, use single repo |
+| Codex permission denied | Set `approval_policy = "never"` in ~/.codex/config.yaml |

-# For Codex: Ensure it supports `e`, `--skip-git-repo-check`, `--json`, `-C`, and `resume`
-# For Claude: Ensure it supports `--output-format stream-json`, `--setting-sources`, `-r`
-# For Gemini: Ensure it supports `-o stream-json`, `-y`, `-r`, `-p`
-
-# Update your backend CLI to the latest version if needed
-```
-
-**JSON parsing errors:**
-```bash
-# If you see "failed to parse JSON output" errors
-
-# Verify the backend outputs stream-json format
-codex e --json "test task"  # Should output newline-delimited JSON
-claude --output-format stream-json -p "test"  # Should output stream JSON
-
-# If not, your backend CLI version may be too old or incompatible
-```
-
-**Infinite recursion with Claude backend:**
-```bash
-# The wrapper prevents this with `--setting-sources ""` flag
-# If you still see recursion, ensure your Claude CLI supports this flag
-
-claude --help | grep "setting-sources"
-
-# If flag is not supported, upgrade Claude CLI
-```
-
-**Session resume failures:**
-```bash
-# Check if session ID is valid
-codex history  # List recent sessions
-claude history
-
-# Ensure backend CLI supports session resumption
-codex resume <session_id> "test"  # Should continue from previous session
-claude -r <session_id> "test"
-
-# If not supported, use new sessions instead of resume mode
-```
-
---
-
-## FAQ (Frequently Asked Questions)
-
-### Q1: `codeagent-wrapper` execution fails with "Unknown event format"
-
-**Problem:**
-```
-Unknown event format: {"type":"turn.started"}
-Unknown event format: {"type":"assistant", ...}
-```
-
-**Solution:**
-This is a logging event format display issue and does not affect actual functionality. It will be fixed in the next version. You can ignore these log outputs.
-
-**Related Issue:** [#96](https://github.com/cexll/myclaude/issues/96)
-
---
-
-### Q2: Gemini cannot read files ignored by `.gitignore`
-
-**Problem:**
-When using `codeagent-wrapper --backend gemini`, files in directories like `.claude/` that are ignored by `.gitignore` cannot be read.
-
-**Solution:**
- **Option 1:** Remove `.claude/` from your `.gitignore` file
- **Option 2:** Ensure files that need to be read are not in `.gitignore` list
-
-**Related Issue:** [#75](https://github.com/cexll/myclaude/issues/75)
-
---
-
-### Q3: `/dev` command parallel execution is very slow
-
-**Problem:**
-Using `/dev` command for simple features takes too long (over 30 minutes) with no visibility into task progress.
-
-**Solution:**
-1. **Check logs:** Review `C:\Users\User\AppData\Local\Temp\codeagent-wrapper-*.log` to identify bottlenecks
-2. **Adjust backend:**
-   - Try faster models like `gpt-5.1-codex-max`
-   - Running in WSL may be significantly faster
-3. **Workspace:** Use a single repository instead of monorepo with multiple sub-projects
-
-**Related Issue:** [#77](https://github.com/cexll/myclaude/issues/77)
-
---
-
-### Q4: Codex permission denied with new Go version
-
-**Problem:**
-After upgrading to the new Go-based Codex implementation, execution fails with permission denied errors.
-
-**Solution:**
-Add the following configuration to `~/.codex/config.yaml` (Windows: `c:\user\.codex\config.toml`):
-```yaml
-model = "gpt-5.1-codex-max"
-model_reasoning_effort = "high"
-model_reasoning_summary = "detailed"
-approval_policy = "never"
-sandbox_mode = "workspace-write"
-disable_response_storage = true
-network_access = true
-```
-
-**Key settings:**
- `approval_policy = "never"` - Remove approval restrictions
- `sandbox_mode = "workspace-write"` - Allow workspace write access
- `network_access = true` - Enable network access
-
-**Related Issue:** [#31](https://github.com/cexll/myclaude/issues/31)
-
---
-
-### Q5: How to disable default bypass/skip-permissions mode
-
-**Background:**
-By default, codeagent-wrapper enables bypass mode for both Codex and Claude backends:
- `CODEX_BYPASS_SANDBOX=true` - Bypasses Codex sandbox restrictions
- `CODEAGENT_SKIP_PERMISSIONS=true` - Skips Claude permission prompts
-
-**To disable (if you need sandbox/permission protection):**
-```bash
-export CODEX_BYPASS_SANDBOX=false
-export CODEAGENT_SKIP_PERMISSIONS=false
-```
-
-Or add to your shell profile (`~/.zshrc` or `~/.bashrc`):
-```bash
-echo 'export CODEX_BYPASS_SANDBOX=false' >> ~/.zshrc
-echo 'export CODEAGENT_SKIP_PERMISSIONS=false' >> ~/.zshrc
-```
-
-**Note:** Disabling bypass mode will require manual approval for certain operations.
-
---
-
-**Still having issues?** Visit [GitHub Issues](https://github.com/cexll/myclaude/issues) to search or report new issues.
-
---
-
-## Documentation
- **[Codeagent-Wrapper Guide](docs/CODEAGENT-WRAPPER.md)** - Multi-backend execution wrapper
- **[Hooks Documentation](docs/HOOKS.md)** - Custom hooks and automation
-
-### Additional Resources
- **[Installation Log](install.log)** - Installation history and troubleshooting
-
---
+See [GitHub Issues](https://github.com/cexll/myclaude/issues) for more.

 ## License

-AGPL-3.0 License - see [LICENSE](LICENSE)
+AGPL-3.0 - see [LICENSE](LICENSE)

 ## Support

- **Issues**: [GitHub Issues](https://github.com/cexll/myclaude/issues)
- **Documentation**: [docs/](docs/)
-
---
-
-**Claude Code + Codex = Better Development** - Orchestration meets execution.
+- [GitHub Issues](https://github.com/cexll/myclaude/issues)
+- [Documentation](docs/)
--- a/README_CN.md
+++ b/README_CN.md
@@ -2,25 +2,11 @@

 [![License: AGPL-3.0](https://img.shields.io/badge/License-AGPL_v3-blue.svg)](https://www.gnu.org/licenses/agpl-3.0)
 [![Claude Code](https://img.shields.io/badge/Claude-Code-blue)](https://claude.ai/code)
-[![Version](https://img.shields.io/badge/Version-5.2-green)](https://github.com/cexll/myclaude)
+[![Version](https://img.shields.io/badge/Version-6.x-green)](https://github.com/cexll/myclaude)

-> AI 驱动的开发自动化 - 多后端执行架构 (Codex/Claude/Gemini)
+> AI 驱动的开发自动化 - 多后端执行架构 (Codex/Claude/Gemini/OpenCode)

-## 核心概念：多后端架构
-
-本系统采用**双智能体架构**与可插拔 AI 后端：
-
-| 角色 | 智能体 | 职责 |
-|------|-------|------|
-| **编排者** | Claude Code | 规划、上下文收集、验证、用户交互 |
-| **执行者** | codeagent-wrapper | 代码编辑、测试执行（Codex/Claude/Gemini 后端）|
-
-**为什么分离？**
- Claude Code 擅长理解上下文和编排复杂工作流
- 专业后端（Codex 擅长代码、Claude 擅长推理、Gemini 擅长原型）专注执行
- 通过 `--backend codex|claude|gemini` 匹配模型与任务
-
-## 快速开始（windows上请在Powershell中执行）
+## 快速开始

 ```bash
 git clone https://github.com/cexll/myclaude.git
@@ -28,72 +14,125 @@ cd myclaude
 python3 install.py --install-dir ~/.claude
 ```

-## 工作流概览
+## 模块概览

-### 0. OmO 多智能体编排器（复杂任务推荐）
+| 模块 | 描述 | 文档 |
+|------|------|------|
+| [do](skills/do/README.md) | **推荐** - 7 阶段功能开发 + codeagent 编排 | `/do` 命令 |
+| [dev](dev-workflow/README.md) | 轻量级开发工作流 + Codex 集成 | `/dev` 命令 |
+| [omo](skills/omo/README.md) | 多智能体编排 + 智能路由 | `/omo` 命令 |
+| [bmad](bmad-agile-workflow/README.md) | BMAD 敏捷工作流 + 6 个专业智能体 | `/bmad-pilot` 命令 |
+| [requirements](requirements-driven-workflow/README.md) | 轻量级需求到代码流水线 | `/requirements-pilot` 命令 |
+| [essentials](development-essentials/README.md) | 核心开发命令和工具 | `/code`, `/debug` 等 |
+| [sparv](skills/sparv/README.md) | SPARV 工作流 (Specify→Plan→Act→Review→Vault) | `/sparv` 命令 |
+| course | 课程开发（组合 dev + product-requirements + test-cases） | 组合模块 |

-**基于风险信号智能路由任务到专业智能体的多智能体编排系统。**
+## 核心架构
+
+| 角色 | 智能体 | 职责 |
+|------|-------|------|
+| **编排者** | Claude Code | 规划、上下文收集、验证 |
+| **执行者** | codeagent-wrapper | 代码编辑、测试执行（Codex/Claude/Gemini/OpenCode 后端）|
+
+## 工作流详解
+
+### do 工作流（推荐）
+
+7 阶段功能开发，通过 codeagent-wrapper 编排多个智能体。**大多数功能开发任务的首选工作流。**

 ```bash
-/omo "分析并修复这个认证 bug"
+/do "添加用户登录功能"
 ```

-**智能体层级：**
-| 智能体 | 角色 | 后端 | 模型 |
-|-------|------|------|------|
-| `oracle` | 技术顾问 | Claude | claude-opus-4-5 |
-| `librarian` | 外部研究 | Claude | claude-sonnet-4-5 |
-| `explore` | 代码库搜索 | OpenCode | grok-code |
-| `develop` | 代码实现 | Codex | gpt-5.2 |
-| `frontend-ui-ux-engineer` | UI/UX 专家 | Gemini | gemini-3-pro |
-| `document-writer` | 文档撰写 | Gemini | gemini-3-flash |
+**7 阶段：**
+| 阶段 | 名称 | 目标 |
+|------|------|------|
+| 1 | Discovery | 理解需求 |
+| 2 | Exploration | 映射代码库模式 |
+| 3 | Clarification | 解决歧义（**强制**）|
+| 4 | Architecture | 设计实现方案 |
+| 5 | Implementation | 构建功能（**需审批**）|
+| 6 | Review | 捕获缺陷 |
+| 7 | Summary | 记录结果 |

-**路由信号（非固定流水线）：**
- 代码位置不明确 → `explore`
- 外部库/API → `librarian`
- 高风险/多文件变更 → `oracle`
- 需要实现 → `develop` / `frontend-ui-ux-engineer`
-
-**常用配方：**
- 解释代码：`explore`
- 位置已知的小修复：直接 `develop`
- Bug 修复，位置未知：`explore → develop`
- 跨模块重构：`explore → oracle → develop`
- 外部 API 集成：`explore + librarian → oracle → develop`
-
-**适用场景：** 复杂 bug 调查、多文件重构、架构决策
+**智能体：**
+- `code-explorer` - 代码追踪、架构映射
+- `code-architect` - 设计方案、文件规划
+- `code-reviewer` - 代码审查、简化建议
+- `develop` - 实现代码、运行测试

 ---

-### 1. Dev 工作流（推荐）
+### Dev 工作流

-**大多数开发任务的首选工作流。**
+轻量级开发工作流，适合简单功能开发。

 ```bash
 /dev "实现 JWT 用户认证"
 ```

 **6 步流程：**
-1. **需求澄清** - 交互式问答明确范围
-2. **Codex 深度分析** - 代码库探索和架构决策
-3. **开发计划生成** - 结构化任务分解和测试要求
-4. **并行执行** - Codex 并发执行任务
-5. **覆盖率验证** - 强制 ≥90% 测试覆盖率
-6. **完成总结** - 文件变更和覆盖率报告
-
-**核心特性：**
- Claude Code 编排，Codex 执行所有代码变更
- 自动任务并行化提升速度
- 强制 90% 测试覆盖率门禁
- 失败自动回滚
-
-**适用场景：** 功能开发、重构、带测试的 bug 修复
+1. 需求澄清 - 交互式问答
+2. Codex 深度分析 - 代码库探索
+3. 开发计划生成 - 结构化任务分解
+4. 并行执行 - Codex 并发执行
+5. 覆盖率验证 - 强制 ≥90%
+6. 完成总结 - 报告生成

 ---

-### 2. BMAD 敏捷工作流
+### OmO 多智能体编排器

-**包含 6 个专业智能体的完整企业敏捷方法论。**
+基于风险信号智能路由任务到专业智能体。
+
+```bash
+/omo "分析并修复这个认证 bug"
+```
+
+**智能体层级：**
+| 智能体 | 角色 | 后端 |
+|-------|------|------|
+| `oracle` | 技术顾问 | Claude |
+| `librarian` | 外部研究 | Claude |
+| `explore` | 代码库搜索 | OpenCode |
+| `develop` | 代码实现 | Codex |
+| `frontend-ui-ux-engineer` | UI/UX 专家 | Gemini |
+| `document-writer` | 文档撰写 | Gemini |
+
+**常用配方：**
+- 解释代码：`explore`
+- 位置已知的小修复：直接 `develop`
+- Bug 修复（位置未知）：`explore → develop`
+- 跨模块重构：`explore → oracle → develop`
+
+---
+
+### SPARV 工作流
+
+极简 5 阶段工作流：Specify → Plan → Act → Review → Vault。
+
+```bash
+/sparv "实现订单导出功能"
+```
+
+**核心规则：**
+- **10 分规格门**：得分 0-10，必须 >=9 才能进入 Plan
+- **2 动作保存**：每 2 次工具调用写入 journal.md
+- **3 失败协议**：连续 3 次失败后停止并上报
+- **EHRB**：高风险操作需明确确认
+
+**评分维度（各 0-2 分）：**
+1. Value - 为什么做，可验证的收益
+2. Scope - MVP + 不在范围内的内容
+3. Acceptance - 可测试的验收标准
+4. Boundaries - 错误/性能/兼容/安全边界
+5. Risk - EHRB/依赖/未知 + 处理方式
+
+---
+
+### BMAD 敏捷工作流
+
+完整企业敏捷方法论 + 6 个专业智能体。

 ```bash
 /bmad-pilot "构建电商结账系统"
@@ -104,43 +143,36 @@ python3 install.py --install-dir ~/.claude
 |-------|------|
 | Product Owner | 需求与用户故事 |
 | Architect | 系统设计与技术决策 |
-| Tech Lead | Sprint 规划与任务分解 |
+| Scrum Master | Sprint 规划与任务分解 |
 | Developer | 实现 |
 | Code Reviewer | 质量保证 |
 | QA Engineer | 测试与验证 |

-**流程：**
-```
-需求 → 架构 → Sprint计划 → 开发 → 审查 → QA
- ↓      ↓       ↓         ↓      ↓      ↓
-PRD.md DESIGN.md SPRINT.md Code REVIEW.md TEST.md
-```
-
-**适用场景：** 大型功能、团队协作、企业项目
+**审批门：**
+- PRD 完成后（90+ 分）需用户审批
+- 架构完成后（90+ 分）需用户审批

 ---

-### 3. 需求驱动工作流
+### 需求驱动工作流

-**轻量级需求到代码流水线。**
+轻量级需求到代码流水线。

 ```bash
 /requirements-pilot "实现 API 限流"
 ```

-**流程：**
-1. 带质量评分的需求生成
-2. 实现规划
-3. 代码生成
-4. 审查和测试
-
-**适用场景：** 快速原型、明确定义的功能
+**100 分质量评分：**
+- 功能清晰度：30 分
+- 技术具体性：25 分
+- 实现完整性：25 分
+- 业务上下文：20 分

 ---

-### 4. 开发基础命令
+### 开发基础命令

-**日常编码任务的直接命令。**
+日常编码任务的直接命令。

 | 命令 | 用途 |
 |------|------|
@@ -152,16 +184,12 @@ PRD.md DESIGN.md SPRINT.md Code REVIEW.md TEST.md
 | `/refactor` | 代码重构 |
 | `/docs` | 编写文档 |

-**适用场景：** 快速任务，无需工作流开销
-
 ---

 ## 安装

-### 模块化安装（推荐）
-
 ```bash
-# 安装所有启用的模块（默认：dev + essentials）
+# 安装所有启用的模块
 python3 install.py --install-dir ~/.claude

 # 安装特定模块
@@ -170,314 +198,77 @@ python3 install.py --module dev
 # 列出可用模块
 python3 install.py --list-modules

-# 强制覆盖现有文件
+# 强制覆盖
 python3 install.py --force
 ```

-### 可用模块
+### 模块配置

-| 模块 | 默认 | 描述 |
-|------|------|------|
-| `dev` | ✓ 启用 | Dev 工作流 + Codex 集成 |
-| `essentials` | ✓ 启用 | 核心开发命令 |
-| `bmad` | 禁用 | 完整 BMAD 敏捷工作流 |
-| `requirements` | 禁用 | 需求驱动工作流 |
-
-### 安装内容
-
-```
-~/.claude/
-├── bin/
-│   └── codeagent-wrapper    # 主可执行文件
-├── CLAUDE.md                # 核心指令和角色定义
-├── commands/                # 斜杠命令 (/dev, /code 等)
-├── agents/                  # 智能体定义
-├── skills/
-│   └── codex/
-│       └── SKILL.md         # Codex 集成技能
-├── config.json              # 配置文件
-└── installed_modules.json   # 安装状态
-```
-
-### 自定义安装目录
-
-默认情况下，myclaude 安装到 `~/.claude`。您可以使用 `INSTALL_DIR` 环境变量自定义安装目录：
-
-```bash
-# 安装到自定义目录
-INSTALL_DIR=/opt/myclaude bash install.sh
-
-# 相应更新您的 PATH
-export PATH="/opt/myclaude/bin:$PATH"
-```
-
-**目录结构：**
- `$INSTALL_DIR/bin/` - codeagent-wrapper 可执行文件
- `$INSTALL_DIR/skills/` - 技能定义
- `$INSTALL_DIR/config.json` - 配置文件
- `$INSTALL_DIR/commands/` - 斜杠命令定义
- `$INSTALL_DIR/agents/` - 智能体定义
-
-**注意：** 使用自定义安装目录时，请确保将 `$INSTALL_DIR/bin` 添加到您的 `PATH` 环境变量中。
-
-### 配置
-
-编辑 `config.json` 自定义：
+编辑 `config.json` 启用/禁用模块：

 ```json
 {
-  "version": "1.0",
-  "install_dir": "~/.claude",
  "modules": {
-    "dev": {
-      "enabled": true,
-      "operations": [
-        {"type": "merge_dir", "source": "dev-workflow"},
-        {"type": "copy_file", "source": "memorys/CLAUDE.md", "target": "CLAUDE.md"},
-        {"type": "copy_file", "source": "skills/codex/SKILL.md", "target": "skills/codex/SKILL.md"},
-        {"type": "run_command", "command": "bash install.sh"}
-      ]
-    }
+    "dev": { "enabled": true },
+    "bmad": { "enabled": false },
+    "requirements": { "enabled": false },
+    "essentials": { "enabled": false },
+    "omo": { "enabled": false },
+    "sparv": { "enabled": false },
+    "do": { "enabled": false },
+    "course": { "enabled": false }
  }
 }
 ```

-**操作类型：**
-| 类型 | 描述 |
-|------|------|
-| `merge_dir` | 合并子目录 (commands/, agents/) 到安装目录 |
-| `copy_dir` | 复制整个目录 |
-| `copy_file` | 复制单个文件到目标路径 |
-| `run_command` | 执行 shell 命令 |
-
---
-
-## Codex 集成
-
-`codex` 技能使 Claude Code 能够将代码执行委托给 Codex CLI。
-
-### 工作流中的使用
-
-```bash
-# 通过技能调用 Codex
-codeagent-wrapper - <<'EOF'
-在 @src/auth.ts 中实现 JWT 验证
-EOF
-```
-
-### 并行执行
-
-```bash
-codeagent-wrapper --parallel <<'EOF'
---TASK---
-id: backend_api
-workdir: /project/backend
---CONTENT---
-实现 /api/users 的 REST 端点
-
---TASK---
-id: frontend_ui
-workdir: /project/frontend
-dependencies: backend_api
---CONTENT---
-创建消费 API 的 React 组件
-EOF
-```
-
-### 安装 Codex Wrapper
-
-```bash
-# 自动（通过 dev 模块）
-python3 install.py --module dev
-
-# 手动
-bash install.sh
-```
-
-#### Windows 系统
-
-Windows 系统会将 `codeagent-wrapper.exe` 安装到 `%USERPROFILE%\bin`。
-
-```powershell
-# PowerShell（推荐）
-powershell -ExecutionPolicy Bypass -File install.ps1
-
-# 批处理（cmd）
-install.bat
-```
-
-**添加到 PATH**（如果安装程序未自动检测）：
-
-```powershell
-# PowerShell - 永久添加（当前用户）
-[Environment]::SetEnvironmentVariable('PATH', "$HOME\bin;" + [Environment]::GetEnvironmentVariable('PATH','User'), 'User')
-
-# PowerShell - 仅当前会话
-$Env:PATH = "$HOME\bin;$Env:PATH"
-```
-
-```batch
-REM cmd.exe - 永久添加（当前用户）（建议使用上面的 PowerShell 方法）
-REM 警告：此命令会展开 %PATH% 包含系统 PATH，导致重复
-REM 注意：使用 reg add 而非 setx 以避免 1024 字符截断限制
-reg add "HKCU\Environment" /v Path /t REG_EXPAND_SZ /d "%USERPROFILE%\bin;%PATH%" /f
-```
-
---
-
 ## 工作流选择指南

-| 场景 | 推荐工作流 |
-|------|----------|
-| 带测试的新功能 | `/dev` |
-| 快速 bug 修复 | `/debug` 或 `/code` |
-| 大型多 Sprint 功能 | `/bmad-pilot` |
-| 原型或 POC | `/requirements-pilot` |
-| 代码审查 | `/review` |
-| 性能问题 | `/optimize` |
+| 场景 | 推荐 |
+|------|------|
+| 功能开发（默认） | `/do` |
+| 轻量级功能 | `/dev` |
+| Bug 调查 + 修复 | `/omo` |
+| 大型企业项目 | `/bmad-pilot` |
+| 快速原型 | `/requirements-pilot` |
+| 简单任务 | `/code`, `/debug` |

---
+## 后端 CLI 要求
+
+| 后端 | 必需功能 |
+|------|----------|
+| Codex | `codex e`, `--json`, `-C`, `resume` |
+| Claude | `--output-format stream-json`, `-r` |
+| Gemini | `-o stream-json`, `-y`, `-r` |

 ## 故障排查

-### 常见问题
-
 **Codex wrapper 未找到：**
 ```bash
-# 安装程序会自动添加 PATH，检查是否已添加
-if [[ ":$PATH:" != *":$HOME/.claude/bin:"* ]]; then
-    echo "PATH not configured. Reinstalling..."
-    bash install.sh
-fi
-
-# 或手动添加（幂等性命令）
-[[ ":$PATH:" != *":$HOME/.claude/bin:"* ]] && echo 'export PATH="$HOME/.claude/bin:$PATH"' >> ~/.zshrc
-```
-
-**权限被拒绝：**
-```bash
-python3 install.py --install-dir ~/.claude --force
+bash install.sh
 ```

 **模块未加载：**
 ```bash
-# 检查安装状态
 cat ~/.claude/installed_modules.json
-
-# 重新安装特定模块
-python3 install.py --module dev --force
+python3 install.py --module <name> --force
 ```

---
+## FAQ

-## 常见问题 (FAQ)
+| 问题 | 解决方案 |
+|------|----------|
+| "Unknown event format" | 日志显示问题，可忽略 |
+| Gemini 无法读取 .gitignore 文件 | 从 .gitignore 移除或使用其他后端 |
+| `/dev` 执行慢 | 检查日志，尝试更快模型，使用单一仓库 |
+| Codex 权限拒绝 | 在 ~/.codex/config.yaml 设置 `approval_policy = "never"` |

-### Q1: `codeagent-wrapper` 执行时报错 "Unknown event format"
-
-**问题描述：**
-执行 `codeagent-wrapper` 时出现错误：
-```
-Unknown event format: {"type":"turn.started"}
-Unknown event format: {"type":"assistant", ...}
-```
-
-**解决方案：**
-这是日志事件流的显示问题，不影响实际功能执行。预计在下个版本中修复。如需排查其他问题，可忽略此日志输出。
-
-**相关 Issue：** [#96](https://github.com/cexll/myclaude/issues/96)
-
---
-
-### Q2: Gemini 无法读取 `.gitignore` 忽略的文件
-
-**问题描述：**
-使用 `codeagent-wrapper --backend gemini` 时，无法读取 `.claude/` 等被 `.gitignore` 忽略的目录中的文件。
-
-**解决方案：**
- **方案一：** 在项目根目录的 `.gitignore` 中取消对 `.claude/` 的忽略
- **方案二：** 确保需要读取的文件不在 `.gitignore` 忽略列表中
-
-**相关 Issue：** [#75](https://github.com/cexll/myclaude/issues/75)
-
---
-
-### Q3: `/dev` 命令并行执行特别慢
-
-**问题描述：**
-使用 `/dev` 命令开发简单功能耗时过长（超过30分钟），无法了解任务执行状态。
-
-**解决方案：**
-1. **检查日志：** 查看 `C:\Users\User\AppData\Local\Temp\codeagent-wrapper-*.log` 分析瓶颈
-2. **调整后端：**
-   - 尝试使用 `gpt-5.1-codex-max` 等更快的模型
-   - 在 WSL 环境下运行速度可能更快
-3. **工作区选择：** 使用独立的代码仓库而非包含多个子项目的 monorepo
-
-**相关 Issue：** [#77](https://github.com/cexll/myclaude/issues/77)
-
---
-
-### Q4: 新版 Go 实现的 Codex 权限不足
-
-**问题描述：**
-升级到新版 Go 实现的 Codex 后，出现权限不足的错误。
-
-**解决方案：**
-在 `~/.codex/config.yaml` 中添加以下配置（Windows: `c:\user\.codex\config.toml`）：
-```yaml
-model = "gpt-5.1-codex-max"
-model_reasoning_effort = "high"
-model_reasoning_summary = "detailed"
-approval_policy = "never"
-sandbox_mode = "workspace-write"
-disable_response_storage = true
-network_access = true
-```
-
-**关键配置说明：**
- `approval_policy = "never"` - 移除审批限制
- `sandbox_mode = "workspace-write"` - 允许工作区写入权限
- `network_access = true` - 启用网络访问
-
-**相关 Issue：** [#31](https://github.com/cexll/myclaude/issues/31)
-
---
-
-### Q5: 执行时遇到权限拒绝或沙箱限制
-
-**问题描述：**
-运行 codeagent-wrapper 时出现权限错误或沙箱限制。
-
-**解决方案：**
-设置以下环境变量：
-```bash
-export CODEX_BYPASS_SANDBOX=true
-export CODEAGENT_SKIP_PERMISSIONS=true
-```
-
-或添加到 shell 配置文件（`~/.zshrc` 或 `~/.bashrc`）：
-```bash
-echo 'export CODEX_BYPASS_SANDBOX=true' >> ~/.zshrc
-echo 'export CODEAGENT_SKIP_PERMISSIONS=true' >> ~/.zshrc
-```
-
-**注意：** 这些设置会绕过安全限制，请仅在可信环境中使用。
-
---
-
-**仍有疑问？** 请访问 [GitHub Issues](https://github.com/cexll/myclaude/issues) 搜索或提交新问题。
-
---
+更多问题请访问 [GitHub Issues](https://github.com/cexll/myclaude/issues)。

 ## 许可证

-AGPL-3.0 License - 查看 [LICENSE](LICENSE)
+AGPL-3.0 - 查看 [LICENSE](LICENSE)

 ## 支持

- **问题反馈**: [GitHub Issues](https://github.com/cexll/myclaude/issues)
- **文档**: [docs/](docs/)
-
---
-
-**Claude Code + Codex = 更好的开发** - 编排遇见执行。
+- [GitHub Issues](https://github.com/cexll/myclaude/issues)
+- [文档](docs/)
--- a/bmad-agile-workflow/README.md
+++ b/bmad-agile-workflow/README.md
@@ -0,0 +1,109 @@
+# bmad - BMAD Agile Workflow
+
+Full enterprise agile methodology with 6 specialized agents, UltraThink analysis, and repository-aware development.
+
+## Installation
+
+```bash
+python install.py --module bmad
+```
+
+## Usage
+
+```bash
+/bmad-pilot <PROJECT_DESCRIPTION> [OPTIONS]
+```
+
+### Options
+
+| Option | Description |
+|--------|-------------|
+| `--skip-tests` | Skip QA testing phase |
+| `--direct-dev` | Skip SM planning, go directly to development |
+| `--skip-scan` | Skip initial repository scanning |
+
+## Workflow Phases
+
+| Phase | Agent | Deliverable | Description |
+|-------|-------|-------------|-------------|
+| 0 | Orchestrator | `00-repo-scan.md` | Repository scanning with UltraThink analysis |
+| 1 | Product Owner (PO) | `01-product-requirements.md` | PRD with 90+ quality score required |
+| 2 | Architect | `02-system-architecture.md` | Technical design with 90+ score required |
+| 3 | Scrum Master (SM) | `03-sprint-plan.md` | Sprint backlog with stories and estimates |
+| 4 | Developer | Implementation code | Multi-sprint implementation |
+| 4.5 | Reviewer | `04-dev-reviewed.md` | Code review (Pass/Pass with Risk/Fail) |
+| 5 | QA Engineer | Test suite | Comprehensive testing and validation |
+
+## Agents
+
+| Agent | Role |
+|-------|------|
+| `bmad-orchestrator` | Repository scanning, workflow coordination |
+| `bmad-po` | Requirements gathering, PRD creation |
+| `bmad-architect` | System design, technology decisions |
+| `bmad-sm` | Sprint planning, task breakdown |
+| `bmad-dev` | Code implementation |
+| `bmad-review` | Code review, quality assessment |
+| `bmad-qa` | Testing, validation |
+
+## Approval Gates
+
+Two mandatory stop points require explicit user approval:
+
+1. **After PRD** (Phase 1 → 2): User must approve requirements before architecture
+2. **After Architecture** (Phase 2 → 3): User must approve design before implementation
+
+## Output Structure
+
+```
+.claude/specs/{feature_name}/
+├── 00-repo-scan.md
+├── 01-product-requirements.md
+├── 02-system-architecture.md
+├── 03-sprint-plan.md
+└── 04-dev-reviewed.md
+```
+
+## UltraThink Methodology
+
+Applied throughout the workflow for deep analysis:
+
+1. **Hypothesis Generation** - Form hypotheses about the problem
+2. **Evidence Collection** - Gather evidence from codebase
+3. **Pattern Recognition** - Identify recurring patterns
+4. **Synthesis** - Create comprehensive understanding
+5. **Validation** - Cross-check findings
+
+## Interactive Confirmation Flow
+
+PO and Architect phases use iterative refinement:
+
+1. Agent produces initial draft + quality score
+2. Orchestrator presents to user with clarification questions
+3. User provides responses
+4. Agent refines until quality >= 90
+5. User confirms to save deliverable
+
+## When to Use
+
+- Large multi-sprint features
+- Enterprise projects requiring documentation
+- Team coordination scenarios
+- Projects needing formal approval gates
+
+## Directory Structure
+
+```
+bmad-agile-workflow/
+├── README.md
+├── commands/
+│   └── bmad-pilot.md
+└── agents/
+    ├── bmad-orchestrator.md
+    ├── bmad-po.md
+    ├── bmad-architect.md
+    ├── bmad-sm.md
+    ├── bmad-dev.md
+    ├── bmad-review.md
+    └── bmad-qa.md
+```
--- a/codeagent-wrapper/.github/workflows/ci.yml
+++ b/codeagent-wrapper/.github/workflows/ci.yml
@@ -0,0 +1,39 @@
+name: CI
+
+on:
+  push:
+    branches: [main, master]
+  pull_request:
+
+permissions:
+  contents: read
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    strategy:
+      fail-fast: false
+      matrix:
+        go-version: ["1.21", "1.22"]
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-go@v5
+        with:
+          go-version: ${{ matrix.go-version }}
+          cache: true
+      - name: Test
+        run: make test
+      - name: Build
+        run: make build
+
+  lint:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-go@v5
+        with:
+          go-version: "1.22"
+          cache: true
+      - name: Lint
+        run: make lint
+
--- a/codeagent-wrapper/.gitignore
+++ b/codeagent-wrapper/.gitignore
@@ -1,4 +1,7 @@
 # Build artifacts
+bin/
+codeagent
+codeagent.exe
 codeagent-wrapper
 codeagent-wrapper.exe
 *.test
@@ -9,3 +12,12 @@ coverage*.out
 cover.out
 cover_*.out
 coverage.html
+
+# Logs
+*.log
+
+# Temp files
+*.tmp
+*.swp
+*~
+.DS_Store
--- a/codeagent-wrapper/Makefile
+++ b/codeagent-wrapper/Makefile
@@ -0,0 +1,37 @@
+GO ?= go
+
+TOOLS_BIN := $(CURDIR)/bin
+TOOLCHAIN ?= go1.22.0
+GOLANGCI_LINT_VERSION := v1.56.2
+STATICCHECK_VERSION := v0.4.7
+
+GOLANGCI_LINT := $(TOOLS_BIN)/golangci-lint
+STATICCHECK := $(TOOLS_BIN)/staticcheck
+
+.PHONY: build test lint clean install
+
+build:
+	$(GO) build -o codeagent ./cmd/codeagent
+	$(GO) build -o codeagent-wrapper ./cmd/codeagent-wrapper
+
+test:
+	$(GO) test ./...
+
+$(GOLANGCI_LINT):
+	@mkdir -p $(TOOLS_BIN)
+	GOTOOLCHAIN=$(TOOLCHAIN) GOBIN=$(TOOLS_BIN) $(GO) install github.com/golangci/golangci-lint/cmd/golangci-lint@$(GOLANGCI_LINT_VERSION)
+
+$(STATICCHECK):
+	@mkdir -p $(TOOLS_BIN)
+	GOTOOLCHAIN=$(TOOLCHAIN) GOBIN=$(TOOLS_BIN) $(GO) install honnef.co/go/tools/cmd/staticcheck@$(STATICCHECK_VERSION)
+
+lint: $(GOLANGCI_LINT) $(STATICCHECK)
+	GOTOOLCHAIN=$(TOOLCHAIN) $(GOLANGCI_LINT) run ./...
+	GOTOOLCHAIN=$(TOOLCHAIN) $(STATICCHECK) ./...
+
+clean:
+	@python3 -c 'import glob, os; paths=["codeagent","codeagent.exe","codeagent-wrapper","codeagent-wrapper.exe","coverage.out","cover.out","coverage.html"]; paths += glob.glob("coverage*.out") + glob.glob("cover_*.out") + glob.glob("*.test"); [os.remove(p) for p in paths if os.path.exists(p)]'
+
+install:
+	$(GO) install ./cmd/codeagent
+	$(GO) install ./cmd/codeagent-wrapper
--- a/codeagent-wrapper/README.md
+++ b/codeagent-wrapper/README.md
@@ -0,0 +1,152 @@
+# codeagent-wrapper
+
+`codeagent-wrapper` 是一个用 Go 编写的“多后端 AI 代码代理”命令行包装器：用统一的 CLI 入口封装不同的 AI 工具后端（Codex / Claude / Gemini / Opencode），并提供一致的参数、配置与会话恢复体验。
+
+入口：`cmd/codeagent/main.go`（生成二进制名：`codeagent`）和 `cmd/codeagent-wrapper/main.go`（生成二进制名：`codeagent-wrapper`）。两者行为一致。
+
+## 功能特性
+
+- 多后端支持：`codex` / `claude` / `gemini` / `opencode`
+- 统一命令行：`codeagent [flags] <task>` / `codeagent resume <session_id> <task> [workdir]`
+- 自动 stdin：遇到换行/特殊字符/超长任务自动走 stdin，避免 shell quoting 地狱；也可显式使用 `-`
+- 配置合并：支持配置文件与 `CODEAGENT_*` 环境变量（viper）
+- Agent 预设：从 `~/.codeagent/models.json` 读取 backend/model/prompt 等预设
+- 并行执行：`--parallel` 从 stdin 读取多任务配置，支持依赖拓扑并发执行
+- 日志清理：`codeagent cleanup` 清理旧日志（日志写入系统临时目录）
+
+## 安装
+
+要求：Go 1.21+。
+
+在仓库根目录执行：
+
+```bash
+go install ./cmd/codeagent
+go install ./cmd/codeagent-wrapper
+```
+
+安装后确认：
+
+```bash
+codeagent version
+codeagent-wrapper version
+```
+
+## 使用示例
+
+最简单用法（默认后端：`codex`）：
+
+```bash
+codeagent "分析 internal/app/cli.go 的入口逻辑，给出改进建议"
+```
+
+指定后端：
+
+```bash
+codeagent --backend claude "解释 internal/executor/parallel_config.go 的并行配置格式"
+```
+
+指定工作目录（第 2 个位置参数）：
+
+```bash
+codeagent "在当前 repo 下搜索潜在数据竞争" .
+```
+
+显式从 stdin 读取 task（使用 `-`）：
+
+```bash
+cat task.txt | codeagent -
+```
+
+恢复会话：
+
+```bash
+codeagent resume <session_id> "继续上次任务"
+```
+
+并行模式（从 stdin 读取任务配置；禁止位置参数）：
+
+```bash
+codeagent --parallel <<'EOF'
+---TASK---
+id: t1
+workdir: .
+backend: codex
+---CONTENT---
+列出本项目的主要模块以及它们的职责。
+---TASK---
+id: t2
+dependencies: t1
+backend: claude
+---CONTENT---
+基于 t1 的结论，提出重构风险点与建议。
+EOF
+```
+
+## 配置说明
+
+### 配置文件
+
+默认查找路径（当 `--config` 为空时）：
+
+- `$HOME/.codeagent/config.(yaml|yml|json|toml|...)`
+
+示例（YAML）：
+
+```yaml
+backend: codex
+model: gpt-4.1
+skip-permissions: false
+```
+
+也可以通过 `--config /path/to/config.yaml` 显式指定。
+
+### 环境变量（`CODEAGENT_*`）
+
+通过 viper 读取并自动映射 `-` 为 `_`，常用项：
+
+- `CODEAGENT_BACKEND`（`codex|claude|gemini|opencode`）
+- `CODEAGENT_MODEL`
+- `CODEAGENT_AGENT`
+- `CODEAGENT_PROMPT_FILE`
+- `CODEAGENT_REASONING_EFFORT`
+- `CODEAGENT_SKIP_PERMISSIONS`
+- `CODEAGENT_FULL_OUTPUT`（并行模式 legacy 输出）
+- `CODEAGENT_MAX_PARALLEL_WORKERS`（0 表示不限制，上限 100）
+
+### Agent 预设（`~/.codeagent/models.json`）
+
+可在 `~/.codeagent/models.json` 定义 agent → backend/model/prompt 等映射，用 `--agent <name>` 选择：
+
+```json
+{
+  "default_backend": "opencode",
+  "default_model": "opencode/grok-code",
+  "agents": {
+    "develop": {
+      "backend": "codex",
+      "model": "gpt-4.1",
+      "prompt_file": "~/.codeagent/prompts/develop.md",
+      "description": "Code development"
+    }
+  }
+}
+```
+
+## 支持的后端
+
+该项目本身不内置模型能力，依赖你本机安装并可在 `PATH` 中找到对应 CLI：
+
+- `codex`：执行 `codex e ...`（默认会添加 `--dangerously-bypass-approvals-and-sandbox`；如需关闭请设置 `CODEX_BYPASS_SANDBOX=false`）
+- `claude`：执行 `claude -p ... --output-format stream-json`（默认会跳过权限提示；如需开启请设置 `CODEAGENT_SKIP_PERMISSIONS=false`）
+- `gemini`：执行 `gemini ... -o stream-json`（可从 `~/.gemini/.env` 加载环境变量）
+- `opencode`：执行 `opencode run --format json`
+
+## 开发
+
+```bash
+make build
+make test
+make lint
+make clean
+```
--- a/codeagent-wrapper/agent_config.go
+++ b/codeagent-wrapper/agent_config.go
@@ -1,79 +0,0 @@
-package main
-
-import (
-	"encoding/json"
-	"fmt"
-	"os"
-	"path/filepath"
-)
-
-type AgentModelConfig struct {
-	Backend     string `json:"backend"`
-	Model       string `json:"model"`
-	PromptFile  string `json:"prompt_file,omitempty"`
-	Description string `json:"description,omitempty"`
-	Yolo        bool   `json:"yolo,omitempty"`
-	Reasoning   string `json:"reasoning,omitempty"`
-}
-
-type ModelsConfig struct {
-	DefaultBackend string                      `json:"default_backend"`
-	DefaultModel   string                      `json:"default_model"`
-	Agents         map[string]AgentModelConfig `json:"agents"`
-}
-
-var defaultModelsConfig = ModelsConfig{
-	DefaultBackend: "opencode",
-	DefaultModel:   "opencode/grok-code",
-	Agents: map[string]AgentModelConfig{
-			"oracle":                  {Backend: "claude", Model: "claude-sonnet-4-20250514", PromptFile: "~/.claude/skills/omo/references/oracle.md", Description: "Technical advisor"},
-			"librarian":               {Backend: "claude", Model: "claude-sonnet-4-5-20250514", PromptFile: "~/.claude/skills/omo/references/librarian.md", Description: "Researcher"},
-			"explore":                 {Backend: "opencode", Model: "opencode/grok-code", PromptFile: "~/.claude/skills/omo/references/explore.md", Description: "Code search"},
-			"develop":                 {Backend: "codex", Model: "", PromptFile: "~/.claude/skills/omo/references/develop.md", Description: "Code development"},
-			"frontend-ui-ux-engineer": {Backend: "gemini", Model: "", PromptFile: "~/.claude/skills/omo/references/frontend-ui-ux-engineer.md", Description: "Frontend engineer"},
-			"document-writer":         {Backend: "gemini", Model: "", PromptFile: "~/.claude/skills/omo/references/document-writer.md", Description: "Documentation"},
-		},
-	}
-
-func loadModelsConfig() *ModelsConfig {
-	home, err := os.UserHomeDir()
-	if err != nil {
-		logWarn(fmt.Sprintf("Failed to resolve home directory for models config: %v; using defaults", err))
-		return &defaultModelsConfig
-	}
-
-	configPath := filepath.Join(home, ".codeagent", "models.json")
-	data, err := os.ReadFile(configPath)
-	if err != nil {
-		if !os.IsNotExist(err) {
-			logWarn(fmt.Sprintf("Failed to read models config %s: %v; using defaults", configPath, err))
-		}
-		return &defaultModelsConfig
-	}
-
-	var cfg ModelsConfig
-	if err := json.Unmarshal(data, &cfg); err != nil {
-		logWarn(fmt.Sprintf("Failed to parse models config %s: %v; using defaults", configPath, err))
-		return &defaultModelsConfig
-	}
-
-	// Merge with defaults
-	for name, agent := range defaultModelsConfig.Agents {
-		if _, exists := cfg.Agents[name]; !exists {
-			if cfg.Agents == nil {
-				cfg.Agents = make(map[string]AgentModelConfig)
-			}
-			cfg.Agents[name] = agent
-		}
-	}
-
-	return &cfg
-}
-
-func resolveAgentConfig(agentName string) (backend, model, promptFile, reasoning string, yolo bool) {
-	cfg := loadModelsConfig()
-	if agent, ok := cfg.Agents[agentName]; ok {
-		return agent.Backend, agent.Model, agent.PromptFile, agent.Reasoning, agent.Yolo
-	}
-	return cfg.DefaultBackend, cfg.DefaultModel, "", "", false
-}
--- a/codeagent-wrapper/backend.go
+++ b/codeagent-wrapper/backend.go
@@ -1,237 +0,0 @@
-package main
-
-import (
-	"encoding/json"
-	"os"
-	"path/filepath"
-	"strings"
-)
-
-// Backend defines the contract for invoking different AI CLI backends.
-// Each backend is responsible for supplying the executable command and
-// building the argument list based on the wrapper config.
-type Backend interface {
-	Name() string
-	BuildArgs(cfg *Config, targetArg string) []string
-	Command() string
-}
-
-type CodexBackend struct{}
-
-func (CodexBackend) Name() string { return "codex" }
-func (CodexBackend) Command() string {
-	return "codex"
-}
-func (CodexBackend) BuildArgs(cfg *Config, targetArg string) []string {
-	return buildCodexArgs(cfg, targetArg)
-}
-
-type ClaudeBackend struct{}
-
-func (ClaudeBackend) Name() string { return "claude" }
-func (ClaudeBackend) Command() string {
-	return "claude"
-}
-func (ClaudeBackend) BuildArgs(cfg *Config, targetArg string) []string {
-	return buildClaudeArgs(cfg, targetArg)
-}
-
-const maxClaudeSettingsBytes = 1 << 20 // 1MB
-
-type minimalClaudeSettings struct {
-	Env   map[string]string
-	Model string
-}
-
-// loadMinimalClaudeSettings 从 ~/.claude/settings.json 只提取安全的最小子集：
-// - env: 只接受字符串类型的值
-// - model: 只接受字符串类型的值
-// 文件缺失/解析失败/超限都返回空。
-func loadMinimalClaudeSettings() minimalClaudeSettings {
-	home, err := os.UserHomeDir()
-	if err != nil || home == "" {
-		return minimalClaudeSettings{}
-	}
-
-	settingPath := filepath.Join(home, ".claude", "settings.json")
-	info, err := os.Stat(settingPath)
-	if err != nil || info.Size() > maxClaudeSettingsBytes {
-		return minimalClaudeSettings{}
-	}
-
-	data, err := os.ReadFile(settingPath)
-	if err != nil {
-		return minimalClaudeSettings{}
-	}
-
-	var cfg struct {
-		Env   map[string]any `json:"env"`
-		Model any            `json:"model"`
-	}
-	if err := json.Unmarshal(data, &cfg); err != nil {
-		return minimalClaudeSettings{}
-	}
-
-	out := minimalClaudeSettings{}
-
-	if model, ok := cfg.Model.(string); ok {
-		out.Model = strings.TrimSpace(model)
-	}
-
-	if len(cfg.Env) == 0 {
-		return out
-	}
-
-	env := make(map[string]string, len(cfg.Env))
-	for k, v := range cfg.Env {
-		s, ok := v.(string)
-		if !ok {
-			continue
-		}
-		env[k] = s
-	}
-	if len(env) == 0 {
-		return out
-	}
-	out.Env = env
-	return out
-}
-
-// loadMinimalEnvSettings is kept for backwards tests; prefer loadMinimalClaudeSettings.
-func loadMinimalEnvSettings() map[string]string {
-	settings := loadMinimalClaudeSettings()
-	if len(settings.Env) == 0 {
-		return nil
-	}
-	return settings.Env
-}
-
-// loadGeminiEnv loads environment variables from ~/.gemini/.env
-// Supports GEMINI_API_KEY, GEMINI_MODEL, GOOGLE_GEMINI_BASE_URL
-// Also sets GEMINI_API_KEY_AUTH_MECHANISM=bearer for third-party API compatibility
-func loadGeminiEnv() map[string]string {
-	home, err := os.UserHomeDir()
-	if err != nil || home == "" {
-		return nil
-	}
-
-	envPath := filepath.Join(home, ".gemini", ".env")
-	data, err := os.ReadFile(envPath)
-	if err != nil {
-		return nil
-	}
-
-	env := make(map[string]string)
-	for _, line := range strings.Split(string(data), "\n") {
-		line = strings.TrimSpace(line)
-		if line == "" || strings.HasPrefix(line, "#") {
-			continue
-		}
-		idx := strings.IndexByte(line, '=')
-		if idx <= 0 {
-			continue
-		}
-		key := strings.TrimSpace(line[:idx])
-		value := strings.TrimSpace(line[idx+1:])
-		if key != "" && value != "" {
-			env[key] = value
-		}
-	}
-
-	// Set bearer auth mechanism for third-party API compatibility
-	if _, ok := env["GEMINI_API_KEY"]; ok {
-		if _, hasAuth := env["GEMINI_API_KEY_AUTH_MECHANISM"]; !hasAuth {
-			env["GEMINI_API_KEY_AUTH_MECHANISM"] = "bearer"
-		}
-	}
-
-	if len(env) == 0 {
-		return nil
-	}
-	return env
-}
-
-func buildClaudeArgs(cfg *Config, targetArg string) []string {
-	if cfg == nil {
-		return nil
-	}
-	args := []string{"-p"}
-	// Default to skip permissions unless CODEAGENT_SKIP_PERMISSIONS=false
-	if cfg.SkipPermissions || cfg.Yolo || envFlagDefaultTrue("CODEAGENT_SKIP_PERMISSIONS") {
-		args = append(args, "--dangerously-skip-permissions")
-	}
-
-	// Prevent infinite recursion: disable all setting sources (user, project, local)
-	// This ensures a clean execution environment without CLAUDE.md or skills that would trigger codeagent
-	args = append(args, "--setting-sources", "")
-
-	if model := strings.TrimSpace(cfg.Model); model != "" {
-		args = append(args, "--model", model)
-	}
-
-	if cfg.Mode == "resume" {
-		if cfg.SessionID != "" {
-			// Claude CLI uses -r <session_id> for resume.
-			args = append(args, "-r", cfg.SessionID)
-		}
-	}
-	// Note: claude CLI doesn't support -C flag; workdir set via cmd.Dir
-
-	args = append(args, "--output-format", "stream-json", "--verbose", targetArg)
-
-	return args
-}
-
-type GeminiBackend struct{}
-
-func (GeminiBackend) Name() string { return "gemini" }
-func (GeminiBackend) Command() string {
-	return "gemini"
-}
-func (GeminiBackend) BuildArgs(cfg *Config, targetArg string) []string {
-	return buildGeminiArgs(cfg, targetArg)
-}
-
-type OpencodeBackend struct{}
-
-func (OpencodeBackend) Name() string    { return "opencode" }
-func (OpencodeBackend) Command() string { return "opencode" }
-func (OpencodeBackend) BuildArgs(cfg *Config, targetArg string) []string {
-	args := []string{"run"}
-	if model := strings.TrimSpace(cfg.Model); model != "" {
-		args = append(args, "-m", model)
-	}
-	if cfg.Mode == "resume" && cfg.SessionID != "" {
-		args = append(args, "-s", cfg.SessionID)
-	}
-	args = append(args, "--format", "json", targetArg)
-	return args
-}
-
-func buildGeminiArgs(cfg *Config, targetArg string) []string {
-	if cfg == nil {
-		return nil
-	}
-	args := []string{"-o", "stream-json", "-y"}
-
-	if model := strings.TrimSpace(cfg.Model); model != "" {
-		args = append(args, "-m", model)
-	}
-
-	if cfg.Mode == "resume" {
-		if cfg.SessionID != "" {
-			args = append(args, "-r", cfg.SessionID)
-		}
-	}
-	// Note: gemini CLI doesn't support -C flag; workdir set via cmd.Dir
-
-	// Use positional argument instead of deprecated -p flag
-	// For stdin mode ("-"), use -p to read from stdin
-	if targetArg == "-" {
-		args = append(args, "-p", targetArg)
-	} else {
-		args = append(args, targetArg)
-	}
-
-	return args
-}
--- a/codeagent-wrapper/bench_test.go
+++ b/codeagent-wrapper/bench_test.go
@@ -1,39 +0,0 @@
-package main
-
-import (
-	"testing"
-)
-
-// BenchmarkLoggerWrite 测试日志写入性能
-func BenchmarkLoggerWrite(b *testing.B) {
-	logger, err := NewLogger()
-	if err != nil {
-		b.Fatal(err)
-	}
-	defer logger.Close()
-
-	b.ResetTimer()
-	for i := 0; i < b.N; i++ {
-		logger.Info("benchmark log message")
-	}
-	b.StopTimer()
-	logger.Flush()
-}
-
-// BenchmarkLoggerConcurrentWrite 测试并发日志写入性能
-func BenchmarkLoggerConcurrentWrite(b *testing.B) {
-	logger, err := NewLogger()
-	if err != nil {
-		b.Fatal(err)
-	}
-	defer logger.Close()
-
-	b.ResetTimer()
-	b.RunParallel(func(pb *testing.PB) {
-		for pb.Next() {
-			logger.Info("concurrent benchmark log message")
-		}
-	})
-	b.StopTimer()
-	logger.Flush()
-}
--- a/codeagent-wrapper/cmd/codeagent/main.go
+++ b/codeagent-wrapper/cmd/codeagent/main.go
@@ -0,0 +1,7 @@
+package main
+
+import app "codeagent-wrapper/internal/app"
+
+func main() {
+	app.Run()
+}
--- a/codeagent-wrapper/config.go
+++ b/codeagent-wrapper/config.go
@@ -1,473 +0,0 @@
-package main
-
-import (
-	"bytes"
-	"context"
-	"fmt"
-	"os"
-	"strconv"
-	"strings"
-)
-
-// Config holds CLI configuration
-type Config struct {
-	Mode               string // "new" or "resume"
-	Task               string
-	SessionID          string
-	WorkDir            string
-	Model              string
-	ReasoningEffort    string
-	ExplicitStdin      bool
-	Timeout            int
-	Backend            string
-	Agent              string
-	PromptFile         string
-	PromptFileExplicit bool
-	SkipPermissions    bool
-	Yolo               bool
-	MaxParallelWorkers int
-}
-
-// ParallelConfig defines the JSON schema for parallel execution
-type ParallelConfig struct {
-	Tasks         []TaskSpec `json:"tasks"`
-	GlobalBackend string     `json:"backend,omitempty"`
-}
-
-// TaskSpec describes an individual task entry in the parallel config
-type TaskSpec struct {
-	ID              string          `json:"id"`
-	Task            string          `json:"task"`
-	WorkDir         string          `json:"workdir,omitempty"`
-	Dependencies    []string        `json:"dependencies,omitempty"`
-	SessionID       string          `json:"session_id,omitempty"`
-	Backend         string          `json:"backend,omitempty"`
-	Model           string          `json:"model,omitempty"`
-	ReasoningEffort string          `json:"reasoning_effort,omitempty"`
-	Agent           string          `json:"agent,omitempty"`
-	PromptFile      string          `json:"prompt_file,omitempty"`
-	SkipPermissions bool            `json:"skip_permissions,omitempty"`
-	Mode            string          `json:"-"`
-	UseStdin        bool            `json:"-"`
-	Context         context.Context `json:"-"`
-}
-
-// TaskResult captures the execution outcome of a task
-type TaskResult struct {
-	TaskID    string `json:"task_id"`
-	ExitCode  int    `json:"exit_code"`
-	Message   string `json:"message"`
-	SessionID string `json:"session_id"`
-	Error     string `json:"error"`
-	LogPath   string `json:"log_path"`
-	// Structured report fields
-	Coverage       string   `json:"coverage,omitempty"`        // extracted coverage percentage (e.g., "92%")
-	CoverageNum    float64  `json:"coverage_num,omitempty"`    // numeric coverage for comparison
-	CoverageTarget float64  `json:"coverage_target,omitempty"` // target coverage (default 90)
-	FilesChanged   []string `json:"files_changed,omitempty"`   // list of changed files
-	KeyOutput      string   `json:"key_output,omitempty"`      // brief summary of what was done
-	TestsPassed    int      `json:"tests_passed,omitempty"`    // number of tests passed
-	TestsFailed    int      `json:"tests_failed,omitempty"`    // number of tests failed
-	sharedLog      bool
-}
-
-var backendRegistry = map[string]Backend{
-	"codex":    CodexBackend{},
-	"claude":   ClaudeBackend{},
-	"gemini":   GeminiBackend{},
-	"opencode": OpencodeBackend{},
-}
-
-func selectBackend(name string) (Backend, error) {
-	key := strings.ToLower(strings.TrimSpace(name))
-	if key == "" {
-		key = defaultBackendName
-	}
-	if backend, ok := backendRegistry[key]; ok {
-		return backend, nil
-	}
-	return nil, fmt.Errorf("unsupported backend %q", name)
-}
-
-func envFlagEnabled(key string) bool {
-	val, ok := os.LookupEnv(key)
-	if !ok {
-		return false
-	}
-	val = strings.TrimSpace(strings.ToLower(val))
-	switch val {
-	case "", "0", "false", "no", "off":
-		return false
-	default:
-		return true
-	}
-}
-
-func parseBoolFlag(val string, defaultValue bool) bool {
-	val = strings.TrimSpace(strings.ToLower(val))
-	switch val {
-	case "1", "true", "yes", "on":
-		return true
-	case "0", "false", "no", "off":
-		return false
-	default:
-		return defaultValue
-	}
-}
-
-// envFlagDefaultTrue returns true unless the env var is explicitly set to false/0/no/off.
-func envFlagDefaultTrue(key string) bool {
-	val, ok := os.LookupEnv(key)
-	if !ok {
-		return true
-	}
-	return parseBoolFlag(val, true)
-}
-
-func validateAgentName(name string) error {
-	if strings.TrimSpace(name) == "" {
-		return fmt.Errorf("agent name is empty")
-	}
-	for _, r := range name {
-		switch {
-		case r >= 'a' && r <= 'z':
-		case r >= 'A' && r <= 'Z':
-		case r >= '0' && r <= '9':
-		case r == '-', r == '_':
-		default:
-			return fmt.Errorf("agent name %q contains invalid character %q", name, r)
-		}
-	}
-	return nil
-}
-
-func parseParallelConfig(data []byte) (*ParallelConfig, error) {
-	trimmed := bytes.TrimSpace(data)
-	if len(trimmed) == 0 {
-		return nil, fmt.Errorf("parallel config is empty")
-	}
-
-	tasks := strings.Split(string(trimmed), "---TASK---")
-	var cfg ParallelConfig
-	seen := make(map[string]struct{})
-
-	taskIndex := 0
-	for _, taskBlock := range tasks {
-		taskBlock = strings.TrimSpace(taskBlock)
-		if taskBlock == "" {
-			continue
-		}
-		taskIndex++
-
-		parts := strings.SplitN(taskBlock, "---CONTENT---", 2)
-		if len(parts) != 2 {
-			return nil, fmt.Errorf("task block #%d missing ---CONTENT--- separator", taskIndex)
-		}
-
-		meta := strings.TrimSpace(parts[0])
-		content := strings.TrimSpace(parts[1])
-
-		task := TaskSpec{WorkDir: defaultWorkdir}
-		agentSpecified := false
-		for _, line := range strings.Split(meta, "\n") {
-			line = strings.TrimSpace(line)
-			if line == "" {
-				continue
-			}
-			kv := strings.SplitN(line, ":", 2)
-			if len(kv) != 2 {
-				continue
-			}
-			key := strings.TrimSpace(kv[0])
-			value := strings.TrimSpace(kv[1])
-
-			switch key {
-			case "id":
-				task.ID = value
-			case "workdir":
-				// Validate workdir: "-" is not a valid directory
-				if value == "-" {
-					return nil, fmt.Errorf("task block #%d has invalid workdir: '-' is not a valid directory path", taskIndex)
-				}
-				task.WorkDir = value
-			case "session_id":
-				task.SessionID = value
-				task.Mode = "resume"
-			case "backend":
-				task.Backend = value
-			case "model":
-				task.Model = value
-			case "reasoning_effort":
-				task.ReasoningEffort = value
-			case "agent":
-				agentSpecified = true
-				task.Agent = value
-			case "skip_permissions", "skip-permissions":
-				if value == "" {
-					task.SkipPermissions = true
-					continue
-				}
-				task.SkipPermissions = parseBoolFlag(value, false)
-			case "dependencies":
-				for _, dep := range strings.Split(value, ",") {
-					dep = strings.TrimSpace(dep)
-					if dep != "" {
-						task.Dependencies = append(task.Dependencies, dep)
-					}
-				}
-			}
-		}
-
-		if task.Mode == "" {
-			task.Mode = "new"
-		}
-
-		if agentSpecified {
-			if strings.TrimSpace(task.Agent) == "" {
-				return nil, fmt.Errorf("task block #%d has empty agent field", taskIndex)
-			}
-			if err := validateAgentName(task.Agent); err != nil {
-				return nil, fmt.Errorf("task block #%d invalid agent name: %w", taskIndex, err)
-			}
-			backend, model, promptFile, reasoning, _ := resolveAgentConfig(task.Agent)
-			if task.Backend == "" {
-				task.Backend = backend
-			}
-			if task.Model == "" {
-				task.Model = model
-			}
-			if task.ReasoningEffort == "" {
-				task.ReasoningEffort = reasoning
-			}
-			task.PromptFile = promptFile
-		}
-
-		if task.ID == "" {
-			return nil, fmt.Errorf("task block #%d missing id field", taskIndex)
-		}
-		if content == "" {
-			return nil, fmt.Errorf("task block #%d (%q) missing content", taskIndex, task.ID)
-		}
-		if task.Mode == "resume" && strings.TrimSpace(task.SessionID) == "" {
-			return nil, fmt.Errorf("task block #%d (%q) has empty session_id", taskIndex, task.ID)
-		}
-		if _, exists := seen[task.ID]; exists {
-			return nil, fmt.Errorf("task block #%d has duplicate id: %s", taskIndex, task.ID)
-		}
-
-		task.Task = content
-		cfg.Tasks = append(cfg.Tasks, task)
-		seen[task.ID] = struct{}{}
-	}
-
-	if len(cfg.Tasks) == 0 {
-		return nil, fmt.Errorf("no tasks found")
-	}
-
-	return &cfg, nil
-}
-
-func parseArgs() (*Config, error) {
-	args := os.Args[1:]
-	if len(args) == 0 {
-		return nil, fmt.Errorf("task required")
-	}
-
-	backendName := defaultBackendName
-	model := ""
-	reasoningEffort := ""
-	agentName := ""
-	promptFile := ""
-	promptFileExplicit := false
-	yolo := false
-	skipPermissions := envFlagEnabled("CODEAGENT_SKIP_PERMISSIONS")
-	filtered := make([]string, 0, len(args))
-	for i := 0; i < len(args); i++ {
-		arg := args[i]
-		switch {
-		case arg == "--agent":
-			if i+1 >= len(args) {
-				return nil, fmt.Errorf("--agent flag requires a value")
-			}
-			value := strings.TrimSpace(args[i+1])
-			if value == "" {
-				return nil, fmt.Errorf("--agent flag requires a value")
-			}
-			if err := validateAgentName(value); err != nil {
-				return nil, fmt.Errorf("--agent flag invalid value: %w", err)
-			}
-			resolvedBackend, resolvedModel, resolvedPromptFile, resolvedReasoning, resolvedYolo := resolveAgentConfig(value)
-			backendName = resolvedBackend
-			model = resolvedModel
-			if !promptFileExplicit {
-				promptFile = resolvedPromptFile
-			}
-			if reasoningEffort == "" {
-				reasoningEffort = resolvedReasoning
-			}
-			yolo = resolvedYolo
-			agentName = value
-			i++
-			continue
-		case strings.HasPrefix(arg, "--agent="):
-			value := strings.TrimSpace(strings.TrimPrefix(arg, "--agent="))
-			if value == "" {
-				return nil, fmt.Errorf("--agent flag requires a value")
-			}
-			if err := validateAgentName(value); err != nil {
-				return nil, fmt.Errorf("--agent flag invalid value: %w", err)
-			}
-			resolvedBackend, resolvedModel, resolvedPromptFile, resolvedReasoning, resolvedYolo := resolveAgentConfig(value)
-			backendName = resolvedBackend
-			model = resolvedModel
-			if !promptFileExplicit {
-				promptFile = resolvedPromptFile
-			}
-			if reasoningEffort == "" {
-				reasoningEffort = resolvedReasoning
-			}
-			yolo = resolvedYolo
-			agentName = value
-			continue
-		case arg == "--prompt-file":
-			if i+1 >= len(args) {
-				return nil, fmt.Errorf("--prompt-file flag requires a value")
-			}
-			value := strings.TrimSpace(args[i+1])
-			if value == "" {
-				return nil, fmt.Errorf("--prompt-file flag requires a value")
-			}
-			promptFile = value
-			promptFileExplicit = true
-			i++
-			continue
-		case strings.HasPrefix(arg, "--prompt-file="):
-			value := strings.TrimSpace(strings.TrimPrefix(arg, "--prompt-file="))
-			if value == "" {
-				return nil, fmt.Errorf("--prompt-file flag requires a value")
-			}
-			promptFile = value
-			promptFileExplicit = true
-			continue
-		case arg == "--backend":
-			if i+1 >= len(args) {
-				return nil, fmt.Errorf("--backend flag requires a value")
-			}
-			backendName = args[i+1]
-			i++
-			continue
-		case strings.HasPrefix(arg, "--backend="):
-			value := strings.TrimPrefix(arg, "--backend=")
-			if value == "" {
-				return nil, fmt.Errorf("--backend flag requires a value")
-			}
-			backendName = value
-			continue
-		case arg == "--skip-permissions", arg == "--dangerously-skip-permissions":
-			skipPermissions = true
-			continue
-		case arg == "--model":
-			if i+1 >= len(args) {
-				return nil, fmt.Errorf("--model flag requires a value")
-			}
-			model = args[i+1]
-			i++
-			continue
-		case strings.HasPrefix(arg, "--model="):
-			value := strings.TrimPrefix(arg, "--model=")
-			if value == "" {
-				return nil, fmt.Errorf("--model flag requires a value")
-			}
-			model = value
-			continue
-		case arg == "--reasoning-effort":
-			if i+1 >= len(args) {
-				return nil, fmt.Errorf("--reasoning-effort flag requires a value")
-			}
-			value := strings.TrimSpace(args[i+1])
-			if value == "" {
-				return nil, fmt.Errorf("--reasoning-effort flag requires a value")
-			}
-			reasoningEffort = value
-			i++
-			continue
-		case strings.HasPrefix(arg, "--reasoning-effort="):
-			value := strings.TrimSpace(strings.TrimPrefix(arg, "--reasoning-effort="))
-			if value == "" {
-				return nil, fmt.Errorf("--reasoning-effort flag requires a value")
-			}
-			reasoningEffort = value
-			continue
-		case strings.HasPrefix(arg, "--skip-permissions="):
-			skipPermissions = parseBoolFlag(strings.TrimPrefix(arg, "--skip-permissions="), skipPermissions)
-			continue
-		case strings.HasPrefix(arg, "--dangerously-skip-permissions="):
-			skipPermissions = parseBoolFlag(strings.TrimPrefix(arg, "--dangerously-skip-permissions="), skipPermissions)
-			continue
-		}
-		filtered = append(filtered, arg)
-	}
-
-	if len(filtered) == 0 {
-		return nil, fmt.Errorf("task required")
-	}
-	args = filtered
-
-	cfg := &Config{WorkDir: defaultWorkdir, Backend: backendName, Agent: agentName, PromptFile: promptFile, PromptFileExplicit: promptFileExplicit, SkipPermissions: skipPermissions, Yolo: yolo, Model: strings.TrimSpace(model), ReasoningEffort: strings.TrimSpace(reasoningEffort)}
-	cfg.MaxParallelWorkers = resolveMaxParallelWorkers()
-
-	if args[0] == "resume" {
-		if len(args) < 3 {
-			return nil, fmt.Errorf("resume mode requires: resume <session_id> <task>")
-		}
-		cfg.Mode = "resume"
-		cfg.SessionID = strings.TrimSpace(args[1])
-		if cfg.SessionID == "" {
-			return nil, fmt.Errorf("resume mode requires non-empty session_id")
-		}
-		cfg.Task = args[2]
-		cfg.ExplicitStdin = (args[2] == "-")
-		if len(args) > 3 {
-			// Validate workdir: "-" is not a valid directory
-			if args[3] == "-" {
-				return nil, fmt.Errorf("invalid workdir: '-' is not a valid directory path")
-			}
-			cfg.WorkDir = args[3]
-		}
-	} else {
-		cfg.Mode = "new"
-		cfg.Task = args[0]
-		cfg.ExplicitStdin = (args[0] == "-")
-		if len(args) > 1 {
-			// Validate workdir: "-" is not a valid directory
-			if args[1] == "-" {
-				return nil, fmt.Errorf("invalid workdir: '-' is not a valid directory path")
-			}
-			cfg.WorkDir = args[1]
-		}
-	}
-
-	return cfg, nil
-}
-
-const maxParallelWorkersLimit = 100
-
-func resolveMaxParallelWorkers() int {
-	raw := strings.TrimSpace(os.Getenv("CODEAGENT_MAX_PARALLEL_WORKERS"))
-	if raw == "" {
-		return 0
-	}
-
-	value, err := strconv.Atoi(raw)
-	if err != nil || value < 0 {
-		logWarn(fmt.Sprintf("Invalid CODEAGENT_MAX_PARALLEL_WORKERS=%q, falling back to unlimited", raw))
-		return 0
-	}
-
-	if value > maxParallelWorkersLimit {
-		logWarn(fmt.Sprintf("CODEAGENT_MAX_PARALLEL_WORKERS=%d exceeds limit, capping at %d", value, maxParallelWorkersLimit))
-		return maxParallelWorkersLimit
-	}
-
-	return value
-}
--- a/codeagent-wrapper/go.mod
+++ b/codeagent-wrapper/go.mod
@@ -1,3 +1,43 @@
 module codeagent-wrapper

 go 1.21
+
+require (
+	github.com/goccy/go-json v0.10.5
+	github.com/rs/zerolog v1.34.0
+	github.com/shirou/gopsutil/v3 v3.24.5
+	github.com/spf13/cobra v1.8.1
+	github.com/spf13/pflag v1.0.5
+	github.com/spf13/viper v1.19.0
+)
+
+require (
+	github.com/fsnotify/fsnotify v1.7.0 // indirect
+	github.com/go-ole/go-ole v1.2.6 // indirect
+	github.com/hashicorp/hcl v1.0.0 // indirect
+	github.com/inconshreveable/mousetrap v1.1.0 // indirect
+	github.com/lufia/plan9stats v0.0.0-20211012122336-39d0f177ccd0 // indirect
+	github.com/magiconair/properties v1.8.7 // indirect
+	github.com/mattn/go-colorable v0.1.13 // indirect
+	github.com/mattn/go-isatty v0.0.19 // indirect
+	github.com/mitchellh/mapstructure v1.5.0 // indirect
+	github.com/pelletier/go-toml/v2 v2.2.2 // indirect
+	github.com/power-devops/perfstat v0.0.0-20210106213030-5aafc221ea8c // indirect
+	github.com/sagikazarmark/locafero v0.4.0 // indirect
+	github.com/sagikazarmark/slog-shim v0.1.0 // indirect
+	github.com/shoenig/go-m1cpu v0.1.6 // indirect
+	github.com/sourcegraph/conc v0.3.0 // indirect
+	github.com/spf13/afero v1.11.0 // indirect
+	github.com/spf13/cast v1.6.0 // indirect
+	github.com/subosito/gotenv v1.6.0 // indirect
+	github.com/tklauser/go-sysconf v0.3.12 // indirect
+	github.com/tklauser/numcpus v0.6.1 // indirect
+	github.com/yusufpapurcu/wmi v1.2.4 // indirect
+	go.uber.org/atomic v1.9.0 // indirect
+	go.uber.org/multierr v1.9.0 // indirect
+	golang.org/x/exp v0.0.0-20230905200255-921286631fa9 // indirect
+	golang.org/x/sys v0.20.0 // indirect
+	golang.org/x/text v0.14.0 // indirect
+	gopkg.in/ini.v1 v1.67.0 // indirect
+	gopkg.in/yaml.v3 v3.0.1 // indirect
+)
--- a/codeagent-wrapper/go.sum
+++ b/codeagent-wrapper/go.sum
@@ -0,0 +1,117 @@
+github.com/coreos/go-systemd/v22 v22.5.0/go.mod h1:Y58oyj3AT4RCenI/lSvhwexgC+NSVTIJ3seZv2GcEnc=
+github.com/cpuguy83/go-md2man/v2 v2.0.4/go.mod h1:tgQtvFlXSQOSOSIRvRPT7W67SCa46tRHOmNcaadrF8o=
+github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
+github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
+github.com/davecgh/go-spew v1.1.2-0.20180830191138-d8f796af33cc h1:U9qPSI2PIWSS1VwoXQT9A3Wy9MM3WgvqSxFWenqJduM=
+github.com/davecgh/go-spew v1.1.2-0.20180830191138-d8f796af33cc/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
+github.com/frankban/quicktest v1.14.6 h1:7Xjx+VpznH+oBnejlPUj8oUpdxnVs4f8XU8WnHkI4W8=
+github.com/frankban/quicktest v1.14.6/go.mod h1:4ptaffx2x8+WTWXmUCuVU6aPUX1/Mz7zb5vbUoiM6w0=
+github.com/fsnotify/fsnotify v1.7.0 h1:8JEhPFa5W2WU7YfeZzPNqzMP6Lwt7L2715Ggo0nosvA=
+github.com/fsnotify/fsnotify v1.7.0/go.mod h1:40Bi/Hjc2AVfZrqy+aj+yEI+/bRxZnMJyTJwOpGvigM=
+github.com/go-ole/go-ole v1.2.6 h1:/Fpf6oFPoeFik9ty7siob0G6Ke8QvQEuVcuChpwXzpY=
+github.com/go-ole/go-ole v1.2.6/go.mod h1:pprOEPIfldk/42T2oK7lQ4v4JSDwmV0As9GaiUsvbm0=
+github.com/goccy/go-json v0.10.5 h1:Fq85nIqj+gXn/S5ahsiTlK3TmC85qgirsdTP/+DeaC4=
+github.com/goccy/go-json v0.10.5/go.mod h1:oq7eo15ShAhp70Anwd5lgX2pLfOS3QCiwU/PULtXL6M=
+github.com/godbus/dbus/v5 v5.0.4/go.mod h1:xhWf0FNVPg57R7Z0UbKHbJfkEywrmjJnf7w5xrFpKfA=
+github.com/google/go-cmp v0.5.6/go.mod h1:v8dTdLbMG2kIc/vJvl+f65V22dbkXbowE6jgT/gNBxE=
+github.com/google/go-cmp v0.6.0 h1:ofyhxvXcZhMsU5ulbFiLKl/XBFqE1GSq7atu8tAmTRI=
+github.com/google/go-cmp v0.6.0/go.mod h1:17dUlkBOakJ0+DkrSSNjCkIjxS6bF9zb3elmeNGIjoY=
+github.com/hashicorp/hcl v1.0.0 h1:0Anlzjpi4vEasTeNFn2mLJgTSwt0+6sfsiTG8qcWGx4=
+github.com/hashicorp/hcl v1.0.0/go.mod h1:E5yfLk+7swimpb2L/Alb/PJmXilQ/rhwaUYs4T20WEQ=
+github.com/inconshreveable/mousetrap v1.1.0 h1:wN+x4NVGpMsO7ErUn/mUI3vEoE6Jt13X2s0bqwp9tc8=
+github.com/inconshreveable/mousetrap v1.1.0/go.mod h1:vpF70FUmC8bwa3OWnCshd2FqLfsEA9PFc4w1p2J65bw=
+github.com/kr/pretty v0.3.1 h1:flRD4NNwYAUpkphVc1HcthR4KEIFJ65n8Mw5qdRn3LE=
+github.com/kr/pretty v0.3.1/go.mod h1:hoEshYVHaxMs3cyo3Yncou5ZscifuDolrwPKZanG3xk=
+github.com/kr/text v0.2.0 h1:5Nx0Ya0ZqY2ygV366QzturHI13Jq95ApcVaJBhpS+AY=
+github.com/kr/text v0.2.0/go.mod h1:eLer722TekiGuMkidMxC/pM04lWEeraHUUmBw8l2grE=
+github.com/lufia/plan9stats v0.0.0-20211012122336-39d0f177ccd0 h1:6E+4a0GO5zZEnZ81pIr0yLvtUWk2if982qA3F3QD6H4=
+github.com/lufia/plan9stats v0.0.0-20211012122336-39d0f177ccd0/go.mod h1:zJYVVT2jmtg6P3p1VtQj7WsuWi/y4VnjVBn7F8KPB3I=
+github.com/magiconair/properties v1.8.7 h1:IeQXZAiQcpL9mgcAe1Nu6cX9LLw6ExEHKjN0VQdvPDY=
+github.com/magiconair/properties v1.8.7/go.mod h1:Dhd985XPs7jluiymwWYZ0G4Z61jb3vdS329zhj2hYo0=
+github.com/mattn/go-colorable v0.1.13 h1:fFA4WZxdEF4tXPZVKMLwD8oUnCTTo08duU7wxecdEvA=
+github.com/mattn/go-colorable v0.1.13/go.mod h1:7S9/ev0klgBDR4GtXTXX8a3vIGJpMovkB8vQcUbaXHg=
+github.com/mattn/go-isatty v0.0.16/go.mod h1:kYGgaQfpe5nmfYZH+SKPsOc2e4SrIfOl2e/yFXSvRLM=
+github.com/mattn/go-isatty v0.0.19 h1:JITubQf0MOLdlGRuRq+jtsDlekdYPia9ZFsB8h/APPA=
+github.com/mattn/go-isatty v0.0.19/go.mod h1:W+V8PltTTMOvKvAeJH7IuucS94S2C6jfK/D7dTCTo3Y=
+github.com/mitchellh/mapstructure v1.5.0 h1:jeMsZIYE/09sWLaz43PL7Gy6RuMjD2eJVyuac5Z2hdY=
+github.com/mitchellh/mapstructure v1.5.0/go.mod h1:bFUtVrKA4DC2yAKiSyO/QUcy7e+RRV2QTWOzhPopBRo=
+github.com/pelletier/go-toml/v2 v2.2.2 h1:aYUidT7k73Pcl9nb2gScu7NSrKCSHIDE89b3+6Wq+LM=
+github.com/pelletier/go-toml/v2 v2.2.2/go.mod h1:1t835xjRzz80PqgE6HHgN2JOsmgYu/h4qDAS4n929Rs=
+github.com/pkg/errors v0.9.1/go.mod h1:bwawxfHBFNV+L2hUp1rHADufV3IMtnDRdf1r5NINEl0=
+github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
+github.com/pmezard/go-difflib v1.0.1-0.20181226105442-5d4384ee4fb2 h1:Jamvg5psRIccs7FGNTlIRMkT8wgtp5eCXdBlqhYGL6U=
+github.com/pmezard/go-difflib v1.0.1-0.20181226105442-5d4384ee4fb2/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
+github.com/power-devops/perfstat v0.0.0-20210106213030-5aafc221ea8c h1:ncq/mPwQF4JjgDlrVEn3C11VoGHZN7m8qihwgMEtzYw=
+github.com/power-devops/perfstat v0.0.0-20210106213030-5aafc221ea8c/go.mod h1:OmDBASR4679mdNQnz2pUhc2G8CO2JrUAVFDRBDP/hJE=
+github.com/rogpeppe/go-internal v1.9.0 h1:73kH8U+JUqXU8lRuOHeVHaa/SZPifC7BkcraZVejAe8=
+github.com/rogpeppe/go-internal v1.9.0/go.mod h1:WtVeX8xhTBvf0smdhujwtBcq4Qrzq/fJaraNFVN+nFs=
+github.com/rs/xid v1.6.0/go.mod h1:7XoLgs4eV+QndskICGsho+ADou8ySMSjJKDIan90Nz0=
+github.com/rs/zerolog v1.34.0 h1:k43nTLIwcTVQAncfCw4KZ2VY6ukYoZaBPNOE8txlOeY=
+github.com/rs/zerolog v1.34.0/go.mod h1:bJsvje4Z08ROH4Nhs5iH600c3IkWhwp44iRc54W6wYQ=
+github.com/russross/blackfriday/v2 v2.1.0/go.mod h1:+Rmxgy9KzJVeS9/2gXHxylqXiyQDYRxCVz55jmeOWTM=
+github.com/sagikazarmark/locafero v0.4.0 h1:HApY1R9zGo4DBgr7dqsTH/JJxLTTsOt7u6keLGt6kNQ=
+github.com/sagikazarmark/locafero v0.4.0/go.mod h1:Pe1W6UlPYUk/+wc/6KFhbORCfqzgYEpgQ3O5fPuL3H4=
+github.com/sagikazarmark/slog-shim v0.1.0 h1:diDBnUNK9N/354PgrxMywXnAwEr1QZcOr6gto+ugjYE=
+github.com/sagikazarmark/slog-shim v0.1.0/go.mod h1:SrcSrq8aKtyuqEI1uvTDTK1arOWRIczQRv+GVI1AkeQ=
+github.com/shirou/gopsutil/v3 v3.24.5 h1:i0t8kL+kQTvpAYToeuiVk3TgDeKOFioZO3Ztz/iZ9pI=
+github.com/shirou/gopsutil/v3 v3.24.5/go.mod h1:bsoOS1aStSs9ErQ1WWfxllSeS1K5D+U30r2NfcubMVk=
+github.com/shoenig/go-m1cpu v0.1.6 h1:nxdKQNcEB6vzgA2E2bvzKIYRuNj7XNJ4S/aRSwKzFtM=
+github.com/shoenig/go-m1cpu v0.1.6/go.mod h1:1JJMcUBvfNwpq05QDQVAnx3gUHr9IYF7GNg9SUEw2VQ=
+github.com/shoenig/test v0.6.4 h1:kVTaSd7WLz5WZ2IaoM0RSzRsUD+m8wRR+5qvntpn4LU=
+github.com/shoenig/test v0.6.4/go.mod h1:byHiCGXqrVaflBLAMq/srcZIHynQPQgeyvkvXnjqq0k=
+github.com/sourcegraph/conc v0.3.0 h1:OQTbbt6P72L20UqAkXXuLOj79LfEanQ+YQFNpLA9ySo=
+github.com/sourcegraph/conc v0.3.0/go.mod h1:Sdozi7LEKbFPqYX2/J+iBAM6HpqSLTASQIKqDmF7Mt0=
+github.com/spf13/afero v1.11.0 h1:WJQKhtpdm3v2IzqG8VMqrr6Rf3UYpEF239Jy9wNepM8=
+github.com/spf13/afero v1.11.0/go.mod h1:GH9Y3pIexgf1MTIWtNGyogA5MwRIDXGUr+hbWNoBjkY=
+github.com/spf13/cast v1.6.0 h1:GEiTHELF+vaR5dhz3VqZfFSzZjYbgeKDpBxQVS4GYJ0=
+github.com/spf13/cast v1.6.0/go.mod h1:ancEpBxwJDODSW/UG4rDrAqiKolqNNh2DX3mk86cAdo=
+github.com/spf13/cobra v1.8.1 h1:e5/vxKd/rZsfSJMUX1agtjeTDf+qv1/JdBF8gg5k9ZM=
+github.com/spf13/cobra v1.8.1/go.mod h1:wHxEcudfqmLYa8iTfL+OuZPbBZkmvliBWKIezN3kD9Y=
+github.com/spf13/pflag v1.0.5 h1:iy+VFUOCP1a+8yFto/drg2CJ5u0yRoB7fZw3DKv/JXA=
+github.com/spf13/pflag v1.0.5/go.mod h1:McXfInJRrz4CZXVZOBLb0bTZqETkiAhM9Iw0y3An2Bg=
+github.com/spf13/viper v1.19.0 h1:RWq5SEjt8o25SROyN3z2OrDB9l7RPd3lwTWU8EcEdcI=
+github.com/spf13/viper v1.19.0/go.mod h1:GQUN9bilAbhU/jgc1bKs99f/suXKeUMct8Adx5+Ntkg=
+github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
+github.com/stretchr/objx v0.4.0/go.mod h1:YvHI0jy2hoMjB+UWwv71VJQ9isScKT/TqJzVSSt89Yw=
+github.com/stretchr/objx v0.5.0/go.mod h1:Yh+to48EsGEfYuaHDzXPcE3xhTkx73EhmCGUpEOglKo=
+github.com/stretchr/objx v0.5.2/go.mod h1:FRsXN1f5AsAjCGJKqEizvkpNtU+EGNCLh3NxZ/8L+MA=
+github.com/stretchr/testify v1.3.0/go.mod h1:M5WIy9Dh21IEIfnGCwXGc5bZfKNJtfHm1UVUgZn+9EI=
+github.com/stretchr/testify v1.7.1/go.mod h1:6Fq8oRcR53rry900zMqJjRRixrwX3KX962/h/Wwjteg=
+github.com/stretchr/testify v1.8.0/go.mod h1:yNjHg4UonilssWZ8iaSj1OCr/vHnekPRkoO+kdMU+MU=
+github.com/stretchr/testify v1.8.4/go.mod h1:sz/lmYIOXD/1dqDmKjjqLyZ2RngseejIcXlSw2iwfAo=
+github.com/stretchr/testify v1.9.0 h1:HtqpIVDClZ4nwg75+f6Lvsy/wHu+3BoSGCbBAcpTsTg=
+github.com/stretchr/testify v1.9.0/go.mod h1:r2ic/lqez/lEtzL7wO/rwa5dbSLXVDPFyf8C91i36aY=
+github.com/subosito/gotenv v1.6.0 h1:9NlTDc1FTs4qu0DDq7AEtTPNw6SVm7uBMsUCUjABIf8=
+github.com/subosito/gotenv v1.6.0/go.mod h1:Dk4QP5c2W3ibzajGcXpNraDfq2IrhjMIvMSWPKKo0FU=
+github.com/tklauser/go-sysconf v0.3.12 h1:0QaGUFOdQaIVdPgfITYzaTegZvdCjmYO52cSFAEVmqU=
+github.com/tklauser/go-sysconf v0.3.12/go.mod h1:Ho14jnntGE1fpdOqQEEaiKRpvIavV0hSfmBq8nJbHYI=
+github.com/tklauser/numcpus v0.6.1 h1:ng9scYS7az0Bk4OZLvrNXNSAO2Pxr1XXRAPyjhIx+Fk=
+github.com/tklauser/numcpus v0.6.1/go.mod h1:1XfjsgE2zo8GVw7POkMbHENHzVg3GzmoZ9fESEdAacY=
+github.com/yusufpapurcu/wmi v1.2.4 h1:zFUKzehAFReQwLys1b/iSMl+JQGSCSjtVqQn9bBrPo0=
+github.com/yusufpapurcu/wmi v1.2.4/go.mod h1:SBZ9tNy3G9/m5Oi98Zks0QjeHVDvuK0qfxQmPyzfmi0=
+go.uber.org/atomic v1.9.0 h1:ECmE8Bn/WFTYwEW/bpKD3M8VtR/zQVbavAoalC1PYyE=
+go.uber.org/atomic v1.9.0/go.mod h1:fEN4uk6kAWBTFdckzkM89CLk9XfWZrxpCo0nPH17wJc=
+go.uber.org/multierr v1.9.0 h1:7fIwc/ZtS0q++VgcfqFDxSBZVv/Xo49/SYnDFupUwlI=
+go.uber.org/multierr v1.9.0/go.mod h1:X2jQV1h+kxSjClGpnseKVIxpmcjrj7MNnI0bnlfKTVQ=
+golang.org/x/exp v0.0.0-20230905200255-921286631fa9 h1:GoHiUyI/Tp2nVkLI2mCxVkOjsbSXD66ic0XW0js0R9g=
+golang.org/x/exp v0.0.0-20230905200255-921286631fa9/go.mod h1:S2oDrQGGwySpoQPVqRShND87VCbxmc6bL1Yd2oYrm6k=
+golang.org/x/sys v0.0.0-20190916202348-b4ddaad3f8a3/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
+golang.org/x/sys v0.0.0-20201204225414-ed752295db88/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
+golang.org/x/sys v0.0.0-20220811171246-fbc7d0a398ab/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
+golang.org/x/sys v0.6.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
+golang.org/x/sys v0.8.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
+golang.org/x/sys v0.11.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
+golang.org/x/sys v0.12.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
+golang.org/x/sys v0.20.0 h1:Od9JTbYCk261bKm4M/mw7AklTlFYIa0bIp9BgSm1S8Y=
+golang.org/x/sys v0.20.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
+golang.org/x/text v0.14.0 h1:ScX5w1eTa3QqT8oi6+ziP7dTV1S2+ALU0bI+0zXKWiQ=
+golang.org/x/text v0.14.0/go.mod h1:18ZOQIKpY8NJVqYksKHtTdi31H5itFRjB5/qKTNYzSU=
+golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543/go.mod h1:I/5z698sn9Ka8TeJc9MKroUUfqBBauWjQqLJ2OPfmY0=
+gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
+gopkg.in/check.v1 v1.0.0-20190902080502-41f04d3bba15 h1:YR8cESwS4TdDjEe65xsg0ogRM/Nc3DYOhEAlW+xobZo=
+gopkg.in/check.v1 v1.0.0-20190902080502-41f04d3bba15/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
+gopkg.in/ini.v1 v1.67.0 h1:Dgnx+6+nfE+IfzjUEISNeydPJh9AXNNsWbGP9KzCsOA=
+gopkg.in/ini.v1 v1.67.0/go.mod h1:pNLf8WUiyNEtQjuu5G5vTm06TEv9tsIgeAvK8hOrP4k=
+gopkg.in/yaml.v3 v3.0.0-20200313102051-9f266ea9e77c/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=
+gopkg.in/yaml.v3 v3.0.1 h1:fxVm/GzAzEWqLHuvctI91KS9hhNmmWOoWu0XTYJS7CA=
+gopkg.in/yaml.v3 v3.0.1/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=
--- a/codeagent-wrapper/internal/app/agent_validation_test.go
+++ b/codeagent-wrapper/internal/app/agent_validation_test.go
@@ -1,4 +1,4 @@
-package main
+package wrapper

 import (
 	"context"
@@ -6,6 +6,9 @@ import (
 	"path/filepath"
 	"testing"
 	"time"
+
+	config "codeagent-wrapper/internal/config"
+	executor "codeagent-wrapper/internal/executor"
 )

 func TestValidateAgentName(t *testing.T) {
@@ -28,7 +31,7 @@ func TestValidateAgentName(t *testing.T) {

 	for _, tt := range tests {
 		t.Run(tt.name, func(t *testing.T) {
-			err := validateAgentName(tt.input)
+			err := config.ValidateAgentName(tt.input)
 			if (err != nil) != tt.wantErr {
 				t.Fatalf("validateAgentName(%q) err=%v, wantErr=%v", tt.input, err, tt.wantErr)
 			}
@@ -59,6 +62,8 @@ func TestParseParallelConfig_ResolvesAgentPromptFile(t *testing.T) {
 	home := t.TempDir()
 	t.Setenv("HOME", home)
 	t.Setenv("USERPROFILE", home)
+	t.Cleanup(config.ResetModelsConfigCacheForTest)
+	config.ResetModelsConfigCacheForTest()

 	configDir := filepath.Join(home, ".codeagent")
 	if err := os.MkdirAll(configDir, 0o755); err != nil {
@@ -117,10 +122,8 @@ func TestDefaultRunCodexTaskFn_AppliesAgentPromptFile(t *testing.T) {
 		WaitDelay: 2 * time.Millisecond,
 	})

-	newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
-		return fake
-	}
-	selectBackendFn = func(name string) (Backend, error) {
+	_ = executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner { return fake })
+	_ = executor.SetSelectBackendFn(func(name string) (Backend, error) {
 		return testBackend{
 			name:    name,
 			command: "fake-cmd",
@@ -128,7 +131,7 @@ func TestDefaultRunCodexTaskFn_AppliesAgentPromptFile(t *testing.T) {
 				return []string{targetArg}
 			},
 		}, nil
-	}
+	})

 	res := defaultRunCodexTaskFn(TaskSpec{
 		ID:         "t",
--- a/codeagent-wrapper/internal/app/app.go
+++ b/codeagent-wrapper/internal/app/app.go
@@ -0,0 +1,278 @@
+package wrapper
+
+import (
+	"fmt"
+	"io"
+	"os"
+	"path/filepath"
+	"strings"
+	"time"
+)
+
+const (
+	version               = "6.1.2"
+	defaultWorkdir        = "."
+	defaultTimeout        = 7200 // seconds (2 hours)
+	defaultCoverageTarget = 90.0
+	codexLogLineLimit     = 1000
+	stdinSpecialChars     = "\n\\\"'`$"
+	stderrCaptureLimit    = 4 * 1024
+	defaultBackendName    = "codex"
+	defaultCodexCommand   = "codex"
+
+	// stdout close reasons
+	stdoutCloseReasonWait  = "wait-done"
+	stdoutCloseReasonDrain = "drain-timeout"
+	stdoutCloseReasonCtx   = "context-cancel"
+	stdoutDrainTimeout     = 100 * time.Millisecond
+)
+
+// Test hooks for dependency injection
+var (
+	stdinReader         io.Reader = os.Stdin
+	isTerminalFn                  = defaultIsTerminal
+	codexCommand                  = defaultCodexCommand
+	cleanupHook         func()
+	startupCleanupAsync = true
+
+	buildCodexArgsFn   = buildCodexArgs
+	selectBackendFn    = selectBackend
+	cleanupLogsFn      = cleanupOldLogs
+	defaultBuildArgsFn = buildCodexArgs
+	runTaskFn          = runCodexTask
+	exitFn             = os.Exit
+)
+
+func runStartupCleanup() {
+	if cleanupLogsFn == nil {
+		return
+	}
+	defer func() {
+		if r := recover(); r != nil {
+			logWarn(fmt.Sprintf("cleanupOldLogs panic: %v", r))
+		}
+	}()
+	if _, err := cleanupLogsFn(); err != nil {
+		logWarn(fmt.Sprintf("cleanupOldLogs error: %v", err))
+	}
+}
+
+func scheduleStartupCleanup() {
+	if !startupCleanupAsync {
+		runStartupCleanup()
+		return
+	}
+	if cleanupLogsFn == nil {
+		return
+	}
+	fn := cleanupLogsFn
+	go func() {
+		defer func() {
+			if r := recover(); r != nil {
+				logWarn(fmt.Sprintf("cleanupOldLogs panic: %v", r))
+			}
+		}()
+		if _, err := fn(); err != nil {
+			logWarn(fmt.Sprintf("cleanupOldLogs error: %v", err))
+		}
+	}()
+}
+
+func runCleanupMode() int {
+	if cleanupLogsFn == nil {
+		fmt.Fprintln(os.Stderr, "Cleanup failed: log cleanup function not configured")
+		return 1
+	}
+
+	stats, err := cleanupLogsFn()
+	if err != nil {
+		fmt.Fprintf(os.Stderr, "Cleanup failed: %v\n", err)
+		return 1
+	}
+
+	fmt.Println("Cleanup completed")
+	fmt.Printf("Files scanned: %d\n", stats.Scanned)
+	fmt.Printf("Files deleted: %d\n", stats.Deleted)
+	if len(stats.DeletedFiles) > 0 {
+		for _, f := range stats.DeletedFiles {
+			fmt.Printf("  - %s\n", f)
+		}
+	}
+	fmt.Printf("Files kept: %d\n", stats.Kept)
+	if len(stats.KeptFiles) > 0 {
+		for _, f := range stats.KeptFiles {
+			fmt.Printf("  - %s\n", f)
+		}
+	}
+	if stats.Errors > 0 {
+		fmt.Printf("Deletion errors: %d\n", stats.Errors)
+	}
+	return 0
+}
+
+func readAgentPromptFile(path string, allowOutsideClaudeDir bool) (string, error) {
+	raw := strings.TrimSpace(path)
+	if raw == "" {
+		return "", nil
+	}
+
+	expanded := raw
+	if raw == "~" || strings.HasPrefix(raw, "~/") || strings.HasPrefix(raw, "~\\") {
+		home, err := os.UserHomeDir()
+		if err != nil {
+			return "", err
+		}
+		if raw == "~" {
+			expanded = home
+		} else {
+			expanded = home + raw[1:]
+		}
+	}
+
+	absPath, err := filepath.Abs(expanded)
+	if err != nil {
+		return "", err
+	}
+	absPath = filepath.Clean(absPath)
+
+	home, err := os.UserHomeDir()
+	if err != nil {
+		if !allowOutsideClaudeDir {
+			return "", err
+		}
+		logWarn(fmt.Sprintf("Failed to resolve home directory for prompt file validation: %v; proceeding without restriction", err))
+	} else {
+		allowedDirs := []string{
+			filepath.Clean(filepath.Join(home, ".claude")),
+			filepath.Clean(filepath.Join(home, ".codeagent", "agents")),
+		}
+		for i := range allowedDirs {
+			allowedAbs, err := filepath.Abs(allowedDirs[i])
+			if err == nil {
+				allowedDirs[i] = filepath.Clean(allowedAbs)
+			}
+		}
+
+		isWithinDir := func(path, dir string) bool {
+			rel, err := filepath.Rel(dir, path)
+			if err != nil {
+				return false
+			}
+			rel = filepath.Clean(rel)
+			if rel == "." {
+				return true
+			}
+			if rel == ".." {
+				return false
+			}
+			prefix := ".." + string(os.PathSeparator)
+			return !strings.HasPrefix(rel, prefix)
+		}
+
+		if !allowOutsideClaudeDir {
+			withinAllowed := false
+			for _, dir := range allowedDirs {
+				if isWithinDir(absPath, dir) {
+					withinAllowed = true
+					break
+				}
+			}
+			if !withinAllowed {
+				logWarn(fmt.Sprintf("Refusing to read prompt file outside allowed dirs (%s): %s", strings.Join(allowedDirs, ", "), absPath))
+				return "", fmt.Errorf("prompt file must be under ~/.claude or ~/.codeagent/agents")
+			}
+
+			resolvedPath, errPath := filepath.EvalSymlinks(absPath)
+			if errPath == nil {
+				resolvedPath = filepath.Clean(resolvedPath)
+				resolvedAllowed := make([]string, 0, len(allowedDirs))
+				for _, dir := range allowedDirs {
+					resolvedBase, errBase := filepath.EvalSymlinks(dir)
+					if errBase != nil {
+						continue
+					}
+					resolvedAllowed = append(resolvedAllowed, filepath.Clean(resolvedBase))
+				}
+				if len(resolvedAllowed) > 0 {
+					withinResolved := false
+					for _, dir := range resolvedAllowed {
+						if isWithinDir(resolvedPath, dir) {
+							withinResolved = true
+							break
+						}
+					}
+					if !withinResolved {
+						logWarn(fmt.Sprintf("Refusing to read prompt file outside allowed dirs (%s) (resolved): %s", strings.Join(resolvedAllowed, ", "), resolvedPath))
+						return "", fmt.Errorf("prompt file must be under ~/.claude or ~/.codeagent/agents")
+					}
+				}
+			}
+		} else {
+			withinAllowed := false
+			for _, dir := range allowedDirs {
+				if isWithinDir(absPath, dir) {
+					withinAllowed = true
+					break
+				}
+			}
+			if !withinAllowed {
+				logWarn(fmt.Sprintf("Reading prompt file outside allowed dirs (%s): %s", strings.Join(allowedDirs, ", "), absPath))
+			}
+		}
+	}
+
+	data, err := os.ReadFile(absPath)
+	if err != nil {
+		return "", err
+	}
+	return strings.TrimRight(string(data), "\r\n"), nil
+}
+
+func wrapTaskWithAgentPrompt(prompt string, task string) string {
+	return "<agent-prompt>\n" + prompt + "\n</agent-prompt>\n\n" + task
+}
+
+func runCleanupHook() {
+	if logger := activeLogger(); logger != nil {
+		logger.Flush()
+	}
+	if cleanupHook != nil {
+		cleanupHook()
+	}
+}
+
+func printHelp() {
+	name := currentWrapperName()
+	help := fmt.Sprintf(`%[1]s - Go wrapper for AI CLI backends
+
+Usage:
+    %[1]s "task" [workdir]
+    %[1]s --backend claude "task" [workdir]
+    %[1]s --prompt-file /path/to/prompt.md "task" [workdir]
+    %[1]s - [workdir]              Read task from stdin
+    %[1]s resume <session_id> "task" [workdir]
+    %[1]s resume <session_id> - [workdir]
+    %[1]s --parallel               Run tasks in parallel (config from stdin)
+    %[1]s --parallel --full-output Run tasks in parallel with full output (legacy)
+    %[1]s --version
+    %[1]s --help
+
+Parallel mode examples:
+    %[1]s --parallel < tasks.txt
+    echo '...' | %[1]s --parallel
+    %[1]s --parallel --full-output < tasks.txt
+    %[1]s --parallel <<'EOF'
+
+Environment Variables:
+    CODEX_TIMEOUT         Timeout in milliseconds (default: 7200000)
+    CODEAGENT_ASCII_MODE  Use ASCII symbols instead of Unicode (PASS/WARN/FAIL)
+
+Exit Codes:
+    0    Success
+    1    General error (missing args, no output)
+    124  Timeout
+    127  backend command not found
+    130  Interrupted (Ctrl+C)
+    *    Passthrough from backend process`, name)
+	fmt.Println(help)
+}
--- a/codeagent-wrapper/internal/app/backend.go
+++ b/codeagent-wrapper/internal/app/backend.go
@@ -0,0 +1,9 @@
+package wrapper
+
+import backend "codeagent-wrapper/internal/backend"
+
+type Backend = backend.Backend
+type CodexBackend = backend.CodexBackend
+type ClaudeBackend = backend.ClaudeBackend
+type GeminiBackend = backend.GeminiBackend
+type OpencodeBackend = backend.OpencodeBackend
--- a/codeagent-wrapper/internal/app/backend_init.go
+++ b/codeagent-wrapper/internal/app/backend_init.go
@@ -0,0 +1,7 @@
+package wrapper
+
+import backend "codeagent-wrapper/internal/backend"
+
+func init() {
+	backend.SetLogFuncs(logWarn, logError)
+}
--- a/codeagent-wrapper/internal/app/backend_registry.go
+++ b/codeagent-wrapper/internal/app/backend_registry.go
@@ -0,0 +1,5 @@
+package wrapper
+
+import backend "codeagent-wrapper/internal/backend"
+
+func selectBackend(name string) (Backend, error) { return backend.Select(name) }
--- a/codeagent-wrapper/internal/app/bench_test.go
+++ b/codeagent-wrapper/internal/app/bench_test.go
@@ -0,0 +1,103 @@
+package wrapper
+
+import (
+	"bytes"
+	"os"
+	"testing"
+
+	config "codeagent-wrapper/internal/config"
+)
+
+var (
+	benchCmdSink      any
+	benchConfigSink   *Config
+	benchMessageSink  string
+	benchThreadIDSink string
+)
+
+// BenchmarkStartup_NewRootCommand measures CLI startup overhead (command+flags construction).
+func BenchmarkStartup_NewRootCommand(b *testing.B) {
+	b.ReportAllocs()
+	for i := 0; i < b.N; i++ {
+		benchCmdSink = newRootCommand()
+	}
+}
+
+// BenchmarkConfigParse_ParseArgs measures config parsing from argv/env (steady-state).
+func BenchmarkConfigParse_ParseArgs(b *testing.B) {
+	home := b.TempDir()
+	b.Setenv("HOME", home)
+	b.Setenv("USERPROFILE", home)
+
+	config.ResetModelsConfigCacheForTest()
+	b.Cleanup(config.ResetModelsConfigCacheForTest)
+
+	origArgs := os.Args
+	os.Args = []string{"codeagent-wrapper", "--agent", "develop", "task"}
+	b.Cleanup(func() { os.Args = origArgs })
+
+	if _, err := parseArgs(); err != nil {
+		b.Fatalf("warmup parseArgs() error: %v", err)
+	}
+
+	b.ReportAllocs()
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		cfg, err := parseArgs()
+		if err != nil {
+			b.Fatalf("parseArgs() error: %v", err)
+		}
+		benchConfigSink = cfg
+	}
+}
+
+// BenchmarkJSONParse_ParseJSONStreamInternal measures line-delimited JSON stream parsing.
+func BenchmarkJSONParse_ParseJSONStreamInternal(b *testing.B) {
+	stream := []byte(
+		`{"type":"thread.started","thread_id":"t"}` + "\n" +
+			`{"type":"item.completed","item":{"type":"agent_message","text":"hello"}}` + "\n" +
+			`{"type":"thread.completed","thread_id":"t"}` + "\n",
+	)
+	b.SetBytes(int64(len(stream)))
+
+	b.ReportAllocs()
+	for i := 0; i < b.N; i++ {
+		message, threadID := parseJSONStreamInternal(bytes.NewReader(stream), nil, nil, nil, nil)
+		benchMessageSink = message
+		benchThreadIDSink = threadID
+	}
+}
+
+// BenchmarkLoggerWrite 测试日志写入性能
+func BenchmarkLoggerWrite(b *testing.B) {
+	logger, err := NewLogger()
+	if err != nil {
+		b.Fatal(err)
+	}
+	defer logger.Close()
+
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		logger.Info("benchmark log message")
+	}
+	b.StopTimer()
+	logger.Flush()
+}
+
+// BenchmarkLoggerConcurrentWrite 测试并发日志写入性能
+func BenchmarkLoggerConcurrentWrite(b *testing.B) {
+	logger, err := NewLogger()
+	if err != nil {
+		b.Fatal(err)
+	}
+	defer logger.Close()
+
+	b.ResetTimer()
+	b.RunParallel(func(pb *testing.PB) {
+		for pb.Next() {
+			logger.Info("concurrent benchmark log message")
+		}
+	})
+	b.StopTimer()
+	logger.Flush()
+}
--- a/codeagent-wrapper/internal/app/cli.go
+++ b/codeagent-wrapper/internal/app/cli.go
@@ -0,0 +1,664 @@
+package wrapper
+
+import (
+	"errors"
+	"fmt"
+	"io"
+	"os"
+	"reflect"
+	"strings"
+
+	config "codeagent-wrapper/internal/config"
+
+	"github.com/spf13/cobra"
+	"github.com/spf13/pflag"
+	"github.com/spf13/viper"
+)
+
+type exitError struct {
+	code int
+}
+
+func (e exitError) Error() string {
+	return fmt.Sprintf("exit %d", e.code)
+}
+
+type cliOptions struct {
+	Backend         string
+	Model           string
+	ReasoningEffort string
+	Agent           string
+	PromptFile      string
+	SkipPermissions bool
+
+	Parallel   bool
+	FullOutput bool
+
+	Cleanup    bool
+	Version    bool
+	ConfigFile string
+}
+
+func Main() {
+	Run()
+}
+
+// Run is the program entrypoint for cmd/codeagent/main.go.
+func Run() {
+	exitFn(run())
+}
+
+func run() int {
+	cmd := newRootCommand()
+	cmd.SetArgs(os.Args[1:])
+	if err := cmd.Execute(); err != nil {
+		var ee exitError
+		if errors.As(err, &ee) {
+			return ee.code
+		}
+		return 1
+	}
+	return 0
+}
+
+func newRootCommand() *cobra.Command {
+	name := currentWrapperName()
+	opts := &cliOptions{}
+
+	cmd := &cobra.Command{
+		Use:           fmt.Sprintf("%s [flags] <task>|resume <session_id> <task> [workdir]", name),
+		Short:         "Go wrapper for AI CLI backends",
+		SilenceErrors: true,
+		SilenceUsage:  true,
+		Args:          cobra.ArbitraryArgs,
+		RunE: func(cmd *cobra.Command, args []string) error {
+			if opts.Version {
+				fmt.Printf("%s version %s\n", name, version)
+				return nil
+			}
+			if opts.Cleanup {
+				code := runCleanupMode()
+				if code == 0 {
+					return nil
+				}
+				return exitError{code: code}
+			}
+
+			exitCode := runWithLoggerAndCleanup(func() int {
+				v, err := config.NewViper(opts.ConfigFile)
+				if err != nil {
+					logError(err.Error())
+					return 1
+				}
+
+				if opts.Parallel {
+					return runParallelMode(cmd, args, opts, v, name)
+				}
+
+				logInfo("Script started")
+
+				cfg, err := buildSingleConfig(cmd, args, os.Args[1:], opts, v)
+				if err != nil {
+					logError(err.Error())
+					return 1
+				}
+				logInfo(fmt.Sprintf("Parsed args: mode=%s, task_len=%d, backend=%s", cfg.Mode, len(cfg.Task), cfg.Backend))
+				return runSingleMode(cfg, name)
+			})
+
+			if exitCode == 0 {
+				return nil
+			}
+			return exitError{code: exitCode}
+		},
+	}
+	cmd.CompletionOptions.DisableDefaultCmd = true
+
+	addRootFlags(cmd.Flags(), opts)
+	cmd.AddCommand(newVersionCommand(name), newCleanupCommand())
+
+	return cmd
+}
+
+func addRootFlags(fs *pflag.FlagSet, opts *cliOptions) {
+	fs.StringVar(&opts.ConfigFile, "config", "", "Config file path (default: $HOME/.codeagent/config.*)")
+	fs.BoolVarP(&opts.Version, "version", "v", false, "Print version and exit")
+	fs.BoolVar(&opts.Cleanup, "cleanup", false, "Clean up old logs and exit")
+
+	fs.BoolVar(&opts.Parallel, "parallel", false, "Run tasks in parallel (config from stdin)")
+	fs.BoolVar(&opts.FullOutput, "full-output", false, "Parallel mode: include full task output (legacy)")
+
+	fs.StringVar(&opts.Backend, "backend", defaultBackendName, "Backend to use (codex, claude, gemini, opencode)")
+	fs.StringVar(&opts.Model, "model", "", "Model override")
+	fs.StringVar(&opts.ReasoningEffort, "reasoning-effort", "", "Reasoning effort (backend-specific)")
+	fs.StringVar(&opts.Agent, "agent", "", "Agent preset name (from ~/.codeagent/models.json)")
+	fs.StringVar(&opts.PromptFile, "prompt-file", "", "Prompt file path")
+
+	fs.BoolVar(&opts.SkipPermissions, "skip-permissions", false, "Skip permissions prompts (also via CODEAGENT_SKIP_PERMISSIONS)")
+	fs.BoolVar(&opts.SkipPermissions, "dangerously-skip-permissions", false, "Alias for --skip-permissions")
+}
+
+func newVersionCommand(name string) *cobra.Command {
+	return &cobra.Command{
+		Use:           "version",
+		Short:         "Print version and exit",
+		SilenceErrors: true,
+		SilenceUsage:  true,
+		RunE: func(cmd *cobra.Command, args []string) error {
+			fmt.Printf("%s version %s\n", name, version)
+			return nil
+		},
+	}
+}
+
+func newCleanupCommand() *cobra.Command {
+	return &cobra.Command{
+		Use:           "cleanup",
+		Short:         "Clean up old logs and exit",
+		SilenceErrors: true,
+		SilenceUsage:  true,
+		RunE: func(cmd *cobra.Command, args []string) error {
+			code := runCleanupMode()
+			if code == 0 {
+				return nil
+			}
+			return exitError{code: code}
+		},
+	}
+}
+
+func runWithLoggerAndCleanup(fn func() int) (exitCode int) {
+	logger, err := NewLogger()
+	if err != nil {
+		fmt.Fprintf(os.Stderr, "ERROR: failed to initialize logger: %v\n", err)
+		return 1
+	}
+	setLogger(logger)
+
+	defer func() {
+		logger := activeLogger()
+		if logger != nil {
+			logger.Flush()
+		}
+		if err := closeLogger(); err != nil {
+			fmt.Fprintf(os.Stderr, "ERROR: failed to close logger: %v\n", err)
+		}
+		if logger == nil {
+			return
+		}
+
+		if exitCode != 0 {
+			if entries := logger.ExtractRecentErrors(10); len(entries) > 0 {
+				fmt.Fprintln(os.Stderr, "\n=== Recent Errors ===")
+				for _, entry := range entries {
+					fmt.Fprintln(os.Stderr, entry)
+				}
+				fmt.Fprintf(os.Stderr, "Log file: %s (deleted)\n", logger.Path())
+			}
+		}
+		_ = logger.RemoveLogFile()
+	}()
+	defer runCleanupHook()
+
+	// Clean up stale logs from previous runs.
+	scheduleStartupCleanup()
+
+	return fn()
+}
+
+func parseArgs() (*Config, error) {
+	opts := &cliOptions{}
+	cmd := &cobra.Command{SilenceErrors: true, SilenceUsage: true, Args: cobra.ArbitraryArgs}
+	addRootFlags(cmd.Flags(), opts)
+
+	rawArgv := os.Args[1:]
+	if err := cmd.ParseFlags(rawArgv); err != nil {
+		return nil, err
+	}
+	args := cmd.Flags().Args()
+
+	v, err := config.NewViper(opts.ConfigFile)
+	if err != nil {
+		return nil, err
+	}
+
+	return buildSingleConfig(cmd, args, rawArgv, opts, v)
+}
+
+func buildSingleConfig(cmd *cobra.Command, args []string, rawArgv []string, opts *cliOptions, v *viper.Viper) (*Config, error) {
+	backendName := defaultBackendName
+	model := ""
+	reasoningEffort := ""
+	agentName := ""
+	promptFile := ""
+	promptFileExplicit := false
+	yolo := false
+
+	if cmd.Flags().Changed("agent") {
+		agentName = strings.TrimSpace(opts.Agent)
+		if agentName == "" {
+			return nil, fmt.Errorf("--agent flag requires a value")
+		}
+		if err := config.ValidateAgentName(agentName); err != nil {
+			return nil, fmt.Errorf("--agent flag invalid value: %w", err)
+		}
+	} else {
+		agentName = strings.TrimSpace(v.GetString("agent"))
+		if agentName != "" {
+			if err := config.ValidateAgentName(agentName); err != nil {
+				return nil, fmt.Errorf("--agent flag invalid value: %w", err)
+			}
+		}
+	}
+
+	var resolvedBackend, resolvedModel, resolvedPromptFile, resolvedReasoning string
+	if agentName != "" {
+		var resolvedYolo bool
+		resolvedBackend, resolvedModel, resolvedPromptFile, resolvedReasoning, _, _, resolvedYolo = config.ResolveAgentConfig(agentName)
+		yolo = resolvedYolo
+	}
+
+	if cmd.Flags().Changed("prompt-file") {
+		promptFile = strings.TrimSpace(opts.PromptFile)
+		if promptFile == "" {
+			return nil, fmt.Errorf("--prompt-file flag requires a value")
+		}
+		promptFileExplicit = true
+	} else if val := strings.TrimSpace(v.GetString("prompt-file")); val != "" {
+		promptFile = val
+		promptFileExplicit = true
+	} else {
+		promptFile = resolvedPromptFile
+	}
+
+	agentFlagChanged := cmd.Flags().Changed("agent")
+	backendFlagChanged := cmd.Flags().Changed("backend")
+	if backendFlagChanged {
+		backendName = strings.TrimSpace(opts.Backend)
+		if backendName == "" {
+			return nil, fmt.Errorf("--backend flag requires a value")
+		}
+	}
+
+	switch {
+	case agentFlagChanged && backendFlagChanged && lastFlagIndex(rawArgv, "agent") > lastFlagIndex(rawArgv, "backend"):
+		backendName = resolvedBackend
+	case !backendFlagChanged && agentName != "":
+		backendName = resolvedBackend
+	case !backendFlagChanged:
+		if val := strings.TrimSpace(v.GetString("backend")); val != "" {
+			backendName = val
+		}
+	}
+
+	modelFlagChanged := cmd.Flags().Changed("model")
+	if modelFlagChanged {
+		model = strings.TrimSpace(opts.Model)
+		if model == "" {
+			return nil, fmt.Errorf("--model flag requires a value")
+		}
+	}
+
+	switch {
+	case agentFlagChanged && modelFlagChanged && lastFlagIndex(rawArgv, "agent") > lastFlagIndex(rawArgv, "model"):
+		model = strings.TrimSpace(resolvedModel)
+	case !modelFlagChanged && agentName != "":
+		model = strings.TrimSpace(resolvedModel)
+	case !modelFlagChanged:
+		model = strings.TrimSpace(v.GetString("model"))
+	}
+
+	if cmd.Flags().Changed("reasoning-effort") {
+		reasoningEffort = strings.TrimSpace(opts.ReasoningEffort)
+		if reasoningEffort == "" {
+			return nil, fmt.Errorf("--reasoning-effort flag requires a value")
+		}
+	} else if val := strings.TrimSpace(v.GetString("reasoning-effort")); val != "" {
+		reasoningEffort = val
+	} else if agentName != "" {
+		reasoningEffort = strings.TrimSpace(resolvedReasoning)
+	}
+
+	skipChanged := cmd.Flags().Changed("skip-permissions") || cmd.Flags().Changed("dangerously-skip-permissions")
+	skipPermissions := false
+	if skipChanged {
+		skipPermissions = opts.SkipPermissions
+	} else {
+		skipPermissions = v.GetBool("skip-permissions")
+	}
+
+	if len(args) == 0 {
+		return nil, fmt.Errorf("task required")
+	}
+
+	cfg := &Config{
+		WorkDir:            defaultWorkdir,
+		Backend:            backendName,
+		Agent:              agentName,
+		PromptFile:         promptFile,
+		PromptFileExplicit: promptFileExplicit,
+		SkipPermissions:    skipPermissions,
+		Yolo:               yolo,
+		Model:              model,
+		ReasoningEffort:    reasoningEffort,
+		MaxParallelWorkers: config.ResolveMaxParallelWorkers(),
+	}
+
+	if args[0] == "resume" {
+		if len(args) < 3 {
+			return nil, fmt.Errorf("resume mode requires: resume <session_id> <task>")
+		}
+		cfg.Mode = "resume"
+		cfg.SessionID = strings.TrimSpace(args[1])
+		if cfg.SessionID == "" {
+			return nil, fmt.Errorf("resume mode requires non-empty session_id")
+		}
+		cfg.Task = args[2]
+		cfg.ExplicitStdin = (args[2] == "-")
+		if len(args) > 3 {
+			if args[3] == "-" {
+				return nil, fmt.Errorf("invalid workdir: '-' is not a valid directory path")
+			}
+			cfg.WorkDir = args[3]
+		}
+	} else {
+		cfg.Mode = "new"
+		cfg.Task = args[0]
+		cfg.ExplicitStdin = (args[0] == "-")
+		if len(args) > 1 {
+			if args[1] == "-" {
+				return nil, fmt.Errorf("invalid workdir: '-' is not a valid directory path")
+			}
+			cfg.WorkDir = args[1]
+		}
+	}
+
+	return cfg, nil
+}
+
+func lastFlagIndex(argv []string, name string) int {
+	if len(argv) == 0 {
+		return -1
+	}
+	name = strings.TrimSpace(name)
+	if name == "" {
+		return -1
+	}
+
+	needle := "--" + name
+	prefix := needle + "="
+	last := -1
+	for i, arg := range argv {
+		if arg == needle || strings.HasPrefix(arg, prefix) {
+			last = i
+		}
+	}
+	return last
+}
+
+func runParallelMode(cmd *cobra.Command, args []string, opts *cliOptions, v *viper.Viper, name string) int {
+	if len(args) > 0 {
+		fmt.Fprintln(os.Stderr, "ERROR: --parallel reads its task configuration from stdin; no positional arguments are allowed.")
+		fmt.Fprintln(os.Stderr, "Usage examples:")
+		fmt.Fprintf(os.Stderr, "  %s --parallel < tasks.txt\n", name)
+		fmt.Fprintf(os.Stderr, "  echo '...' | %s --parallel\n", name)
+		fmt.Fprintf(os.Stderr, "  %s --parallel <<'EOF'\n", name)
+		fmt.Fprintf(os.Stderr, "  %s --parallel --full-output <<'EOF'  # include full task output\n", name)
+		return 1
+	}
+
+	if cmd.Flags().Changed("agent") || cmd.Flags().Changed("prompt-file") || cmd.Flags().Changed("reasoning-effort") {
+		fmt.Fprintln(os.Stderr, "ERROR: --parallel reads its task configuration from stdin; only --backend, --model, --full-output and --skip-permissions are allowed.")
+		return 1
+	}
+
+	backendName := defaultBackendName
+	if cmd.Flags().Changed("backend") {
+		backendName = strings.TrimSpace(opts.Backend)
+		if backendName == "" {
+			fmt.Fprintln(os.Stderr, "ERROR: --backend flag requires a value")
+			return 1
+		}
+	} else if val := strings.TrimSpace(v.GetString("backend")); val != "" {
+		backendName = val
+	}
+
+	model := ""
+	if cmd.Flags().Changed("model") {
+		model = strings.TrimSpace(opts.Model)
+		if model == "" {
+			fmt.Fprintln(os.Stderr, "ERROR: --model flag requires a value")
+			return 1
+		}
+	} else {
+		model = strings.TrimSpace(v.GetString("model"))
+	}
+
+	fullOutput := opts.FullOutput
+	if !cmd.Flags().Changed("full-output") && v.IsSet("full-output") {
+		fullOutput = v.GetBool("full-output")
+	}
+
+	skipChanged := cmd.Flags().Changed("skip-permissions") || cmd.Flags().Changed("dangerously-skip-permissions")
+	skipPermissions := false
+	if skipChanged {
+		skipPermissions = opts.SkipPermissions
+	} else {
+		skipPermissions = v.GetBool("skip-permissions")
+	}
+
+	backend, err := selectBackendFn(backendName)
+	if err != nil {
+		fmt.Fprintf(os.Stderr, "ERROR: %v\n", err)
+		return 1
+	}
+	backendName = backend.Name()
+
+	data, err := io.ReadAll(stdinReader)
+	if err != nil {
+		fmt.Fprintf(os.Stderr, "ERROR: failed to read stdin: %v\n", err)
+		return 1
+	}
+
+	cfg, err := parseParallelConfig(data)
+	if err != nil {
+		fmt.Fprintf(os.Stderr, "ERROR: %v\n", err)
+		return 1
+	}
+
+	cfg.GlobalBackend = backendName
+	model = strings.TrimSpace(model)
+	for i := range cfg.Tasks {
+		if strings.TrimSpace(cfg.Tasks[i].Backend) == "" {
+			cfg.Tasks[i].Backend = backendName
+		}
+		if strings.TrimSpace(cfg.Tasks[i].Model) == "" && model != "" {
+			cfg.Tasks[i].Model = model
+		}
+		cfg.Tasks[i].SkipPermissions = cfg.Tasks[i].SkipPermissions || skipPermissions
+	}
+
+	timeoutSec := resolveTimeout()
+	layers, err := topologicalSort(cfg.Tasks)
+	if err != nil {
+		fmt.Fprintf(os.Stderr, "ERROR: %v\n", err)
+		return 1
+	}
+
+	results := executeConcurrent(layers, timeoutSec)
+
+	for i := range results {
+		results[i].CoverageTarget = defaultCoverageTarget
+		if results[i].Message == "" {
+			continue
+		}
+
+		lines := strings.Split(results[i].Message, "\n")
+		results[i].Coverage = extractCoverageFromLines(lines)
+		results[i].CoverageNum = extractCoverageNum(results[i].Coverage)
+		results[i].FilesChanged = extractFilesChangedFromLines(lines)
+		results[i].TestsPassed, results[i].TestsFailed = extractTestResultsFromLines(lines)
+		results[i].KeyOutput = extractKeyOutputFromLines(lines, 150)
+	}
+
+	fmt.Println(generateFinalOutputWithMode(results, !fullOutput))
+
+	exitCode := 0
+	for _, res := range results {
+		if res.ExitCode != 0 {
+			exitCode = res.ExitCode
+		}
+	}
+	return exitCode
+}
+
+func runSingleMode(cfg *Config, name string) int {
+	backend, err := selectBackendFn(cfg.Backend)
+	if err != nil {
+		logError(err.Error())
+		return 1
+	}
+	cfg.Backend = backend.Name()
+
+	cmdInjected := codexCommand != defaultCodexCommand
+	argsInjected := buildCodexArgsFn != nil && reflect.ValueOf(buildCodexArgsFn).Pointer() != reflect.ValueOf(defaultBuildArgsFn).Pointer()
+
+	if backend.Name() != defaultBackendName || !cmdInjected {
+		codexCommand = backend.Command()
+	}
+	if backend.Name() != defaultBackendName || !argsInjected {
+		buildCodexArgsFn = backend.BuildArgs
+	}
+	logInfo(fmt.Sprintf("Selected backend: %s", backend.Name()))
+
+	timeoutSec := resolveTimeout()
+	logInfo(fmt.Sprintf("Timeout: %ds", timeoutSec))
+	cfg.Timeout = timeoutSec
+
+	var taskText string
+	var piped bool
+
+	if cfg.ExplicitStdin {
+		logInfo("Explicit stdin mode: reading task from stdin")
+		data, err := io.ReadAll(stdinReader)
+		if err != nil {
+			logError("Failed to read stdin: " + err.Error())
+			return 1
+		}
+		taskText = string(data)
+		if taskText == "" {
+			logError("Explicit stdin mode requires task input from stdin")
+			return 1
+		}
+		piped = !isTerminal()
+	} else {
+		pipedTask, err := readPipedTask()
+		if err != nil {
+			logError("Failed to read piped stdin: " + err.Error())
+			return 1
+		}
+		piped = pipedTask != ""
+		if piped {
+			taskText = pipedTask
+		} else {
+			taskText = cfg.Task
+		}
+	}
+
+	if strings.TrimSpace(cfg.PromptFile) != "" {
+		prompt, err := readAgentPromptFile(cfg.PromptFile, cfg.PromptFileExplicit)
+		if err != nil {
+			logError("Failed to read prompt file: " + err.Error())
+			return 1
+		}
+		taskText = wrapTaskWithAgentPrompt(prompt, taskText)
+	}
+
+	useStdin := cfg.ExplicitStdin || shouldUseStdin(taskText, piped)
+
+	targetArg := taskText
+	if useStdin {
+		targetArg = "-"
+	}
+	codexArgs := buildCodexArgsFn(cfg, targetArg)
+
+	logger := activeLogger()
+	if logger == nil {
+		fmt.Fprintln(os.Stderr, "ERROR: logger is not initialized")
+		return 1
+	}
+
+	fmt.Fprintf(os.Stderr, "[%s]\n", name)
+	fmt.Fprintf(os.Stderr, "  Backend: %s\n", cfg.Backend)
+	fmt.Fprintf(os.Stderr, "  Command: %s %s\n", codexCommand, strings.Join(codexArgs, " "))
+	fmt.Fprintf(os.Stderr, "  PID: %d\n", os.Getpid())
+	fmt.Fprintf(os.Stderr, "  Log: %s\n", logger.Path())
+
+	if useStdin {
+		var reasons []string
+		if piped {
+			reasons = append(reasons, "piped input")
+		}
+		if cfg.ExplicitStdin {
+			reasons = append(reasons, "explicit \"-\"")
+		}
+		if strings.Contains(taskText, "\n") {
+			reasons = append(reasons, "newline")
+		}
+		if strings.Contains(taskText, "\\") {
+			reasons = append(reasons, "backslash")
+		}
+		if strings.Contains(taskText, "\"") {
+			reasons = append(reasons, "double-quote")
+		}
+		if strings.Contains(taskText, "'") {
+			reasons = append(reasons, "single-quote")
+		}
+		if strings.Contains(taskText, "`") {
+			reasons = append(reasons, "backtick")
+		}
+		if strings.Contains(taskText, "$") {
+			reasons = append(reasons, "dollar")
+		}
+		if len(taskText) > 800 {
+			reasons = append(reasons, "length>800")
+		}
+		if len(reasons) > 0 {
+			logWarn(fmt.Sprintf("Using stdin mode for task due to: %s", strings.Join(reasons, ", ")))
+		}
+	}
+
+	logInfo(fmt.Sprintf("%s running...", cfg.Backend))
+
+	taskSpec := TaskSpec{
+		Task:            taskText,
+		WorkDir:         cfg.WorkDir,
+		Mode:            cfg.Mode,
+		SessionID:       cfg.SessionID,
+		Backend:         cfg.Backend,
+		Model:           cfg.Model,
+		ReasoningEffort: cfg.ReasoningEffort,
+		Agent:           cfg.Agent,
+		SkipPermissions: cfg.SkipPermissions,
+		UseStdin:        useStdin,
+	}
+
+	result := runTaskFn(taskSpec, false, cfg.Timeout)
+
+	if result.ExitCode != 0 {
+		return result.ExitCode
+	}
+
+	// Validate that we got a meaningful output message
+	if strings.TrimSpace(result.Message) == "" {
+		logError(fmt.Sprintf("no output message: backend=%s returned empty result.Message with exit_code=0", cfg.Backend))
+		return 1
+	}
+
+	fmt.Println(result.Message)
+	if result.SessionID != "" {
+		fmt.Printf("\n---\nSESSION_ID: %s\n", result.SessionID)
+	}
+
+	return 0
+}
--- a/codeagent-wrapper/internal/app/concurrent_stress_test.go
+++ b/codeagent-wrapper/internal/app/concurrent_stress_test.go
@@ -1,4 +1,4 @@
-package main
+package wrapper

 import (
 	"bufio"
@@ -11,9 +11,20 @@ import (
 	"sync/atomic"
 	"testing"
 	"time"
+
+	"github.com/goccy/go-json"
 )

 func stripTimestampPrefix(line string) string {
+	line = strings.TrimSpace(line)
+	if strings.HasPrefix(line, "{") {
+		var evt struct {
+			Message string `json:"message"`
+		}
+		if err := json.Unmarshal([]byte(line), &evt); err == nil && evt.Message != "" {
+			return evt.Message
+		}
+	}
 	if !strings.HasPrefix(line, "[") {
 		return line
 	}
--- a/codeagent-wrapper/internal/app/config_alias.go
+++ b/codeagent-wrapper/internal/app/config_alias.go
@@ -0,0 +1,7 @@
+package wrapper
+
+import config "codeagent-wrapper/internal/config"
+
+// Keep the existing Config name throughout the codebase, but source the
+// implementation from internal/config.
+type Config = config.Config
--- a/codeagent-wrapper/internal/app/executor_alias.go
+++ b/codeagent-wrapper/internal/app/executor_alias.go
@@ -0,0 +1,54 @@
+package wrapper
+
+import (
+	"context"
+
+	backend "codeagent-wrapper/internal/backend"
+	config "codeagent-wrapper/internal/config"
+	executor "codeagent-wrapper/internal/executor"
+)
+
+// defaultRunCodexTaskFn is the default implementation of runCodexTaskFn (exposed for test reset).
+func defaultRunCodexTaskFn(task TaskSpec, timeout int) TaskResult {
+	return executor.DefaultRunCodexTaskFn(task, timeout)
+}
+
+var runCodexTaskFn = defaultRunCodexTaskFn
+
+func topologicalSort(tasks []TaskSpec) ([][]TaskSpec, error) {
+	return executor.TopologicalSort(tasks)
+}
+
+func executeConcurrent(layers [][]TaskSpec, timeout int) []TaskResult {
+	maxWorkers := config.ResolveMaxParallelWorkers()
+	return executeConcurrentWithContext(context.Background(), layers, timeout, maxWorkers)
+}
+
+func executeConcurrentWithContext(parentCtx context.Context, layers [][]TaskSpec, timeout int, maxWorkers int) []TaskResult {
+	return executor.ExecuteConcurrentWithContext(parentCtx, layers, timeout, maxWorkers, runCodexTaskFn)
+}
+
+func generateFinalOutput(results []TaskResult) string {
+	return executor.GenerateFinalOutput(results)
+}
+
+func generateFinalOutputWithMode(results []TaskResult, summaryOnly bool) string {
+	return executor.GenerateFinalOutputWithMode(results, summaryOnly)
+}
+
+func buildCodexArgs(cfg *Config, targetArg string) []string {
+	return backend.BuildCodexArgs(cfg, targetArg)
+}
+
+func runCodexTask(taskSpec TaskSpec, silent bool, timeoutSec int) TaskResult {
+	return runCodexTaskWithContext(context.Background(), taskSpec, nil, nil, false, silent, timeoutSec)
+}
+
+func runCodexProcess(parentCtx context.Context, codexArgs []string, taskText string, useStdin bool, timeoutSec int) (message, threadID string, exitCode int) {
+	res := runCodexTaskWithContext(parentCtx, TaskSpec{Task: taskText, WorkDir: defaultWorkdir, Mode: "new", UseStdin: useStdin}, nil, codexArgs, true, false, timeoutSec)
+	return res.Message, res.SessionID, res.ExitCode
+}
+
+func runCodexTaskWithContext(parentCtx context.Context, taskSpec TaskSpec, backend Backend, customArgs []string, useCustomArgs bool, silent bool, timeoutSec int) TaskResult {
+	return executor.RunCodexTaskWithContext(parentCtx, taskSpec, backend, codexCommand, buildCodexArgsFn, customArgs, useCustomArgs, silent, timeoutSec)
+}
--- a/codeagent-wrapper/internal/app/executor_concurrent_test.go
+++ b/codeagent-wrapper/internal/app/executor_concurrent_test.go
@@ -1,4 +1,4 @@
-package main
+package wrapper

 import (
 	"bufio"
@@ -15,9 +15,10 @@ import (
 	"strings"
 	"sync"
 	"sync/atomic"
-	"syscall"
 	"testing"
 	"time"
+
+	executor "codeagent-wrapper/internal/executor"
 )

 var executorTestTaskCounter atomic.Int64
@@ -91,7 +92,7 @@ func (rc *reasonReadCloser) record(reason string) {
 type execFakeRunner struct {
 	stdout          io.ReadCloser
 	stderr          io.ReadCloser
-	process         processHandle
+	process         executor.ProcessHandle
 	stdin           io.WriteCloser
 	dir             string
 	env             map[string]string
@@ -158,7 +159,7 @@ func (f *execFakeRunner) SetEnv(env map[string]string) {
 		f.env[k] = v
 	}
 }
-func (f *execFakeRunner) Process() processHandle {
+func (f *execFakeRunner) Process() executor.ProcessHandle {
 	if f.process != nil {
 		return f.process
 	}
@@ -168,225 +169,15 @@ func (f *execFakeRunner) Process() processHandle {
 	return &execFakeProcess{pid: 1}
 }

-func TestExecutorHelperCoverage(t *testing.T) {
-	t.Run("realCmdAndProcess", func(t *testing.T) {
-		rc := &realCmd{}
-		if err := rc.Start(); err == nil {
-			t.Fatalf("expected error for nil command")
-		}
-		if err := rc.Wait(); err == nil {
-			t.Fatalf("expected error for nil command")
-		}
-		if _, err := rc.StdoutPipe(); err == nil {
-			t.Fatalf("expected error for nil command")
-		}
-		if _, err := rc.StderrPipe(); err == nil {
-			t.Fatalf("expected error for nil command")
-		}
-		if _, err := rc.StdinPipe(); err == nil {
-			t.Fatalf("expected error for nil command")
-		}
-		rc.SetStderr(io.Discard)
-		if rc.Process() != nil {
-			t.Fatalf("expected nil process")
-		}
-		rcWithCmd := &realCmd{cmd: &exec.Cmd{}}
-		rcWithCmd.SetStderr(io.Discard)
-		rcWithCmd.SetDir("/tmp")
-		if rcWithCmd.cmd.Dir != "/tmp" {
-			t.Fatalf("expected SetDir to set cmd.Dir, got %q", rcWithCmd.cmd.Dir)
-		}
-		echoCmd := exec.Command("echo", "ok")
-		rcProc := &realCmd{cmd: echoCmd}
-		stdoutPipe, err := rcProc.StdoutPipe()
-		if err != nil {
-			t.Fatalf("StdoutPipe error: %v", err)
-		}
-		stderrPipe, err := rcProc.StderrPipe()
-		if err != nil {
-			t.Fatalf("StderrPipe error: %v", err)
-		}
-		stdinPipe, err := rcProc.StdinPipe()
-		if err != nil {
-			t.Fatalf("StdinPipe error: %v", err)
-		}
-		if err := rcProc.Start(); err != nil {
-			t.Fatalf("Start failed: %v", err)
-		}
-		_, _ = stdinPipe.Write([]byte{})
-		_ = stdinPipe.Close()
-		procHandle := rcProc.Process()
-		if procHandle == nil {
-			t.Fatalf("expected process handle")
-		}
-		_ = procHandle.Signal(syscall.SIGTERM)
-		_ = procHandle.Kill()
-		_ = rcProc.Wait()
-		_, _ = io.ReadAll(stdoutPipe)
-		_, _ = io.ReadAll(stderrPipe)
-
-		rp := &realProcess{}
-		if rp.Pid() != 0 {
-			t.Fatalf("nil process should have pid 0")
-		}
-		if rp.Kill() != nil {
-			t.Fatalf("nil process Kill should be nil")
-		}
-		if rp.Signal(syscall.SIGTERM) != nil {
-			t.Fatalf("nil process Signal should be nil")
-		}
-		rpLive := &realProcess{proc: &os.Process{Pid: 99}}
-		if rpLive.Pid() != 99 {
-			t.Fatalf("expected pid 99, got %d", rpLive.Pid())
-		}
-		_ = rpLive.Kill()
-		_ = rpLive.Signal(syscall.SIGTERM)
-	})
-
-	t.Run("topologicalSortAndSkip", func(t *testing.T) {
-		layers, err := topologicalSort([]TaskSpec{{ID: "root"}, {ID: "child", Dependencies: []string{"root"}}})
-		if err != nil || len(layers) != 2 {
-			t.Fatalf("unexpected topological sort result: layers=%d err=%v", len(layers), err)
-		}
-		if _, err := topologicalSort([]TaskSpec{{ID: "cycle", Dependencies: []string{"cycle"}}}); err == nil {
-			t.Fatalf("expected cycle detection error")
-		}
-
-		failed := map[string]TaskResult{"root": {ExitCode: 1}}
-		if skip, _ := shouldSkipTask(TaskSpec{ID: "child", Dependencies: []string{"root"}}, failed); !skip {
-			t.Fatalf("should skip when dependency failed")
-		}
-		if skip, _ := shouldSkipTask(TaskSpec{ID: "leaf"}, failed); skip {
-			t.Fatalf("should not skip task without dependencies")
-		}
-		if skip, _ := shouldSkipTask(TaskSpec{ID: "child-ok", Dependencies: []string{"root"}}, map[string]TaskResult{}); skip {
-			t.Fatalf("should not skip when dependencies succeeded")
-		}
-	})
-
-	t.Run("cancelledTaskResult", func(t *testing.T) {
-		ctx, cancel := context.WithCancel(context.Background())
-		cancel()
-		res := cancelledTaskResult("t1", ctx)
-		if res.ExitCode != 130 {
-			t.Fatalf("expected cancel exit code, got %d", res.ExitCode)
-		}
-
-		timeoutCtx, timeoutCancel := context.WithTimeout(context.Background(), 0)
-		defer timeoutCancel()
-		res = cancelledTaskResult("t2", timeoutCtx)
-		if res.ExitCode != 124 {
-			t.Fatalf("expected timeout exit code, got %d", res.ExitCode)
-		}
-	})
-
-	t.Run("generateFinalOutputAndArgs", func(t *testing.T) {
-		const key = "CODEX_BYPASS_SANDBOX"
-		t.Setenv(key, "false")
-
-		out := generateFinalOutput([]TaskResult{
-			{TaskID: "ok", ExitCode: 0},
-			{TaskID: "fail", ExitCode: 1, Error: "boom"},
-		})
-		if !strings.Contains(out, "ok") || !strings.Contains(out, "fail") {
-			t.Fatalf("unexpected summary output: %s", out)
-		}
-		// Test summary mode (default) - should have new format with ### headers
-		out = generateFinalOutput([]TaskResult{{TaskID: "rich", ExitCode: 0, SessionID: "sess", LogPath: "/tmp/log", Message: "hello"}})
-		if !strings.Contains(out, "### rich") {
-			t.Fatalf("summary output missing task header: %s", out)
-		}
-		// Test full output mode - should have Session and Message
-		out = generateFinalOutputWithMode([]TaskResult{{TaskID: "rich", ExitCode: 0, SessionID: "sess", LogPath: "/tmp/log", Message: "hello"}}, false)
-		if !strings.Contains(out, "Session: sess") || !strings.Contains(out, "Log: /tmp/log") || !strings.Contains(out, "hello") {
-			t.Fatalf("full output missing fields: %s", out)
-		}
-
-		args := buildCodexArgs(&Config{Mode: "new", WorkDir: "/tmp"}, "task")
-		if !slices.Equal(args, []string{"e", "--skip-git-repo-check", "-C", "/tmp", "--json", "task"}) {
-			t.Fatalf("unexpected codex args: %+v", args)
-		}
-		args = buildCodexArgs(&Config{Mode: "resume", SessionID: "sess"}, "target")
-		if !slices.Equal(args, []string{"e", "--skip-git-repo-check", "--json", "resume", "sess", "target"}) {
-			t.Fatalf("unexpected resume args: %+v", args)
-		}
-	})
-
-	t.Run("generateFinalOutputASCIIMode", func(t *testing.T) {
-		t.Setenv("CODEAGENT_ASCII_MODE", "true")
-
-		results := []TaskResult{
-			{TaskID: "ok", ExitCode: 0, Coverage: "92%", CoverageNum: 92, CoverageTarget: 90, KeyOutput: "done"},
-			{TaskID: "warn", ExitCode: 0, Coverage: "80%", CoverageNum: 80, CoverageTarget: 90, KeyOutput: "did"},
-			{TaskID: "bad", ExitCode: 2, Error: "boom"},
-		}
-		out := generateFinalOutput(results)
-
-		for _, sym := range []string{"PASS", "WARN", "FAIL"} {
-			if !strings.Contains(out, sym) {
-				t.Fatalf("ASCII mode should include %q, got: %s", sym, out)
-			}
-		}
-		for _, sym := range []string{"✓", "⚠️", "✗"} {
-			if strings.Contains(out, sym) {
-				t.Fatalf("ASCII mode should not include %q, got: %s", sym, out)
-			}
-		}
-	})
-
-	t.Run("generateFinalOutputUnicodeMode", func(t *testing.T) {
-		t.Setenv("CODEAGENT_ASCII_MODE", "false")
-
-		results := []TaskResult{
-			{TaskID: "ok", ExitCode: 0, Coverage: "92%", CoverageNum: 92, CoverageTarget: 90, KeyOutput: "done"},
-			{TaskID: "warn", ExitCode: 0, Coverage: "80%", CoverageNum: 80, CoverageTarget: 90, KeyOutput: "did"},
-			{TaskID: "bad", ExitCode: 2, Error: "boom"},
-		}
-		out := generateFinalOutput(results)
-
-		for _, sym := range []string{"✓", "⚠️", "✗"} {
-			if !strings.Contains(out, sym) {
-				t.Fatalf("Unicode mode should include %q, got: %s", sym, out)
-			}
-		}
-	})
-
-	t.Run("executeConcurrentWrapper", func(t *testing.T) {
-		orig := runCodexTaskFn
-		defer func() { runCodexTaskFn = orig }()
-		runCodexTaskFn = func(task TaskSpec, timeout int) TaskResult {
-			return TaskResult{TaskID: task.ID, ExitCode: 0, Message: "done"}
-		}
-		t.Setenv("CODEAGENT_MAX_PARALLEL_WORKERS", "1")
-
-		results := executeConcurrent([][]TaskSpec{{{ID: "wrap"}}}, 1)
-		if len(results) != 1 || results[0].TaskID != "wrap" {
-			t.Fatalf("unexpected wrapper results: %+v", results)
-		}
-
-		unbounded := executeConcurrentWithContext(context.Background(), [][]TaskSpec{{{ID: "unbounded"}}}, 1, 0)
-		if len(unbounded) != 1 || unbounded[0].ExitCode != 0 {
-			t.Fatalf("unexpected unbounded result: %+v", unbounded)
-		}
-
-		ctx, cancel := context.WithCancel(context.Background())
-		cancel()
-		cancelled := executeConcurrentWithContext(ctx, [][]TaskSpec{{{ID: "cancel"}}}, 1, 1)
-		if cancelled[0].ExitCode == 0 {
-			t.Fatalf("expected cancelled result, got %+v", cancelled[0])
-		}
-	})
-}
-
 func TestExecutorRunCodexTaskWithContext(t *testing.T) {
-	origRunner := newCommandRunner
-	defer func() { newCommandRunner = origRunner }()
+	defer resetTestHooks()

 	t.Run("resumeMissingSessionID", func(t *testing.T) {
-		newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
+		executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner {
 			t.Fatalf("unexpected command execution for invalid resume config")
 			return nil
-		}
+		})
+		t.Cleanup(func() { executor.SetNewCommandRunner(nil) })

 		res := runCodexTaskWithContext(context.Background(), TaskSpec{Task: "payload", WorkDir: ".", Mode: "resume"}, nil, nil, false, false, 1)
 		if res.ExitCode == 0 || !strings.Contains(res.Error, "session_id") {
@@ -396,13 +187,14 @@ func TestExecutorRunCodexTaskWithContext(t *testing.T) {

 	t.Run("success", func(t *testing.T) {
 		var firstStdout *reasonReadCloser
-		newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
+		executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner {
 			rc := newReasonReadCloser(`{"type":"item.completed","item":{"type":"agent_message","text":"hello"}}`)
 			if firstStdout == nil {
 				firstStdout = rc
 			}
 			return &execFakeRunner{stdout: rc, process: &execFakeProcess{pid: 1234}}
-		}
+		})
+		t.Cleanup(func() { executor.SetNewCommandRunner(nil) })

 		res := runCodexTaskWithContext(context.Background(), TaskSpec{ID: "task-1", Task: "payload", WorkDir: "."}, nil, nil, false, false, 1)
 		if res.Error != "" || res.Message != "hello" || res.ExitCode != 0 {
@@ -432,17 +224,18 @@ func TestExecutorRunCodexTaskWithContext(t *testing.T) {
 	})

 	t.Run("startErrors", func(t *testing.T) {
-		newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
+		executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner {
 			return &execFakeRunner{startErr: errors.New("executable file not found"), process: &execFakeProcess{pid: 1}}
-		}
+		})
+		t.Cleanup(func() { executor.SetNewCommandRunner(nil) })
 		res := runCodexTaskWithContext(context.Background(), TaskSpec{Task: "payload", WorkDir: "."}, nil, nil, false, false, 1)
 		if res.ExitCode != 127 {
 			t.Fatalf("expected missing executable exit code, got %d", res.ExitCode)
 		}

-		newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
+		executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner {
 			return &execFakeRunner{startErr: errors.New("start failed"), process: &execFakeProcess{pid: 2}}
-		}
+		})
 		res = runCodexTaskWithContext(context.Background(), TaskSpec{Task: "payload", WorkDir: "."}, nil, nil, false, false, 1)
 		if res.ExitCode == 0 {
 			t.Fatalf("expected non-zero exit on start failure")
@@ -450,13 +243,14 @@ func TestExecutorRunCodexTaskWithContext(t *testing.T) {
 	})

 	t.Run("timeoutAndPipes", func(t *testing.T) {
-		newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
+		executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner {
 			return &execFakeRunner{
 				stdout:    newReasonReadCloser(`{"type":"item.completed","item":{"type":"agent_message","text":"slow"}}`),
 				process:   &execFakeProcess{pid: 5},
 				waitDelay: 20 * time.Millisecond,
 			}
-		}
+		})
+		t.Cleanup(func() { executor.SetNewCommandRunner(nil) })
 		res := runCodexTaskWithContext(context.Background(), TaskSpec{Task: "payload", WorkDir: ".", UseStdin: true}, nil, nil, false, false, 0)
 		if res.ExitCode == 0 {
 			t.Fatalf("expected timeout result, got %+v", res)
@@ -464,17 +258,18 @@ func TestExecutorRunCodexTaskWithContext(t *testing.T) {
 	})

 	t.Run("pipeErrors", func(t *testing.T) {
-		newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
+		executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner {
 			return &execFakeRunner{stdoutErr: errors.New("stdout fail"), process: &execFakeProcess{pid: 6}}
-		}
+		})
+		t.Cleanup(func() { executor.SetNewCommandRunner(nil) })
 		res := runCodexTaskWithContext(context.Background(), TaskSpec{Task: "payload", WorkDir: "."}, nil, nil, false, false, 1)
 		if res.ExitCode == 0 {
 			t.Fatalf("expected failure on stdout pipe error")
 		}

-		newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
+		executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner {
 			return &execFakeRunner{stdinErr: errors.New("stdin fail"), process: &execFakeProcess{pid: 7}}
-		}
+		})
 		res = runCodexTaskWithContext(context.Background(), TaskSpec{Task: "payload", WorkDir: ".", UseStdin: true}, nil, nil, false, false, 1)
 		if res.ExitCode == 0 {
 			t.Fatalf("expected failure on stdin pipe error")
@@ -487,13 +282,14 @@ func TestExecutorRunCodexTaskWithContext(t *testing.T) {
 		if exitErr == nil {
 			t.Fatalf("expected exec.ExitError")
 		}
-		newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
+		executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner {
 			return &execFakeRunner{
 				stdout:  newReasonReadCloser(`{"type":"item.completed","item":{"type":"agent_message","text":"ignored"}}`),
 				process: &execFakeProcess{pid: 8},
 				waitErr: exitErr,
 			}
-		}
+		})
+		t.Cleanup(func() { executor.SetNewCommandRunner(nil) })
 		res := runCodexTaskWithContext(context.Background(), TaskSpec{Task: "payload", WorkDir: "."}, nil, nil, false, false, 1)
 		if res.ExitCode == 0 {
 			t.Fatalf("expected non-zero exit on wait error")
@@ -501,13 +297,14 @@ func TestExecutorRunCodexTaskWithContext(t *testing.T) {
 	})

 	t.Run("contextCancelled", func(t *testing.T) {
-		newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
+		executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner {
 			return &execFakeRunner{
 				stdout:    newReasonReadCloser(`{"type":"item.completed","item":{"type":"agent_message","text":"cancel"}}`),
 				process:   &execFakeProcess{pid: 9},
 				waitDelay: 10 * time.Millisecond,
 			}
-		}
+		})
+		t.Cleanup(func() { executor.SetNewCommandRunner(nil) })
 		ctx, cancel := context.WithCancel(context.Background())
 		cancel()
 		res := runCodexTaskWithContext(ctx, TaskSpec{Task: "payload", WorkDir: "."}, nil, nil, false, false, 1)
@@ -517,12 +314,13 @@ func TestExecutorRunCodexTaskWithContext(t *testing.T) {
 	})

 	t.Run("silentLogger", func(t *testing.T) {
-		newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
+		executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner {
 			return &execFakeRunner{
 				stdout:  newReasonReadCloser(`{"type":"item.completed","item":{"type":"agent_message","text":"quiet"}}`),
 				process: &execFakeProcess{pid: 10},
 			}
-		}
+		})
+		t.Cleanup(func() { executor.SetNewCommandRunner(nil) })
 		_ = closeLogger()
 		res := runCodexTaskWithContext(context.Background(), TaskSpec{Task: "payload", WorkDir: "."}, nil, nil, false, true, 1)
 		if res.ExitCode != 0 || res.LogPath == "" {
@@ -532,12 +330,13 @@ func TestExecutorRunCodexTaskWithContext(t *testing.T) {
 	})

 	t.Run("injectedLogger", func(t *testing.T) {
-		newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
+		executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner {
 			return &execFakeRunner{
 				stdout:  newReasonReadCloser(`{"type":"item.completed","item":{"type":"agent_message","text":"injected"}}`),
 				process: &execFakeProcess{pid: 12},
 			}
-		}
+		})
+		t.Cleanup(func() { executor.SetNewCommandRunner(nil) })
 		_ = closeLogger()

 		injected, err := NewLoggerWithSuffix("executor-injected")
@@ -549,7 +348,7 @@ func TestExecutorRunCodexTaskWithContext(t *testing.T) {
 			_ = os.Remove(injected.Path())
 		}()

-		ctx := withTaskLogger(context.Background(), injected)
+		ctx := executor.WithTaskLogger(context.Background(), injected)
 		res := runCodexTaskWithContext(ctx, TaskSpec{ID: "task-injected", Task: "payload", WorkDir: "."}, nil, nil, false, true, 1)
 		if res.ExitCode != 0 || res.LogPath != injected.Path() {
 			t.Fatalf("expected injected logger path, got %+v", res)
@@ -569,12 +368,13 @@ func TestExecutorRunCodexTaskWithContext(t *testing.T) {
 	})

 	t.Run("contextLoggerWithoutParent", func(t *testing.T) {
-		newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
+		executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner {
 			return &execFakeRunner{
 				stdout:  newReasonReadCloser(`{"type":"item.completed","item":{"type":"agent_message","text":"ctx"}}`),
 				process: &execFakeProcess{pid: 14},
 			}
-		}
+		})
+		t.Cleanup(func() { executor.SetNewCommandRunner(nil) })
 		_ = closeLogger()

 		taskLogger, err := NewLoggerWithSuffix("executor-taskctx")
@@ -586,8 +386,8 @@ func TestExecutorRunCodexTaskWithContext(t *testing.T) {
 			_ = os.Remove(taskLogger.Path())
 		})

-		ctx := withTaskLogger(context.Background(), taskLogger)
-		res := runCodexTaskWithContext(nil, TaskSpec{ID: "task-context", Task: "payload", WorkDir: ".", Context: ctx}, nil, nil, false, true, 1)
+		ctx := executor.WithTaskLogger(context.Background(), taskLogger)
+		res := runCodexTaskWithContext(context.TODO(), TaskSpec{ID: "task-context", Task: "payload", WorkDir: ".", Context: ctx}, nil, nil, false, true, 1)
 		if res.ExitCode != 0 || res.LogPath != taskLogger.Path() {
 			t.Fatalf("expected task logger to be reused from spec context, got %+v", res)
 		}
@@ -607,16 +407,17 @@ func TestExecutorRunCodexTaskWithContext(t *testing.T) {

 	t.Run("backendSetsDirAndNilContext", func(t *testing.T) {
 		var rc *execFakeRunner
-		newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
+		executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner {
 			rc = &execFakeRunner{
 				stdout:  newReasonReadCloser(`{"type":"item.completed","item":{"type":"agent_message","text":"backend"}}`),
 				process: &execFakeProcess{pid: 13},
 			}
 			return rc
-		}
+		})
+		t.Cleanup(func() { executor.SetNewCommandRunner(nil) })

 		_ = closeLogger()
-		res := runCodexTaskWithContext(nil, TaskSpec{ID: "task-backend", Task: "payload", WorkDir: "/tmp"}, ClaudeBackend{}, nil, false, false, 1)
+		res := runCodexTaskWithContext(context.TODO(), TaskSpec{ID: "task-backend", Task: "payload", WorkDir: "/tmp"}, ClaudeBackend{}, nil, false, false, 1)
 		if res.ExitCode != 0 || res.Message != "backend" {
 			t.Fatalf("unexpected result: %+v", res)
 		}
@@ -628,13 +429,14 @@ func TestExecutorRunCodexTaskWithContext(t *testing.T) {
 	t.Run("claudeSkipPermissionsPropagatesFromTaskSpec", func(t *testing.T) {
 		t.Setenv("CODEAGENT_SKIP_PERMISSIONS", "false")
 		var gotArgs []string
-		newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
+		executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner {
 			gotArgs = append([]string(nil), args...)
 			return &execFakeRunner{
 				stdout:  newReasonReadCloser(`{"type":"item.completed","item":{"type":"agent_message","text":"ok"}}`),
 				process: &execFakeProcess{pid: 15},
 			}
-		}
+		})
+		t.Cleanup(func() { executor.SetNewCommandRunner(nil) })

 		_ = closeLogger()
 		res := runCodexTaskWithContext(context.Background(), TaskSpec{ID: "task-skip", Task: "payload", WorkDir: ".", SkipPermissions: true}, ClaudeBackend{}, nil, false, false, 1)
@@ -647,12 +449,13 @@ func TestExecutorRunCodexTaskWithContext(t *testing.T) {
 	})

 	t.Run("missingMessage", func(t *testing.T) {
-		newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
+		executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner {
 			return &execFakeRunner{
 				stdout:  newReasonReadCloser(`{"type":"item.completed","item":{"type":"task","text":"noop"}}`),
 				process: &execFakeProcess{pid: 11},
 			}
-		}
+		})
+		t.Cleanup(func() { executor.SetNewCommandRunner(nil) })
 		res := runCodexTaskWithContext(context.Background(), TaskSpec{Task: "payload", WorkDir: "."}, nil, nil, false, false, 1)
 		if res.ExitCode == 0 {
 			t.Fatalf("expected failure when no agent_message returned")
@@ -678,7 +481,7 @@ func TestExecutorParallelLogIsolation(t *testing.T) {

 	origRun := runCodexTaskFn
 	runCodexTaskFn = func(task TaskSpec, timeout int) TaskResult {
-		logger := taskLoggerFromContext(task.Context)
+		logger := executor.TaskLoggerFromContext(task.Context)
 		if logger == nil {
 			return TaskResult{TaskID: task.ID, ExitCode: 1, Error: "missing task logger"}
 		}
@@ -702,7 +505,7 @@ func TestExecutorParallelLogIsolation(t *testing.T) {
 	os.Stderr = stderrW
 	defer func() { os.Stderr = oldStderr }()

-	results := executeConcurrentWithContext(nil, [][]TaskSpec{{{ID: taskA}, {ID: taskB}}}, 1, -1)
+	results := executeConcurrentWithContext(context.TODO(), [][]TaskSpec{{{ID: taskA}, {ID: taskB}}}, 1, -1)

 	_ = stderrW.Close()
 	os.Stderr = oldStderr
@@ -768,7 +571,7 @@ func TestConcurrentExecutorParallelLogIsolationAndClosure(t *testing.T) {
 	t.Setenv("TMPDIR", tempDir)

 	oldArgs := os.Args
-	os.Args = []string{defaultWrapperName}
+	os.Args = []string{wrapperName}
 	t.Cleanup(func() { os.Args = oldArgs })

 	mainLogger, err := NewLoggerWithSuffix("concurrent-main")
@@ -814,7 +617,7 @@ func TestConcurrentExecutorParallelLogIsolationAndClosure(t *testing.T) {
 	runCodexTaskFn = func(task TaskSpec, timeout int) TaskResult {
 		readyCh <- struct{}{}

-		logger := taskLoggerFromContext(task.Context)
+		logger := executor.TaskLoggerFromContext(task.Context)
 		loggerCh <- taskLoggerInfo{taskID: task.ID, logger: logger}
 		if logger == nil {
 			return TaskResult{TaskID: task.ID, ExitCode: 1, Error: "missing task logger"}
@@ -901,15 +704,9 @@ func TestConcurrentExecutorParallelLogIsolationAndClosure(t *testing.T) {
 	}

 	for taskID, logger := range loggers {
-		if !logger.closed.Load() {
+		if !logger.IsClosed() {
 			t.Fatalf("expected task logger to be closed for %q", taskID)
 		}
-		if logger.file == nil {
-			t.Fatalf("expected task logger file to be non-nil for %q", taskID)
-		}
-		if _, err := logger.file.Write([]byte("x")); err == nil {
-			t.Fatalf("expected task logger file to be closed for %q", taskID)
-		}
 	}

 	mainLogger.Flush()
@@ -979,10 +776,10 @@ func parseTaskIDFromLogLine(line string) (string, bool) {
 }

 func TestExecutorTaskLoggerContext(t *testing.T) {
-	if taskLoggerFromContext(nil) != nil {
-		t.Fatalf("expected nil logger from nil context")
+	if executor.TaskLoggerFromContext(context.TODO()) != nil {
+		t.Fatalf("expected nil logger from TODO context")
 	}
-	if taskLoggerFromContext(context.Background()) != nil {
+	if executor.TaskLoggerFromContext(context.Background()) != nil {
 		t.Fatalf("expected nil logger when context has no logger")
 	}

@@ -995,12 +792,12 @@ func TestExecutorTaskLoggerContext(t *testing.T) {
 		_ = os.Remove(logger.Path())
 	}()

-	ctx := withTaskLogger(context.Background(), logger)
-	if got := taskLoggerFromContext(ctx); got != logger {
+	ctx := executor.WithTaskLogger(context.Background(), logger)
+	if got := executor.TaskLoggerFromContext(ctx); got != logger {
 		t.Fatalf("expected logger roundtrip, got %v", got)
 	}

-	if taskLoggerFromContext(withTaskLogger(context.Background(), nil)) != nil {
+	if executor.TaskLoggerFromContext(executor.WithTaskLogger(context.Background(), nil)) != nil {
 		t.Fatalf("expected nil logger when injected logger is nil")
 	}
 }
@@ -1157,7 +954,7 @@ func TestExecutorExecuteConcurrentWithContextBranches(t *testing.T) {

 		orig := runCodexTaskFn
 		runCodexTaskFn = func(task TaskSpec, timeout int) TaskResult {
-			logger := taskLoggerFromContext(task.Context)
+			logger := executor.TaskLoggerFromContext(task.Context)
 			if logger != mainLogger {
 				return TaskResult{TaskID: task.ID, ExitCode: 1, Error: "unexpected logger"}
 			}
@@ -1191,9 +988,6 @@ func TestExecutorExecuteConcurrentWithContextBranches(t *testing.T) {
 			if res.LogPath != mainLogger.Path() {
 				t.Fatalf("shared log path mismatch: got %q want %q", res.LogPath, mainLogger.Path())
 			}
-			if !res.sharedLog {
-				t.Fatalf("expected sharedLog flag for %+v", res)
-			}
 			if !strings.Contains(stderrOut, "Log (shared)") {
 				t.Fatalf("stderr missing shared marker: %s", stderrOut)
 			}
@@ -1222,7 +1016,7 @@ func TestExecutorExecuteConcurrentWithContextBranches(t *testing.T) {

 		orig := runCodexTaskFn
 		runCodexTaskFn = func(task TaskSpec, timeout int) TaskResult {
-			logger := taskLoggerFromContext(task.Context)
+			logger := executor.TaskLoggerFromContext(task.Context)
 			if logger == nil {
 				return TaskResult{TaskID: task.ID, ExitCode: 1, Error: "missing logger"}
 			}
@@ -1260,7 +1054,14 @@ func TestExecutorExecuteConcurrentWithContextBranches(t *testing.T) {
 			if err != nil {
 				t.Fatalf("failed to read log %q: %v", res.LogPath, err)
 			}
-			if !strings.Contains(string(data), "TASK="+res.TaskID) {
+			found := false
+			for _, line := range strings.Split(string(data), "\n") {
+				if strings.Contains(stripTimestampPrefix(line), "TASK="+res.TaskID) {
+					found = true
+					break
+				}
+			}
+			if !found {
 				t.Fatalf("log for %q missing task marker, content: %s", res.TaskID, string(data))
 			}
 			_ = os.Remove(res.LogPath)
@@ -1268,147 +1069,6 @@ func TestExecutorExecuteConcurrentWithContextBranches(t *testing.T) {
 	})
 }

-func TestExecutorSignalAndTermination(t *testing.T) {
-	forceKillDelay.Store(0)
-	defer forceKillDelay.Store(5)
-
-	proc := &execFakeProcess{pid: 42}
-	cmd := &execFakeRunner{process: proc}
-
-	origNotify := signalNotifyFn
-	origStop := signalStopFn
-	defer func() {
-		signalNotifyFn = origNotify
-		signalStopFn = origStop
-	}()
-
-	signalNotifyFn = func(c chan<- os.Signal, sigs ...os.Signal) {
-		go func() { c <- syscall.SIGINT }()
-	}
-	signalStopFn = func(c chan<- os.Signal) {}
-
-	forwardSignals(context.Background(), cmd, func(string) {})
-	time.Sleep(20 * time.Millisecond)
-
-	proc.mu.Lock()
-	signalled := len(proc.signals)
-	proc.mu.Unlock()
-	if runtime.GOOS != "windows" && signalled == 0 {
-		t.Fatalf("process did not receive signal")
-	}
-	if proc.killed.Load() == 0 {
-		t.Fatalf("process was not killed after signal")
-	}
-
-	timer := terminateProcess(cmd)
-	if timer == nil {
-		t.Fatalf("terminateProcess returned nil timer")
-	}
-	timer.Stop()
-
-	ft := terminateCommand(cmd)
-	if ft == nil {
-		t.Fatalf("terminateCommand returned nil")
-	}
-	ft.Stop()
-
-	cmdKill := &execFakeRunner{process: &execFakeProcess{pid: 50}}
-	ftKill := terminateCommand(cmdKill)
-	time.Sleep(10 * time.Millisecond)
-	if p, ok := cmdKill.process.(*execFakeProcess); ok && p.killed.Load() == 0 {
-		t.Fatalf("terminateCommand did not kill process")
-	}
-	ftKill.Stop()
-
-	cmdKill2 := &execFakeRunner{process: &execFakeProcess{pid: 51}}
-	timer2 := terminateProcess(cmdKill2)
-	time.Sleep(10 * time.Millisecond)
-	if p, ok := cmdKill2.process.(*execFakeProcess); ok && p.killed.Load() == 0 {
-		t.Fatalf("terminateProcess did not kill process")
-	}
-	timer2.Stop()
-
-	if terminateCommand(nil) != nil {
-		t.Fatalf("terminateCommand should return nil for nil cmd")
-	}
-	if terminateCommand(&execFakeRunner{allowNilProcess: true}) != nil {
-		t.Fatalf("terminateCommand should return nil when process is nil")
-	}
-	if terminateProcess(nil) != nil {
-		t.Fatalf("terminateProcess should return nil for nil cmd")
-	}
-	if terminateProcess(&execFakeRunner{allowNilProcess: true}) != nil {
-		t.Fatalf("terminateProcess should return nil when process is nil")
-	}
-
-	signalNotifyFn = func(c chan<- os.Signal, sigs ...os.Signal) {}
-	ctxDone, cancelDone := context.WithCancel(context.Background())
-	cancelDone()
-	forwardSignals(ctxDone, &execFakeRunner{process: &execFakeProcess{pid: 70}}, func(string) {})
-}
-
-func TestExecutorCancelReasonAndCloseWithReason(t *testing.T) {
-	if reason := cancelReason("", nil); !strings.Contains(reason, "Context") {
-		t.Fatalf("unexpected cancelReason for nil ctx: %s", reason)
-	}
-	ctx, cancel := context.WithTimeout(context.Background(), 0)
-	defer cancel()
-	if !strings.Contains(cancelReason("cmd", ctx), "timeout") {
-		t.Fatalf("expected timeout reason")
-	}
-	cancelCtx, cancelFn := context.WithCancel(context.Background())
-	cancelFn()
-	if !strings.Contains(cancelReason("cmd", cancelCtx), "Execution cancelled") {
-		t.Fatalf("expected cancellation reason")
-	}
-	if !strings.Contains(cancelReason("", cancelCtx), "codex") {
-		t.Fatalf("expected default command name in cancel reason")
-	}
-
-	rc := &reasonReadCloser{r: strings.NewReader("data"), closedC: make(chan struct{}, 1)}
-	closeWithReason(rc, "why")
-	select {
-	case <-rc.closedC:
-	default:
-		t.Fatalf("CloseWithReason was not called")
-	}
-
-	plain := io.NopCloser(strings.NewReader("x"))
-	closeWithReason(plain, "noop")
-	closeWithReason(nil, "noop")
-}
-
-func TestExecutorForceKillTimerStop(t *testing.T) {
-	done := make(chan struct{}, 1)
-	ft := &forceKillTimer{timer: time.AfterFunc(50*time.Millisecond, func() { done <- struct{}{} }), done: done}
-	ft.Stop()
-
-	done2 := make(chan struct{}, 1)
-	ft2 := &forceKillTimer{timer: time.AfterFunc(0, func() { done2 <- struct{}{} }), done: done2}
-	time.Sleep(10 * time.Millisecond)
-	ft2.Stop()
-
-	var nilTimer *forceKillTimer
-	nilTimer.Stop()
-	(&forceKillTimer{}).Stop()
-}
-
-func TestExecutorForwardSignalsDefaults(t *testing.T) {
-	origNotify := signalNotifyFn
-	origStop := signalStopFn
-	signalNotifyFn = nil
-	signalStopFn = nil
-	defer func() {
-		signalNotifyFn = origNotify
-		signalStopFn = origStop
-	}()
-
-	ctx, cancel := context.WithCancel(context.Background())
-	cancel()
-	forwardSignals(ctx, &execFakeRunner{process: &execFakeProcess{pid: 80}}, func(string) {})
-	time.Sleep(10 * time.Millisecond)
-}
-
 func TestExecutorSharedLogFalseWhenCustomLogPath(t *testing.T) {
 	devNull, err := os.OpenFile(os.DevNull, os.O_WRONLY, 0)
 	if err != nil {
@@ -1464,10 +1124,9 @@ func TestExecutorSharedLogFalseWhenCustomLogPath(t *testing.T) {
 	}

 	res := results[0]
-	// 关键断言：即使 handle.shared=true（因为 task logger 创建失败），
-	// 但因为 LogPath 不等于主 logger 的路径，sharedLog 应为 false
-	if res.sharedLog {
-		t.Fatalf("expected sharedLog=false when LogPath differs from shared logger, got true")
+	out := generateFinalOutputWithMode(results, false)
+	if strings.Contains(out, "(shared)") {
+		t.Fatalf("did not expect shared marker when LogPath differs from shared logger, got: %s", out)
 	}

 	// 验证 LogPath 确实是自定义的
--- a/codeagent-wrapper/internal/app/logger.go
+++ b/codeagent-wrapper/internal/app/logger.go
@@ -0,0 +1,26 @@
+package wrapper
+
+import ilogger "codeagent-wrapper/internal/logger"
+
+type Logger = ilogger.Logger
+type CleanupStats = ilogger.CleanupStats
+
+func NewLogger() (*Logger, error) { return ilogger.NewLogger() }
+
+func NewLoggerWithSuffix(suffix string) (*Logger, error) { return ilogger.NewLoggerWithSuffix(suffix) }
+
+func setLogger(l *Logger) { ilogger.SetLogger(l) }
+
+func closeLogger() error { return ilogger.CloseLogger() }
+
+func activeLogger() *Logger { return ilogger.ActiveLogger() }
+
+func logInfo(msg string) { ilogger.LogInfo(msg) }
+
+func logWarn(msg string) { ilogger.LogWarn(msg) }
+
+func logError(msg string) { ilogger.LogError(msg) }
+
+func cleanupOldLogs() (CleanupStats, error) { return ilogger.CleanupOldLogs() }
+
+func sanitizeLogSuffix(raw string) string { return ilogger.SanitizeLogSuffix(raw) }
--- a/codeagent-wrapper/internal/app/main_integration_test.go
+++ b/codeagent-wrapper/internal/app/main_integration_test.go
@@ -1,7 +1,8 @@
-package main
+package wrapper

 import (
 	"bytes"
+	"codeagent-wrapper/internal/logger"
 	"fmt"
 	"io"
 	"os"
@@ -36,7 +37,9 @@ func captureStdout(t *testing.T, fn func()) string {
 	os.Stdout = old

 	var buf bytes.Buffer
-	io.Copy(&buf, r)
+	if _, err := io.Copy(&buf, r); err != nil {
+		t.Fatalf("io.Copy() error = %v", err)
+	}
 	return buf.String()
 }

@@ -57,11 +60,17 @@ func parseIntegrationOutput(t *testing.T, out string) integrationOutput {
 			for _, p := range parts {
 				p = strings.TrimSpace(p)
 				if strings.HasSuffix(p, "tasks") {
-					fmt.Sscanf(p, "%d tasks", &payload.Summary.Total)
+					if _, err := fmt.Sscanf(p, "%d tasks", &payload.Summary.Total); err != nil {
+						t.Fatalf("failed to parse total tasks from %q: %v", p, err)
+					}
 				} else if strings.HasSuffix(p, "passed") {
-					fmt.Sscanf(p, "%d passed", &payload.Summary.Success)
+					if _, err := fmt.Sscanf(p, "%d passed", &payload.Summary.Success); err != nil {
+						t.Fatalf("failed to parse passed tasks from %q: %v", p, err)
+					}
 				} else if strings.HasSuffix(p, "failed") {
-					fmt.Sscanf(p, "%d failed", &payload.Summary.Failed)
+					if _, err := fmt.Sscanf(p, "%d failed", &payload.Summary.Failed); err != nil {
+						t.Fatalf("failed to parse failed tasks from %q: %v", p, err)
+					}
 				}
 			}
 		} else if strings.HasPrefix(line, "Total:") {
@@ -70,11 +79,17 @@ func parseIntegrationOutput(t *testing.T, out string) integrationOutput {
 			for _, p := range parts {
 				p = strings.TrimSpace(p)
 				if strings.HasPrefix(p, "Total:") {
-					fmt.Sscanf(p, "Total: %d", &payload.Summary.Total)
+					if _, err := fmt.Sscanf(p, "Total: %d", &payload.Summary.Total); err != nil {
+						t.Fatalf("failed to parse total tasks from %q: %v", p, err)
+					}
 				} else if strings.HasPrefix(p, "Success:") {
-					fmt.Sscanf(p, "Success: %d", &payload.Summary.Success)
+					if _, err := fmt.Sscanf(p, "Success: %d", &payload.Summary.Success); err != nil {
+						t.Fatalf("failed to parse passed tasks from %q: %v", p, err)
+					}
 				} else if strings.HasPrefix(p, "Failed:") {
-					fmt.Sscanf(p, "Failed: %d", &payload.Summary.Failed)
+					if _, err := fmt.Sscanf(p, "Failed: %d", &payload.Summary.Failed); err != nil {
+						t.Fatalf("failed to parse failed tasks from %q: %v", p, err)
+					}
 				}
 			}
 		} else if line == "## Task Results" {
@@ -94,34 +109,39 @@ func parseIntegrationOutput(t *testing.T, out string) integrationOutput {
 			currentTask = &TaskResult{}

 			taskLine := strings.TrimPrefix(line, "### ")
-			success, warning, failed := getStatusSymbols()
-			// Parse different formats
-			if strings.Contains(taskLine, " "+success) {
-				parts := strings.Split(taskLine, " "+success)
+			parseMarker := func(marker string, exitCode int) bool {
+				needle := " " + marker
+				if !strings.Contains(taskLine, needle) {
+					return false
+				}
+				parts := strings.Split(taskLine, needle)
 				currentTask.TaskID = strings.TrimSpace(parts[0])
-				currentTask.ExitCode = 0
-				// Extract coverage if present
-				if len(parts) > 1 {
+				currentTask.ExitCode = exitCode
+				if exitCode == 0 && len(parts) > 1 {
 					coveragePart := strings.TrimSpace(parts[1])
 					if strings.HasSuffix(coveragePart, "%") {
 						currentTask.Coverage = coveragePart
 					}
 				}
-			} else if strings.Contains(taskLine, " "+warning) {
-				parts := strings.Split(taskLine, " "+warning)
-				currentTask.TaskID = strings.TrimSpace(parts[0])
-				currentTask.ExitCode = 0
-			} else if strings.Contains(taskLine, " "+failed) {
-				parts := strings.Split(taskLine, " "+failed)
-				currentTask.TaskID = strings.TrimSpace(parts[0])
-				currentTask.ExitCode = 1
-			} else {
+				return true
+			}
+
+			switch {
+			case parseMarker("✓", 0), parseMarker("PASS", 0):
+				// ok
+			case parseMarker("⚠️", 0), parseMarker("WARN", 0):
+				// warning
+			case parseMarker("✗", 1), parseMarker("FAIL", 1):
+				// fail
+			default:
 				currentTask.TaskID = taskLine
 			}
 		} else if currentTask != nil && inTaskResults {
 			// Parse task details
 			if strings.HasPrefix(line, "Exit code:") {
-				fmt.Sscanf(line, "Exit code: %d", &currentTask.ExitCode)
+				if _, err := fmt.Sscanf(line, "Exit code: %d", &currentTask.ExitCode); err != nil {
+					t.Fatalf("failed to parse exit code from %q: %v", line, err)
+				}
 			} else if strings.HasPrefix(line, "Error:") {
 				currentTask.Error = strings.TrimPrefix(line, "Error: ")
 			} else if strings.HasPrefix(line, "Log:") {
@@ -147,7 +167,9 @@ func parseIntegrationOutput(t *testing.T, out string) integrationOutput {
 				currentTask.ExitCode = 0
 			} else if strings.HasPrefix(line, "Status: FAILED") {
 				if strings.Contains(line, "exit code") {
-					fmt.Sscanf(line, "Status: FAILED (exit code %d)", &currentTask.ExitCode)
+					if _, err := fmt.Sscanf(line, "Status: FAILED (exit code %d)", &currentTask.ExitCode); err != nil {
+						t.Fatalf("failed to parse exit code from %q: %v", line, err)
+					}
 				} else {
 					currentTask.ExitCode = 1
 				}
@@ -180,6 +202,37 @@ func findResultByID(t *testing.T, payload integrationOutput, id string) TaskResu
 	return TaskResult{}
 }

+func setTempDirEnv(t *testing.T, dir string) string {
+	t.Helper()
+	resolved := dir
+	if eval, err := filepath.EvalSymlinks(dir); err == nil {
+		resolved = eval
+	}
+	t.Setenv("TMPDIR", resolved)
+	t.Setenv("TEMP", resolved)
+	t.Setenv("TMP", resolved)
+	return resolved
+}
+
+func createTempLog(t *testing.T, dir, name string) string {
+	t.Helper()
+	path := filepath.Join(dir, name)
+	if err := os.WriteFile(path, []byte("test"), 0o644); err != nil {
+		t.Fatalf("failed to create temp log %s: %v", path, err)
+	}
+	return path
+}
+
+func stubProcessRunning(t *testing.T, fn func(int) bool) {
+	t.Helper()
+	t.Cleanup(logger.SetProcessRunningCheck(fn))
+}
+
+func stubProcessStartTime(t *testing.T, fn func(int) time.Time) {
+	t.Helper()
+	t.Cleanup(logger.SetProcessStartTimeFn(fn))
+}
+
 func TestRunParallelEndToEnd_OrderAndConcurrency(t *testing.T) {
 	defer resetTestHooks()
 	origRun := runCodexTaskFn
@@ -365,7 +418,7 @@ id: beta
 ---CONTENT---
 task-beta`
 	stdinReader = bytes.NewReader([]byte(input))
-	os.Args = []string{"codex-wrapper", "--parallel"}
+	os.Args = []string{"codeagent-wrapper", "--parallel"}

 	var exitCode int
 	output := captureStdout(t, func() {
@@ -418,9 +471,9 @@ id: d
 ---CONTENT---
 ok-d`
 	stdinReader = bytes.NewReader([]byte(input))
-	os.Args = []string{"codex-wrapper", "--parallel"}
+	os.Args = []string{"codeagent-wrapper", "--parallel"}

-	expectedLog := filepath.Join(tempDir, fmt.Sprintf("codex-wrapper-%d.log", os.Getpid()))
+	expectedLog := filepath.Join(tempDir, fmt.Sprintf("codeagent-wrapper-%d.log", os.Getpid()))

 	origRun := runCodexTaskFn
 	runCodexTaskFn = func(task TaskSpec, timeout int) TaskResult {
@@ -474,9 +527,9 @@ ok-d`

 	// After parallel log isolation fix, each task has its own log file
 	expectedLines := map[string]struct{}{
-		fmt.Sprintf("Task a: Log: %s", filepath.Join(tempDir, fmt.Sprintf("codex-wrapper-%d-a.log", os.Getpid()))): {},
-		fmt.Sprintf("Task b: Log: %s", filepath.Join(tempDir, fmt.Sprintf("codex-wrapper-%d-b.log", os.Getpid()))): {},
-		fmt.Sprintf("Task d: Log: %s", filepath.Join(tempDir, fmt.Sprintf("codex-wrapper-%d-d.log", os.Getpid()))): {},
+		fmt.Sprintf("Task a: Log: %s", filepath.Join(tempDir, fmt.Sprintf("codeagent-wrapper-%d-a.log", os.Getpid()))): {},
+		fmt.Sprintf("Task b: Log: %s", filepath.Join(tempDir, fmt.Sprintf("codeagent-wrapper-%d-b.log", os.Getpid()))): {},
+		fmt.Sprintf("Task d: Log: %s", filepath.Join(tempDir, fmt.Sprintf("codeagent-wrapper-%d-d.log", os.Getpid()))): {},
 	}

 	if len(taskLines) != len(expectedLines) {
@@ -494,7 +547,7 @@ func TestRunNonParallelOutputsIncludeLogPathsIntegration(t *testing.T) {
 	defer resetTestHooks()

 	tempDir := setTempDirEnv(t, t.TempDir())
-	os.Args = []string{"codex-wrapper", "integration-log-check"}
+	os.Args = []string{"codeagent-wrapper", "integration-log-check"}
 	stdinReader = strings.NewReader("")
 	isTerminalFn = func() bool { return true }
 	codexCommand = "echo"
@@ -512,7 +565,7 @@ func TestRunNonParallelOutputsIncludeLogPathsIntegration(t *testing.T) {
 	if exitCode != 0 {
 		t.Fatalf("run() exit=%d, want 0", exitCode)
 	}
-	expectedLog := filepath.Join(tempDir, fmt.Sprintf("codex-wrapper-%d.log", os.Getpid()))
+	expectedLog := filepath.Join(tempDir, fmt.Sprintf("codeagent-wrapper-%d.log", os.Getpid()))
 	wantLine := fmt.Sprintf("Log: %s", expectedLog)
 	if !strings.Contains(stderr, wantLine) {
 		t.Fatalf("stderr missing %q, got: %q", wantLine, stderr)
@@ -693,11 +746,11 @@ func TestRunStartupCleanupRemovesOrphansEndToEnd(t *testing.T) {

 	tempDir := setTempDirEnv(t, t.TempDir())

-	orphanA := createTempLog(t, tempDir, "codex-wrapper-5001.log")
-	orphanB := createTempLog(t, tempDir, "codex-wrapper-5002-extra.log")
-	orphanC := createTempLog(t, tempDir, "codex-wrapper-5003-suffix.log")
+	orphanA := createTempLog(t, tempDir, "codeagent-wrapper-5001.log")
+	orphanB := createTempLog(t, tempDir, "codeagent-wrapper-5002-extra.log")
+	orphanC := createTempLog(t, tempDir, "codeagent-wrapper-5003-suffix.log")
 	runningPID := 81234
-	runningLog := createTempLog(t, tempDir, fmt.Sprintf("codex-wrapper-%d.log", runningPID))
+	runningLog := createTempLog(t, tempDir, fmt.Sprintf("codeagent-wrapper-%d.log", runningPID))
 	unrelated := createTempLog(t, tempDir, "wrapper.log")

 	stubProcessRunning(t, func(pid int) bool {
@@ -713,7 +766,7 @@ func TestRunStartupCleanupRemovesOrphansEndToEnd(t *testing.T) {
 	codexCommand = createFakeCodexScript(t, "tid-startup", "ok")
 	stdinReader = strings.NewReader("")
 	isTerminalFn = func() bool { return true }
-	os.Args = []string{"codex-wrapper", "task"}
+	os.Args = []string{"codeagent-wrapper", "task"}

 	if exit := run(); exit != 0 {
 		t.Fatalf("run() exit=%d, want 0", exit)
@@ -739,7 +792,7 @@ func TestRunStartupCleanupConcurrentWrappers(t *testing.T) {

 	const totalLogs = 40
 	for i := 0; i < totalLogs; i++ {
-		createTempLog(t, tempDir, fmt.Sprintf("codex-wrapper-%d.log", 9000+i))
+		createTempLog(t, tempDir, fmt.Sprintf("codeagent-wrapper-%d.log", 9000+i))
 	}

 	stubProcessRunning(t, func(pid int) bool {
@@ -763,7 +816,7 @@ func TestRunStartupCleanupConcurrentWrappers(t *testing.T) {
 	close(start)
 	wg.Wait()

-	matches, err := filepath.Glob(filepath.Join(tempDir, "codex-wrapper-*.log"))
+	matches, err := filepath.Glob(filepath.Join(tempDir, "codeagent-wrapper-*.log"))
 	if err != nil {
 		t.Fatalf("glob error: %v", err)
 	}
@@ -777,9 +830,9 @@ func TestRunCleanupFlagEndToEnd_Success(t *testing.T) {

 	tempDir := setTempDirEnv(t, t.TempDir())

-	staleA := createTempLog(t, tempDir, "codex-wrapper-2100.log")
-	staleB := createTempLog(t, tempDir, "codex-wrapper-2200-extra.log")
-	keeper := createTempLog(t, tempDir, "codex-wrapper-2300.log")
+	staleA := createTempLog(t, tempDir, "codeagent-wrapper-2100.log")
+	staleB := createTempLog(t, tempDir, "codeagent-wrapper-2200-extra.log")
+	keeper := createTempLog(t, tempDir, "codeagent-wrapper-2300.log")

 	stubProcessRunning(t, func(pid int) bool {
 		return pid == 2300 || pid == os.Getpid()
@@ -791,7 +844,7 @@ func TestRunCleanupFlagEndToEnd_Success(t *testing.T) {
 		return time.Time{}
 	})

-	os.Args = []string{"codex-wrapper", "--cleanup"}
+	os.Args = []string{"codeagent-wrapper", "--cleanup"}

 	var exitCode int
 	output := captureStdout(t, func() {
@@ -815,10 +868,10 @@ func TestRunCleanupFlagEndToEnd_Success(t *testing.T) {
 	if !strings.Contains(output, "Files kept: 1") {
 		t.Fatalf("missing 'Files kept: 1' in output: %q", output)
 	}
-	if !strings.Contains(output, "codex-wrapper-2100.log") || !strings.Contains(output, "codex-wrapper-2200-extra.log") {
+	if !strings.Contains(output, "codeagent-wrapper-2100.log") || !strings.Contains(output, "codeagent-wrapper-2200-extra.log") {
 		t.Fatalf("missing deleted file names in output: %q", output)
 	}
-	if !strings.Contains(output, "codex-wrapper-2300.log") {
+	if !strings.Contains(output, "codeagent-wrapper-2300.log") {
 		t.Fatalf("missing kept file names in output: %q", output)
 	}

@@ -831,7 +884,7 @@ func TestRunCleanupFlagEndToEnd_Success(t *testing.T) {
 		t.Fatalf("expected kept log to remain, err=%v", err)
 	}

-	currentLog := filepath.Join(tempDir, fmt.Sprintf("codex-wrapper-%d.log", os.Getpid()))
+	currentLog := filepath.Join(tempDir, fmt.Sprintf("codeagent-wrapper-%d.log", os.Getpid()))
 	if _, err := os.Stat(currentLog); err == nil {
 		t.Fatalf("cleanup mode should not create new log file %s", currentLog)
 	} else if !os.IsNotExist(err) {
@@ -850,7 +903,7 @@ func TestRunCleanupFlagEndToEnd_FailureDoesNotAffectStartup(t *testing.T) {
 		return CleanupStats{Scanned: 1}, fmt.Errorf("permission denied")
 	}

-	os.Args = []string{"codex-wrapper", "--cleanup"}
+	os.Args = []string{"codeagent-wrapper", "--cleanup"}

 	var exitCode int
 	errOutput := captureStderr(t, func() {
@@ -867,7 +920,7 @@ func TestRunCleanupFlagEndToEnd_FailureDoesNotAffectStartup(t *testing.T) {
 		t.Fatalf("cleanup called %d times, want 1", calls)
 	}

-	currentLog := filepath.Join(tempDir, fmt.Sprintf("codex-wrapper-%d.log", os.Getpid()))
+	currentLog := filepath.Join(tempDir, fmt.Sprintf("codeagent-wrapper-%d.log", os.Getpid()))
 	if _, err := os.Stat(currentLog); err == nil {
 		t.Fatalf("cleanup failure should not create new log file %s", currentLog)
 	} else if !os.IsNotExist(err) {
@@ -880,7 +933,7 @@ func TestRunCleanupFlagEndToEnd_FailureDoesNotAffectStartup(t *testing.T) {
 	codexCommand = createFakeCodexScript(t, "tid-cleanup-e2e", "ok")
 	stdinReader = strings.NewReader("")
 	isTerminalFn = func() bool { return true }
-	os.Args = []string{"codex-wrapper", "post-cleanup task"}
+	os.Args = []string{"codeagent-wrapper", "post-cleanup task"}

 	var normalExit int
 	normalOutput := captureStdout(t, func() {
--- a/codeagent-wrapper/internal/app/main_test.go
+++ b/codeagent-wrapper/internal/app/main_test.go
@@ -1,10 +1,9 @@
-package main
+package wrapper

 import (
 	"bufio"
 	"bytes"
 	"context"
-	"encoding/json"
 	"errors"
 	"fmt"
 	"io"
@@ -19,6 +18,11 @@ import (
 	"syscall"
 	"testing"
 	"time"
+
+	config "codeagent-wrapper/internal/config"
+	executor "codeagent-wrapper/internal/executor"
+
+	"github.com/goccy/go-json"
 )

 // Helper to reset test hooks
@@ -28,17 +32,15 @@ func resetTestHooks() {
 	codexCommand = "codex"
 	cleanupHook = nil
 	cleanupLogsFn = cleanupOldLogs
-	signalNotifyFn = signal.Notify
-	signalStopFn = signal.Stop
+	startupCleanupAsync = false
+	config.ResetModelsConfigCacheForTest()
+	_ = executor.SetSelectBackendFn(nil)
 	buildCodexArgsFn = buildCodexArgs
 	selectBackendFn = selectBackend
-	commandContext = exec.CommandContext
-	newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
-		return &realCmd{cmd: commandContext(ctx, name, args...)}
-	}
-	forceKillDelay.Store(5)
-	closeLogger()
-	executablePathFn = os.Executable
+	_ = executor.SetCommandContextFn(nil)
+	_ = executor.SetNewCommandRunner(nil)
+	_ = executor.SetForceKillDelay(5)
+	_ = closeLogger()
 	runTaskFn = runCodexTask
 	runCodexTaskFn = defaultRunCodexTaskFn
 	exitFn = os.Exit
@@ -86,6 +88,8 @@ func (t testBackend) Command() string {
 	return "echo"
 }

+func (t testBackend) Env(baseURL, apiKey string) map[string]string { return nil }
+
 func withBackend(command string, argsFn func(*Config, string) []string) func() {
 	prev := selectBackendFn
 	selectBackendFn = func(name string) (Backend, error) {
@@ -107,7 +111,7 @@ func restoreStdoutPipe(c *capturedStdout) {
 	}
 	c.writer.Close()
 	os.Stdout = c.old
-	io.Copy(&c.buf, c.reader)
+	_, _ = io.Copy(&c.buf, c.reader)
 }

 func (c *capturedStdout) String() string {
@@ -127,7 +131,9 @@ func captureOutput(t *testing.T, fn func()) string {
 	os.Stdout = old

 	var buf bytes.Buffer
-	io.Copy(&buf, r)
+	if _, err := io.Copy(&buf, r); err != nil {
+		t.Fatalf("io.Copy() error = %v", err)
+	}
 	return buf.String()
 }

@@ -141,7 +147,9 @@ func captureStderr(t *testing.T, fn func()) string {
 	os.Stderr = old

 	var buf bytes.Buffer
-	io.Copy(&buf, r)
+	if _, err := io.Copy(&buf, r); err != nil {
+		t.Fatalf("io.Copy() error = %v", err)
+	}
 	return buf.String()
 }

@@ -262,7 +270,7 @@ func (d *drainBlockingCmd) SetEnv(env map[string]string) {
 	d.inner.SetEnv(env)
 }

-func (d *drainBlockingCmd) Process() processHandle {
+func (d *drainBlockingCmd) Process() executor.ProcessHandle {
 	return d.inner.Process()
 }

@@ -553,7 +561,7 @@ func (f *fakeCmd) SetEnv(env map[string]string) {
 	}
 }

-func (f *fakeCmd) Process() processHandle {
+func (f *fakeCmd) Process() executor.ProcessHandle {
 	if f == nil {
 		return nil
 	}
@@ -728,9 +736,7 @@ func TestFakeCmdInfra(t *testing.T) {
 			WaitDelay: 5 * time.Millisecond,
 		})

-		newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
-			return fake
-		}
+		_ = executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner { return fake })
 		buildCodexArgsFn = func(cfg *Config, targetArg string) []string {
 			return []string{targetArg}
 		}
@@ -774,9 +780,7 @@ func TestRunCodexTask_WaitBeforeParse(t *testing.T) {
 		WaitDelay: waitDelay,
 	})

-	newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
-		return fake
-	}
+	_ = executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner { return fake })
 	buildCodexArgsFn = func(cfg *Config, targetArg string) []string {
 		return []string{targetArg}
 	}
@@ -821,9 +825,7 @@ func TestRunCodexTask_ParseStall(t *testing.T) {
 	})

 	blockingCmd := newDrainBlockingCmd(fake)
-	newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
-		return blockingCmd
-	}
+	_ = executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner { return blockingCmd })
 	buildCodexArgsFn = func(cfg *Config, targetArg string) []string {
 		return []string{targetArg}
 	}
@@ -878,7 +880,7 @@ func TestRunCodexTask_ParseStall(t *testing.T) {

 func TestRunCodexTask_ContextTimeout(t *testing.T) {
 	defer resetTestHooks()
-	forceKillDelay.Store(0)
+	_ = executor.SetForceKillDelay(0)

 	fake := newFakeCmd(fakeCmdConfig{
 		KeepStdoutOpen:      true,
@@ -887,9 +889,7 @@ func TestRunCodexTask_ContextTimeout(t *testing.T) {
 		ReleaseWaitOnSignal: false,
 	})

-	newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
-		return fake
-	}
+	_ = executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner { return fake })
 	buildCodexArgsFn = func(cfg *Config, targetArg string) []string {
 		return []string{targetArg}
 	}
@@ -898,14 +898,6 @@ func TestRunCodexTask_ContextTimeout(t *testing.T) {
 	ctx, cancel := context.WithTimeout(context.Background(), 200*time.Millisecond)
 	defer cancel()

-	var capturedTimer *forceKillTimer
-	terminateCommandFn = func(cmd commandRunner) *forceKillTimer {
-		timer := terminateCommand(cmd)
-		capturedTimer = timer
-		return timer
-	}
-	defer func() { terminateCommandFn = terminateCommand }()
-
 	result := runCodexTaskWithContext(ctx, TaskSpec{Task: "ctx-timeout", WorkDir: defaultWorkdir}, nil, nil, false, false, 60)

 	if result.ExitCode != 124 {
@@ -929,15 +921,6 @@ func TestRunCodexTask_ContextTimeout(t *testing.T) {
 			t.Fatalf("expected Kill to eventually run, got 0")
 		}
 	}
-	if capturedTimer == nil {
-		t.Fatalf("forceKillTimer not captured")
-	}
-	if !capturedTimer.stopped.Load() {
-		t.Fatalf("forceKillTimer.Stop was not called")
-	}
-	if !capturedTimer.drained.Load() {
-		t.Fatalf("forceKillTimer drain logic did not run")
-	}
 	if fake.stdout == nil {
 		t.Fatalf("stdout reader not initialized")
 	}
@@ -948,7 +931,7 @@ func TestRunCodexTask_ContextTimeout(t *testing.T) {

 func TestRunCodexTask_ForcesStopAfterCompletion(t *testing.T) {
 	defer resetTestHooks()
-	forceKillDelay.Store(0)
+	_ = executor.SetForceKillDelay(0)

 	fake := newFakeCmd(fakeCmdConfig{
 		StdoutPlan: []fakeStdoutEvent{
@@ -961,9 +944,7 @@ func TestRunCodexTask_ForcesStopAfterCompletion(t *testing.T) {
 		ReleaseWaitOnKill:   true,
 	})

-	newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
-		return fake
-	}
+	_ = executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner { return fake })
 	buildCodexArgsFn = func(cfg *Config, targetArg string) []string { return []string{targetArg} }
 	codexCommand = "fake-cmd"

@@ -988,7 +969,7 @@ func TestRunCodexTask_ForcesStopAfterCompletion(t *testing.T) {

 func TestRunCodexTask_ForcesStopAfterTurnCompleted(t *testing.T) {
 	defer resetTestHooks()
-	forceKillDelay.Store(0)
+	_ = executor.SetForceKillDelay(0)

 	fake := newFakeCmd(fakeCmdConfig{
 		StdoutPlan: []fakeStdoutEvent{
@@ -1001,9 +982,7 @@ func TestRunCodexTask_ForcesStopAfterTurnCompleted(t *testing.T) {
 		ReleaseWaitOnKill:   true,
 	})

-	newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
-		return fake
-	}
+	_ = executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner { return fake })
 	buildCodexArgsFn = func(cfg *Config, targetArg string) []string { return []string{targetArg} }
 	codexCommand = "fake-cmd"

@@ -1028,7 +1007,7 @@ func TestRunCodexTask_ForcesStopAfterTurnCompleted(t *testing.T) {

 func TestRunCodexTask_DoesNotTerminateBeforeThreadCompleted(t *testing.T) {
 	defer resetTestHooks()
-	forceKillDelay.Store(0)
+	_ = executor.SetForceKillDelay(0)

 	fake := newFakeCmd(fakeCmdConfig{
 		StdoutPlan: []fakeStdoutEvent{
@@ -1042,9 +1021,7 @@ func TestRunCodexTask_DoesNotTerminateBeforeThreadCompleted(t *testing.T) {
 		ReleaseWaitOnKill:   true,
 	})

-	newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
-		return fake
-	}
+	_ = executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner { return fake })
 	buildCodexArgsFn = func(cfg *Config, targetArg string) []string { return []string{targetArg} }
 	codexCommand = "fake-cmd"

@@ -1498,7 +1475,7 @@ func TestBackendParseBoolFlag(t *testing.T) {

 	for _, tt := range tests {
 		t.Run(tt.name, func(t *testing.T) {
-			if got := parseBoolFlag(tt.val, tt.def); got != tt.want {
+			if got := config.ParseBoolFlag(tt.val, tt.def); got != tt.want {
 				t.Fatalf("parseBoolFlag(%q,%v) = %v, want %v", tt.val, tt.def, got, tt.want)
 			}
 		})
@@ -1508,17 +1485,17 @@ func TestBackendParseBoolFlag(t *testing.T) {
 func TestBackendEnvFlagEnabled(t *testing.T) {
 	const key = "TEST_FLAG_ENABLED"
 	t.Setenv(key, "")
-	if envFlagEnabled(key) {
+	if config.EnvFlagEnabled(key) {
 		t.Fatalf("envFlagEnabled should be false when unset")
 	}

 	t.Setenv(key, "true")
-	if !envFlagEnabled(key) {
+	if !config.EnvFlagEnabled(key) {
 		t.Fatalf("envFlagEnabled should be true for 'true'")
 	}

 	t.Setenv(key, "no")
-	if envFlagEnabled(key) {
+	if config.EnvFlagEnabled(key) {
 		t.Fatalf("envFlagEnabled should be false for 'no'")
 	}
 }
@@ -1708,8 +1685,8 @@ func TestClaudeModel_DefaultsFromSettings(t *testing.T) {
 		t.Fatalf("WriteFile: %v", err)
 	}

-	makeRunner := func(gotName *string, gotArgs *[]string, fake **fakeCmd) func(context.Context, string, ...string) commandRunner {
-		return func(ctx context.Context, name string, args ...string) commandRunner {
+	makeRunner := func(gotName *string, gotArgs *[]string, fake **fakeCmd) func(context.Context, string, ...string) executor.CommandRunner {
+		return func(ctx context.Context, name string, args ...string) executor.CommandRunner {
 			*gotName = name
 			*gotArgs = append([]string(nil), args...)
 			cmd := newFakeCmd(fakeCmdConfig{
@@ -1729,9 +1706,8 @@ func TestClaudeModel_DefaultsFromSettings(t *testing.T) {
 			gotArgs []string
 			fake    *fakeCmd
 		)
-		origRunner := newCommandRunner
-		newCommandRunner = makeRunner(&gotName, &gotArgs, &fake)
-		t.Cleanup(func() { newCommandRunner = origRunner })
+		restore := executor.SetNewCommandRunner(makeRunner(&gotName, &gotArgs, &fake))
+		t.Cleanup(restore)

 		res := runCodexTaskWithContext(context.Background(), TaskSpec{Task: "hi", Mode: "new", WorkDir: defaultWorkdir}, ClaudeBackend{}, nil, false, true, 5)
 		if res.ExitCode != 0 || res.Message != "ok" {
@@ -1761,9 +1737,8 @@ func TestClaudeModel_DefaultsFromSettings(t *testing.T) {
 			gotArgs []string
 			fake    *fakeCmd
 		)
-		origRunner := newCommandRunner
-		newCommandRunner = makeRunner(&gotName, &gotArgs, &fake)
-		t.Cleanup(func() { newCommandRunner = origRunner })
+		restore := executor.SetNewCommandRunner(makeRunner(&gotName, &gotArgs, &fake))
+		t.Cleanup(restore)

 		res := runCodexTaskWithContext(context.Background(), TaskSpec{Task: "hi", Mode: "new", WorkDir: defaultWorkdir, Model: "sonnet"}, ClaudeBackend{}, nil, false, true, 5)
 		if res.ExitCode != 0 || res.Message != "ok" {
@@ -1787,9 +1762,8 @@ func TestClaudeModel_DefaultsFromSettings(t *testing.T) {
 			gotArgs []string
 			fake    *fakeCmd
 		)
-		origRunner := newCommandRunner
-		newCommandRunner = makeRunner(&gotName, &gotArgs, &fake)
-		t.Cleanup(func() { newCommandRunner = origRunner })
+		restore := executor.SetNewCommandRunner(makeRunner(&gotName, &gotArgs, &fake))
+		t.Cleanup(restore)

 		res := runCodexTaskWithContext(context.Background(), TaskSpec{Task: "hi", Mode: "resume", SessionID: "sid-123", WorkDir: defaultWorkdir}, ClaudeBackend{}, nil, false, true, 5)
 		if res.ExitCode != 0 || res.Message != "ok" {
@@ -1939,6 +1913,37 @@ func TestRun_PassesReasoningEffortToTaskSpec(t *testing.T) {
 	}
 }

+func TestRun_NoOutputMessage_ReturnsExitCode1AndWritesStderr(t *testing.T) {
+	defer resetTestHooks()
+	cleanupLogsFn = func() (CleanupStats, error) { return CleanupStats{}, nil }
+	t.Setenv("TMPDIR", t.TempDir())
+
+	selectBackendFn = func(name string) (Backend, error) {
+		return testBackend{name: name, command: "echo"}, nil
+	}
+
+	runTaskFn = func(task TaskSpec, silent bool, timeout int) TaskResult {
+		return TaskResult{ExitCode: 0, Message: ""}
+	}
+
+	isTerminalFn = func() bool { return true }
+	stdinReader = strings.NewReader("")
+
+	os.Args = []string{"codeagent-wrapper", "task"}
+
+	var code int
+	errOutput := captureStderr(t, func() {
+		code = run()
+	})
+
+	if code != 1 {
+		t.Fatalf("run() exit=%d, want 1", code)
+	}
+	if !strings.Contains(errOutput, "no output message") {
+		t.Fatalf("stderr missing sentinel error text; got:\n%s", errOutput)
+	}
+}
+
 func TestRunBuildCodexArgs_NewMode(t *testing.T) {
 	const key = "CODEX_BYPASS_SANDBOX"
 	t.Setenv(key, "false")
@@ -1970,7 +1975,7 @@ func TestRunBuildCodexArgs_NewMode_WithReasoningEffort(t *testing.T) {
 	args := buildCodexArgs(cfg, "my task")
 	expected := []string{
 		"e",
-		"--reasoning-effort", "high",
+		"-c", "model_reasoning_effort=high",
 		"--skip-git-repo-check",
 		"-C", "/test/dir",
 		"--json",
@@ -1991,8 +1996,7 @@ func TestRunCodexTaskWithContext_CodexReasoningEffort(t *testing.T) {
 	t.Setenv("CODEX_BYPASS_SANDBOX", "false")

 	var gotArgs []string
-	origRunner := newCommandRunner
-	newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
+	restore := executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner {
 		gotArgs = append([]string(nil), args...)
 		return newFakeCmd(fakeCmdConfig{
 			PID: 123,
@@ -2000,8 +2004,8 @@ func TestRunCodexTaskWithContext_CodexReasoningEffort(t *testing.T) {
 				{Data: "{\"type\":\"result\",\"session_id\":\"sid\",\"result\":\"ok\"}\n"},
 			},
 		})
-	}
-	t.Cleanup(func() { newCommandRunner = origRunner })
+	})
+	t.Cleanup(restore)

 	res := runCodexTaskWithContext(context.Background(), TaskSpec{Task: "hi", Mode: "new", WorkDir: defaultWorkdir, ReasoningEffort: "high"}, nil, nil, false, true, 5)
 	if res.ExitCode != 0 || res.Message != "ok" {
@@ -2010,13 +2014,13 @@ func TestRunCodexTaskWithContext_CodexReasoningEffort(t *testing.T) {

 	found := false
 	for i := 0; i+1 < len(gotArgs); i++ {
-		if gotArgs[i] == "--reasoning-effort" && gotArgs[i+1] == "high" {
+		if gotArgs[i] == "-c" && gotArgs[i+1] == "model_reasoning_effort=high" {
 			found = true
 			break
 		}
 	}
 	if !found {
-		t.Fatalf("expected --reasoning-effort high in args, got %v", gotArgs)
+		t.Fatalf("expected -c model_reasoning_effort=high in args, got %v", gotArgs)
 	}
 }

@@ -2071,7 +2075,7 @@ func TestRunBuildCodexArgs_BypassSandboxEnvTrue(t *testing.T) {
 		t.Fatalf("NewLogger() error = %v", err)
 	}
 	setLogger(logger)
-	defer closeLogger()
+	defer func() { _ = closeLogger() }()

 	t.Setenv("CODEX_BYPASS_SANDBOX", "true")

@@ -2716,7 +2720,7 @@ func TestRunLogFunctions(t *testing.T) {
 		t.Fatalf("NewLogger() error = %v", err)
 	}
 	setLogger(logger)
-	defer closeLogger()
+	defer func() { _ = closeLogger() }()

 	logInfo("info message")
 	logWarn("warn message")
@@ -2751,25 +2755,38 @@ func TestLoggerPathAndRemoveNil(t *testing.T) {
 }

 func TestLoggerLogDropOnDone(t *testing.T) {
-	logger := &Logger{
-		ch:   make(chan logEntry),
-		done: make(chan struct{}),
-	}
-	close(logger.done)
-	logger.log("INFO", "dropped")
-	logger.pendingWG.Wait()
+	t.Skip("internal logger behavior moved to internal/logger; exercise via public methods instead")
 }

 func TestLoggerLogAfterClose(t *testing.T) {
 	defer resetTestHooks()
+	tempDir := t.TempDir()
+	t.Setenv("TMPDIR", tempDir)
+
 	logger, err := NewLogger()
 	if err != nil {
 		t.Fatalf("NewLogger error: %v", err)
 	}
+	logPath := logger.Path()
+	t.Cleanup(func() { _ = os.Remove(logPath) })
+
+	logger.Info("before close")
+	logger.Flush()
+
 	if err := logger.Close(); err != nil {
 		t.Fatalf("Close error: %v", err)
 	}
-	logger.log("INFO", "should be ignored")
+
+	logger.Info("should be ignored")
+	logger.Flush()
+
+	data, err := os.ReadFile(logPath)
+	if err != nil {
+		t.Fatalf("failed to read log file: %v", err)
+	}
+	if strings.Contains(string(data), "should be ignored") {
+		t.Fatalf("expected log message to be dropped after Close, got: %s", string(data))
+	}
 }

 func TestLogWriterLogLine(t *testing.T) {
@@ -2788,7 +2805,7 @@ func TestLogWriterLogLine(t *testing.T) {
 	if !strings.Contains(string(data), "P:abc") {
 		t.Fatalf("log output missing truncated entry, got %q", string(data))
 	}
-	closeLogger()
+	_ = closeLogger()
 }

 func TestNewLogWriterDefaultMaxLen(t *testing.T) {
@@ -2807,7 +2824,9 @@ func TestBackendPrintHelp(t *testing.T) {
 	os.Stdout = oldStdout

 	var buf bytes.Buffer
-	io.Copy(&buf, r)
+	if _, err := io.Copy(&buf, r); err != nil {
+		t.Fatalf("io.Copy() error = %v", err)
+	}
 	output := buf.String()

 	expected := []string{"codeagent-wrapper", "Usage:", "resume", "CODEX_TIMEOUT", "Exit Codes:"}
@@ -2929,12 +2948,12 @@ func TestRunCodexTaskFn_UsesTaskBackend(t *testing.T) {

 	var seenName string
 	var seenArgs []string
-	newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
+	_ = executor.SetNewCommandRunner(func(ctx context.Context, name string, args ...string) executor.CommandRunner {
 		seenName = name
 		seenArgs = append([]string(nil), args...)
 		return fake
-	}
-	selectBackendFn = func(name string) (Backend, error) {
+	})
+	_ = executor.SetSelectBackendFn(func(name string) (Backend, error) {
 		return testBackend{
 			name:    strings.ToLower(name),
 			command: "custom-cli",
@@ -2942,7 +2961,7 @@ func TestRunCodexTaskFn_UsesTaskBackend(t *testing.T) {
 				return []string{"do", targetArg}
 			},
 		}, nil
-	}
+	})

 	res := runCodexTaskFn(TaskSpec{ID: "task-1", Task: "payload", Backend: "Custom"}, 5)

@@ -2966,9 +2985,9 @@ func TestRunCodexTaskFn_UsesTaskBackend(t *testing.T) {
 func TestRunCodexTaskFn_InvalidBackend(t *testing.T) {
 	defer resetTestHooks()

-	selectBackendFn = func(name string) (Backend, error) {
+	_ = executor.SetSelectBackendFn(func(name string) (Backend, error) {
 		return nil, fmt.Errorf("invalid backend: %s", name)
-	}
+	})

 	res := runCodexTaskFn(TaskSpec{ID: "bad-task", Task: "noop", Backend: "unknown"}, 5)
 	if res.ExitCode == 0 {
@@ -3109,11 +3128,11 @@ func TestRunCodexTask_ExitError(t *testing.T) {

 func TestRunCodexTask_StdinPipeError(t *testing.T) {
 	defer resetTestHooks()
-	commandContext = func(ctx context.Context, name string, args ...string) *exec.Cmd {
+	_ = executor.SetCommandContextFn(func(ctx context.Context, name string, args ...string) *exec.Cmd {
 		cmd := exec.CommandContext(ctx, "cat")
 		cmd.Stdin = os.Stdin
 		return cmd
-	}
+	})
 	buildCodexArgsFn = func(cfg *Config, targetArg string) []string { return []string{} }
 	res := runCodexTask(TaskSpec{Task: "data", UseStdin: true}, false, 1)
 	if res.ExitCode != 1 || !strings.Contains(res.Error, "stdin pipe") {
@@ -3123,11 +3142,11 @@ func TestRunCodexTask_StdinPipeError(t *testing.T) {

 func TestRunCodexTask_StdoutPipeError(t *testing.T) {
 	defer resetTestHooks()
-	commandContext = func(ctx context.Context, name string, args ...string) *exec.Cmd {
+	_ = executor.SetCommandContextFn(func(ctx context.Context, name string, args ...string) *exec.Cmd {
 		cmd := exec.CommandContext(ctx, "echo", "noop")
 		cmd.Stdout = os.Stdout
 		return cmd
-	}
+	})
 	buildCodexArgsFn = func(cfg *Config, targetArg string) []string { return []string{} }
 	res := runCodexTask(TaskSpec{Task: "noop"}, false, 1)
 	if res.ExitCode != 1 || !strings.Contains(res.Error, "stdout pipe") {
@@ -3170,36 +3189,6 @@ func TestRunCodexTask_SignalHandling(t *testing.T) {
 	}
 }

-func TestForwardSignals_ContextCancel(t *testing.T) {
-	ctx, cancel := context.WithCancel(context.Background())
-	defer cancel()
-	forwardSignals(ctx, &realCmd{cmd: &exec.Cmd{}}, func(string) {})
-	cancel()
-	time.Sleep(10 * time.Millisecond)
-}
-
-func TestCancelReason(t *testing.T) {
-	const cmdName = "codex"
-
-	if got := cancelReason(cmdName, nil); got != "Context cancelled" {
-		t.Fatalf("cancelReason(nil) = %q, want %q", got, "Context cancelled")
-	}
-
-	ctxTimeout, cancelTimeout := context.WithTimeout(context.Background(), 1*time.Nanosecond)
-	defer cancelTimeout()
-	<-ctxTimeout.Done()
-	wantTimeout := fmt.Sprintf("%s execution timeout", cmdName)
-	if got := cancelReason(cmdName, ctxTimeout); got != wantTimeout {
-		t.Fatalf("cancelReason(deadline) = %q, want %q", got, wantTimeout)
-	}
-
-	ctxCancelled, cancel := context.WithCancel(context.Background())
-	cancel()
-	if got := cancelReason(cmdName, ctxCancelled); got != "Execution cancelled, terminating codex process" {
-		t.Fatalf("cancelReason(cancelled) = %q, want %q", got, "Execution cancelled, terminating codex process")
-	}
-}
-
 func TestRunCodexProcess(t *testing.T) {
 	defer resetTestHooks()
 	script := createFakeCodexScript(t, "proc-thread", "proc-msg")
@@ -3235,7 +3224,9 @@ func TestRunSilentMode(t *testing.T) {
 		w.Close()
 		os.Stderr = oldStderr
 		var buf bytes.Buffer
-		io.Copy(&buf, r)
+		if _, err := io.Copy(&buf, r); err != nil {
+			t.Fatalf("io.Copy() error = %v", err)
+		}
 		return buf.String()
 	}

@@ -3336,35 +3327,6 @@ func TestParallelTopologicalSortTasks(t *testing.T) {
 	}
 }

-func TestRunShouldSkipTask(t *testing.T) {
-	failed := map[string]TaskResult{"a": {TaskID: "a", ExitCode: 1}, "b": {TaskID: "b", ExitCode: 2}}
-	tests := []struct {
-		name           string
-		task           TaskSpec
-		skip           bool
-		reasonContains []string
-	}{
-		{"no deps", TaskSpec{ID: "c"}, false, nil},
-		{"missing deps not failed", TaskSpec{ID: "d", Dependencies: []string{"x"}}, false, nil},
-		{"single failed dep", TaskSpec{ID: "e", Dependencies: []string{"a"}}, true, []string{"a"}},
-		{"multiple failed deps", TaskSpec{ID: "f", Dependencies: []string{"a", "b"}}, true, []string{"a", "b"}},
-	}
-
-	for _, tt := range tests {
-		t.Run(tt.name, func(t *testing.T) {
-			skip, reason := shouldSkipTask(tt.task, failed)
-			if skip != tt.skip {
-				t.Fatalf("skip=%v, want %v", skip, tt.skip)
-			}
-			for _, expect := range tt.reasonContains {
-				if !strings.Contains(reason, expect) {
-					t.Fatalf("reason %q missing %q", reason, expect)
-				}
-			}
-		})
-	}
-}
-
 func TestRunTopologicalSort_CycleDetection(t *testing.T) {
 	tasks := []TaskSpec{{ID: "a", Dependencies: []string{"b"}}, {ID: "b", Dependencies: []string{"a"}}}
 	if _, err := topologicalSort(tasks); err == nil || !strings.Contains(err.Error(), "cycle detected") {
@@ -3701,7 +3663,7 @@ func TestParallelTriggersCleanup(t *testing.T) {
 	oldArgs := os.Args
 	defer func() { os.Args = oldArgs }()

-	os.Args = []string{"codex-wrapper", "--parallel"}
+	os.Args = []string{"codeagent-wrapper", "--parallel"}
 	stdinReader = strings.NewReader(`---TASK---
 id: only
 ---CONTENT---
@@ -3736,10 +3698,8 @@ func TestVersionFlag(t *testing.T) {
 		}
 	})

-	want := "codeagent-wrapper version 5.5.0\n"
-
-	if output != want {
-		t.Fatalf("output = %q, want %q", output, want)
+	if !strings.HasPrefix(output, "codeagent-wrapper version ") {
+		t.Fatalf("output = %q, want prefix %q", output, "codeagent-wrapper version ")
 	}
 }

@@ -3752,26 +3712,22 @@ func TestVersionShortFlag(t *testing.T) {
 		}
 	})

-	want := "codeagent-wrapper version 5.5.0\n"
-
-	if output != want {
-		t.Fatalf("output = %q, want %q", output, want)
+	if !strings.HasPrefix(output, "codeagent-wrapper version ") {
+		t.Fatalf("output = %q, want prefix %q", output, "codeagent-wrapper version ")
 	}
 }

 func TestVersionLegacyAlias(t *testing.T) {
 	defer resetTestHooks()
-	os.Args = []string{"codex-wrapper", "--version"}
+	os.Args = []string{"codeagent-wrapper", "--version"}
 	output := captureOutput(t, func() {
 		if code := run(); code != 0 {
 			t.Errorf("exit = %d, want 0", code)
 		}
 	})

-	want := "codex-wrapper version 5.5.0\n"
-
-	if output != want {
-		t.Fatalf("output = %q, want %q", output, want)
+	if !strings.HasPrefix(output, "codeagent-wrapper version ") {
+		t.Fatalf("output = %q, want prefix %q", output, "codeagent-wrapper version ")
 	}
 }

@@ -3793,7 +3749,7 @@ func TestRun_HelpShort(t *testing.T) {

 func TestRun_HelpDoesNotTriggerCleanup(t *testing.T) {
 	defer resetTestHooks()
-	os.Args = []string{"codex-wrapper", "--help"}
+	os.Args = []string{"codeagent-wrapper", "--help"}
 	cleanupLogsFn = func() (CleanupStats, error) {
 		t.Fatalf("cleanup should not run for --help")
 		return CleanupStats{}, nil
@@ -3806,7 +3762,7 @@ func TestRun_HelpDoesNotTriggerCleanup(t *testing.T) {

 func TestVersionDoesNotTriggerCleanup(t *testing.T) {
 	defer resetTestHooks()
-	os.Args = []string{"codex-wrapper", "--version"}
+	os.Args = []string{"codeagent-wrapper", "--version"}
 	cleanupLogsFn = func() (CleanupStats, error) {
 		t.Fatalf("cleanup should not run for --version")
 		return CleanupStats{}, nil
@@ -3874,7 +3830,7 @@ func TestVersionCoverageFullRun(t *testing.T) {

 		_ = closeLogger()
 		_ = logger.RemoveLogFile()
-		loggerPtr.Store(nil)
+		setLogger(nil)
 	})

 	t.Run("parseArgsError", func(t *testing.T) {
@@ -4089,7 +4045,7 @@ func TestVersionMainWrapper(t *testing.T) {
 	exitCalled := -1
 	exitFn = func(code int) { exitCalled = code }
 	os.Args = []string{"codeagent-wrapper", "--version"}
-	main()
+	Main()
 	if exitCalled != 0 {
 		t.Fatalf("main exit = %d, want 0", exitCalled)
 	}
@@ -4102,8 +4058,8 @@ func TestBackendCleanupMode_Success(t *testing.T) {
 			Scanned:      5,
 			Deleted:      3,
 			Kept:         2,
-			DeletedFiles: []string{"codex-wrapper-111.log", "codex-wrapper-222.log", "codex-wrapper-333.log"},
-			KeptFiles:    []string{"codex-wrapper-444.log", "codex-wrapper-555.log"},
+			DeletedFiles: []string{"codeagent-wrapper-111.log", "codeagent-wrapper-222.log", "codeagent-wrapper-333.log"},
+			KeptFiles:    []string{"codeagent-wrapper-444.log", "codeagent-wrapper-555.log"},
 		}, nil
 	}

@@ -4114,7 +4070,7 @@ func TestBackendCleanupMode_Success(t *testing.T) {
 	if exitCode != 0 {
 		t.Fatalf("exit = %d, want 0", exitCode)
 	}
-	want := "Cleanup completed\nFiles scanned: 5\nFiles deleted: 3\n  - codex-wrapper-111.log\n  - codex-wrapper-222.log\n  - codex-wrapper-333.log\nFiles kept: 2\n  - codex-wrapper-444.log\n  - codex-wrapper-555.log\n"
+	want := "Cleanup completed\nFiles scanned: 5\nFiles deleted: 3\n  - codeagent-wrapper-111.log\n  - codeagent-wrapper-222.log\n  - codeagent-wrapper-333.log\nFiles kept: 2\n  - codeagent-wrapper-444.log\n  - codeagent-wrapper-555.log\n"
 	if output != want {
 		t.Fatalf("output = %q, want %q", output, want)
 	}
@@ -4128,7 +4084,7 @@ func TestBackendCleanupMode_SuccessWithErrorsLine(t *testing.T) {
 			Deleted:      1,
 			Kept:         0,
 			Errors:       1,
-			DeletedFiles: []string{"codex-wrapper-123.log"},
+			DeletedFiles: []string{"codeagent-wrapper-123.log"},
 		}, nil
 	}

@@ -4139,7 +4095,7 @@ func TestBackendCleanupMode_SuccessWithErrorsLine(t *testing.T) {
 	if exitCode != 0 {
 		t.Fatalf("exit = %d, want 0", exitCode)
 	}
-	want := "Cleanup completed\nFiles scanned: 2\nFiles deleted: 1\n  - codex-wrapper-123.log\nFiles kept: 0\nDeletion errors: 1\n"
+	want := "Cleanup completed\nFiles scanned: 2\nFiles deleted: 1\n  - codeagent-wrapper-123.log\nFiles kept: 0\nDeletion errors: 1\n"
 	if output != want {
 		t.Fatalf("output = %q, want %q", output, want)
 	}
@@ -4208,7 +4164,7 @@ func TestRun_CleanupFlag(t *testing.T) {
 	oldArgs := os.Args
 	defer func() { os.Args = oldArgs }()

-	os.Args = []string{"codex-wrapper", "--cleanup"}
+	os.Args = []string{"codeagent-wrapper", "--cleanup"}

 	calls := 0
 	cleanupLogsFn = func() (CleanupStats, error) {
@@ -4453,7 +4409,7 @@ func TestRun_LoggerRemovedOnSignal(t *testing.T) {
 	defer signal.Reset(syscall.SIGINT, syscall.SIGTERM)

 	// Set shorter delays for faster test
-	forceKillDelay.Store(1)
+	_ = executor.SetForceKillDelay(1)

 	tempDir := t.TempDir()
 	t.Setenv("TMPDIR", tempDir)
@@ -4565,7 +4521,7 @@ func TestRun_CleanupFailureDoesNotBlock(t *testing.T) {
 	codexCommand = createFakeCodexScript(t, "tid-cleanup", "ok")
 	stdinReader = strings.NewReader("")
 	isTerminalFn = func() bool { return true }
-	os.Args = []string{"codex-wrapper", "task"}
+	os.Args = []string{"codeagent-wrapper", "task"}

 	if exit := run(); exit != 0 {
 		t.Fatalf("exit = %d, want 0", exit)
@@ -4704,73 +4660,6 @@ func TestBackendDiscardInvalidJSONBuffer(t *testing.T) {
 	})
 }

-func TestRunForwardSignals(t *testing.T) {
-	defer resetTestHooks()
-
-	if runtime.GOOS == "windows" {
-		t.Skip("sleep command not available on Windows")
-	}
-
-	execCmd := exec.Command("sleep", "5")
-	if err := execCmd.Start(); err != nil {
-		t.Skipf("unable to start sleep command: %v", err)
-	}
-	defer func() {
-		_ = execCmd.Process.Kill()
-		execCmd.Wait()
-	}()
-
-	ctx, cancel := context.WithCancel(context.Background())
-	defer cancel()
-
-	forceKillDelay.Store(0)
-	defer forceKillDelay.Store(5)
-
-	ready := make(chan struct{})
-	var captured chan<- os.Signal
-	signalNotifyFn = func(ch chan<- os.Signal, sig ...os.Signal) {
-		captured = ch
-		close(ready)
-	}
-	signalStopFn = func(ch chan<- os.Signal) {}
-	defer func() {
-		signalNotifyFn = signal.Notify
-		signalStopFn = signal.Stop
-	}()
-
-	var mu sync.Mutex
-	var logs []string
-	cmd := &realCmd{cmd: execCmd}
-	forwardSignals(ctx, cmd, func(msg string) {
-		mu.Lock()
-		defer mu.Unlock()
-		logs = append(logs, msg)
-	})
-
-	select {
-	case <-ready:
-	case <-time.After(500 * time.Millisecond):
-		t.Fatalf("signalNotifyFn not invoked")
-	}
-
-	captured <- syscall.SIGINT
-
-	done := make(chan error, 1)
-	go func() { done <- cmd.Wait() }()
-
-	select {
-	case <-done:
-	case <-time.After(2 * time.Second):
-		t.Fatalf("process did not exit after forwarded signal")
-	}
-
-	mu.Lock()
-	defer mu.Unlock()
-	if len(logs) == 0 {
-		t.Fatalf("expected log entry for forwarded signal")
-	}
-}
-
 // Backend-focused coverage suite to ensure run() paths stay exercised under the focused pattern.
 func TestBackendRunCoverage(t *testing.T) {
 	suite := []struct {
@@ -4810,7 +4699,7 @@ func TestParallelLogPathInSerialMode(t *testing.T) {
 	tempDir := t.TempDir()
 	t.Setenv("TMPDIR", tempDir)

-	os.Args = []string{"codex-wrapper", "do-stuff"}
+	os.Args = []string{"codeagent-wrapper", "do-stuff"}
 	stdinReader = strings.NewReader("")
 	isTerminalFn = func() bool { return true }
 	codexCommand = "echo"
@@ -4827,126 +4716,13 @@ func TestParallelLogPathInSerialMode(t *testing.T) {
 	if exitCode != 0 {
 		t.Fatalf("run() exit = %d, want 0", exitCode)
 	}
-	expectedLog := filepath.Join(tempDir, fmt.Sprintf("codex-wrapper-%d.log", os.Getpid()))
+	expectedLog := filepath.Join(tempDir, fmt.Sprintf("codeagent-wrapper-%d.log", os.Getpid()))
 	wantLine := fmt.Sprintf("Log: %s", expectedLog)
 	if !strings.Contains(stderr, wantLine) {
 		t.Fatalf("stderr missing %q, got: %q", wantLine, stderr)
 	}
 }

-func TestRealProcessNilSafety(t *testing.T) {
-	var proc *realProcess
-	if pid := proc.Pid(); pid != 0 {
-		t.Fatalf("Pid() = %d, want 0", pid)
-	}
-	if err := proc.Kill(); err != nil {
-		t.Fatalf("Kill() error = %v", err)
-	}
-	if err := proc.Signal(syscall.SIGTERM); err != nil {
-		t.Fatalf("Signal() error = %v", err)
-	}
-}
-
-func TestRealProcessKill(t *testing.T) {
-	if runtime.GOOS == "windows" {
-		t.Skip("sleep command not available on Windows")
-	}
-
-	cmd := exec.Command("sleep", "5")
-	if err := cmd.Start(); err != nil {
-		t.Skipf("unable to start sleep command: %v", err)
-	}
-	waited := false
-	defer func() {
-		if waited {
-			return
-		}
-		if cmd.Process != nil {
-			_ = cmd.Process.Kill()
-			cmd.Wait()
-		}
-	}()
-
-	proc := &realProcess{proc: cmd.Process}
-	if proc.Pid() == 0 {
-		t.Fatalf("Pid() returned 0 for active process")
-	}
-	if err := proc.Kill(); err != nil {
-		t.Fatalf("Kill() error = %v", err)
-	}
-	waitErr := cmd.Wait()
-	waited = true
-	if waitErr == nil {
-		t.Fatalf("Kill() should lead to non-nil wait error")
-	}
-}
-
-func TestRealProcessSignal(t *testing.T) {
-	if runtime.GOOS == "windows" {
-		t.Skip("sleep command not available on Windows")
-	}
-
-	cmd := exec.Command("sleep", "5")
-	if err := cmd.Start(); err != nil {
-		t.Skipf("unable to start sleep command: %v", err)
-	}
-	waited := false
-	defer func() {
-		if waited {
-			return
-		}
-		if cmd.Process != nil {
-			_ = cmd.Process.Kill()
-			cmd.Wait()
-		}
-	}()
-
-	proc := &realProcess{proc: cmd.Process}
-	if err := proc.Signal(syscall.SIGTERM); err != nil {
-		t.Fatalf("Signal() error = %v", err)
-	}
-	waitErr := cmd.Wait()
-	waited = true
-	if waitErr == nil {
-		t.Fatalf("Signal() should lead to non-nil wait error")
-	}
-}
-
-func TestRealCmdProcess(t *testing.T) {
-	rc := &realCmd{}
-	if rc.Process() != nil {
-		t.Fatalf("Process() should return nil when realCmd has no command")
-	}
-	rc = &realCmd{cmd: &exec.Cmd{}}
-	if rc.Process() != nil {
-		t.Fatalf("Process() should return nil when exec.Cmd has no process")
-	}
-
-	if runtime.GOOS == "windows" {
-		return
-	}
-
-	cmd := exec.Command("sleep", "5")
-	if err := cmd.Start(); err != nil {
-		t.Skipf("unable to start sleep command: %v", err)
-	}
-	defer func() {
-		if cmd.Process != nil {
-			_ = cmd.Process.Kill()
-			cmd.Wait()
-		}
-	}()
-
-	rc = &realCmd{cmd: cmd}
-	handle := rc.Process()
-	if handle == nil {
-		t.Fatalf("expected non-nil process handle")
-	}
-	if pid := handle.Pid(); pid == 0 {
-		t.Fatalf("process handle returned pid=0")
-	}
-}
-
 func TestRun_CLI_Success(t *testing.T) {
 	defer resetTestHooks()
 	os.Args = []string{"codeagent-wrapper", "do-things"}
@@ -4988,7 +4764,7 @@ func TestResolveMaxParallelWorkers(t *testing.T) {
 		t.Run(tt.name, func(t *testing.T) {
 			t.Setenv("CODEAGENT_MAX_PARALLEL_WORKERS", tt.envValue)

-			got := resolveMaxParallelWorkers()
+			got := config.ResolveMaxParallelWorkers()
 			if got != tt.want {
 				t.Errorf("resolveMaxParallelWorkers() = %d, want %d", got, tt.want)
 			}
--- a/codeagent-wrapper/internal/app/parallel_config.go
+++ b/codeagent-wrapper/internal/app/parallel_config.go
@@ -0,0 +1,9 @@
+package wrapper
+
+import (
+	executor "codeagent-wrapper/internal/executor"
+)
+
+func parseParallelConfig(data []byte) (*ParallelConfig, error) {
+	return executor.ParseParallelConfig(data)
+}
--- a/codeagent-wrapper/internal/app/parser.go
+++ b/codeagent-wrapper/internal/app/parser.go
@@ -0,0 +1,34 @@
+package wrapper
+
+import (
+	"bufio"
+	"io"
+
+	parser "codeagent-wrapper/internal/parser"
+
+	"github.com/goccy/go-json"
+)
+
+func parseJSONStream(r io.Reader) (message, threadID string) {
+	return parseJSONStreamWithLog(r, logWarn, logInfo)
+}
+
+func parseJSONStreamWithWarn(r io.Reader, warnFn func(string)) (message, threadID string) {
+	return parseJSONStreamWithLog(r, warnFn, logInfo)
+}
+
+func parseJSONStreamWithLog(r io.Reader, warnFn func(string), infoFn func(string)) (message, threadID string) {
+	return parseJSONStreamInternal(r, warnFn, infoFn, nil, nil)
+}
+
+func parseJSONStreamInternal(r io.Reader, warnFn func(string), infoFn func(string), onMessage func(), onComplete func()) (message, threadID string) {
+	return parser.ParseJSONStreamInternal(r, warnFn, infoFn, onMessage, onComplete)
+}
+
+func hasKey(m map[string]json.RawMessage, key string) bool { return parser.HasKey(m, key) }
+
+func discardInvalidJSON(decoder *json.Decoder, reader *bufio.Reader) (*bufio.Reader, error) {
+	return parser.DiscardInvalidJSON(decoder, reader)
+}
+
+func normalizeText(text interface{}) string { return parser.NormalizeText(text) }
--- a/codeagent-wrapper/internal/app/task_types.go
+++ b/codeagent-wrapper/internal/app/task_types.go
@@ -0,0 +1,8 @@
+package wrapper
+
+import executor "codeagent-wrapper/internal/executor"
+
+// Type aliases to keep existing names in the wrapper package.
+type ParallelConfig = executor.ParallelConfig
+type TaskSpec = executor.TaskSpec
+type TaskResult = executor.TaskResult
--- a/codeagent-wrapper/internal/app/terminal_test.go
+++ b/codeagent-wrapper/internal/app/terminal_test.go
@@ -0,0 +1,30 @@
+package wrapper
+
+import (
+	"os"
+	"testing"
+)
+
+func TestDefaultIsTerminalCoverage(t *testing.T) {
+	oldStdin := os.Stdin
+	t.Cleanup(func() { os.Stdin = oldStdin })
+
+	f, err := os.CreateTemp(t.TempDir(), "stdin-*")
+	if err != nil {
+		t.Fatalf("os.CreateTemp() error = %v", err)
+	}
+	defer os.Remove(f.Name())
+
+	os.Stdin = f
+	if got := defaultIsTerminal(); got {
+		t.Fatalf("defaultIsTerminal() = %v, want false for regular file", got)
+	}
+
+	if err := f.Close(); err != nil {
+		t.Fatalf("Close() error = %v", err)
+	}
+	os.Stdin = f
+	if got := defaultIsTerminal(); !got {
+		t.Fatalf("defaultIsTerminal() = %v, want true when Stat fails", got)
+	}
+}
--- a/codeagent-wrapper/internal/app/utils.go
+++ b/codeagent-wrapper/internal/app/utils.go
@@ -1,4 +1,4 @@
-package main
+package wrapper

 import (
 	"bytes"
@@ -7,6 +7,8 @@ import (
 	"os"
 	"strconv"
 	"strings"
+
+	utils "codeagent-wrapper/internal/utils"
 )

 func resolveTimeout() int {
@@ -52,7 +54,7 @@ func shouldUseStdin(taskText string, piped bool) bool {
 	if len(taskText) > 800 {
 		return true
 	}
-	return strings.IndexAny(taskText, stdinSpecialChars) >= 0
+	return strings.ContainsAny(taskText, stdinSpecialChars)
 }

 func defaultIsTerminal() bool {
@@ -196,69 +198,21 @@ func (b *tailBuffer) String() string {
 }

 func truncate(s string, maxLen int) string {
-	if len(s) <= maxLen {
-		return s
-	}
-	if maxLen < 0 {
-		return ""
-	}
-	return s[:maxLen] + "..."
+	return utils.Truncate(s, maxLen)
 }

 // safeTruncate safely truncates string to maxLen, avoiding panic and UTF-8 corruption.
 func safeTruncate(s string, maxLen int) string {
-	if maxLen <= 0 || s == "" {
-		return ""
-	}
-
-	runes := []rune(s)
-	if len(runes) <= maxLen {
-		return s
-	}
-
-	if maxLen < 4 {
-		return string(runes[:1])
-	}
-
-	cutoff := maxLen - 3
-	if cutoff <= 0 {
-		return string(runes[:1])
-	}
-	if len(runes) <= cutoff {
-		return s
-	}
-	return string(runes[:cutoff]) + "..."
+	return utils.SafeTruncate(s, maxLen)
 }

 // sanitizeOutput removes ANSI escape sequences and control characters.
 func sanitizeOutput(s string) string {
-	var result strings.Builder
-	inEscape := false
-	for i := 0; i < len(s); i++ {
-		if s[i] == '\x1b' && i+1 < len(s) && s[i+1] == '[' {
-			inEscape = true
-			i++ // skip '['
-			continue
-		}
-		if inEscape {
-			if (s[i] >= 'A' && s[i] <= 'Z') || (s[i] >= 'a' && s[i] <= 'z') {
-				inEscape = false
-			}
-			continue
-		}
-		// Keep printable chars and common whitespace.
-		if s[i] >= 32 || s[i] == '\n' || s[i] == '\t' {
-			result.WriteByte(s[i])
-		}
-	}
-	return result.String()
+	return utils.SanitizeOutput(s)
 }

 func min(a, b int) int {
-	if a < b {
-		return a
-	}
-	return b
+	return utils.Min(a, b)
 }

 func hello() string {
@@ -381,7 +335,7 @@ func extractFilesChangedFromLines(lines []string) []string {
 		for _, prefix := range []string{"Modified:", "Created:", "Updated:", "Edited:", "Wrote:", "Changed:"} {
 			if strings.HasPrefix(line, prefix) {
 				file := strings.TrimSpace(strings.TrimPrefix(line, prefix))
-				file = strings.Trim(file, "`,\"'()[],:")
+				file = strings.Trim(file, "`\"'()[],:")
 				file = strings.TrimPrefix(file, "@")
 				if file != "" && !seen[file] {
 					files = append(files, file)
@@ -398,7 +352,7 @@ func extractFilesChangedFromLines(lines []string) []string {
 		// Pattern 2: Tokens that look like file paths (allow root files, strip @ prefix).
 		parts := strings.Fields(line)
 		for _, part := range parts {
-			part = strings.Trim(part, "`,\"'()[],:")
+			part = strings.Trim(part, "`\"'()[],:")
 			part = strings.TrimPrefix(part, "@")
 			for _, ext := range exts {
 				if strings.HasSuffix(part, ext) && !seen[part] {
@@ -567,116 +521,3 @@ func extractKeyOutputFromLines(lines []string, maxLen int) string {
 	clean := strings.TrimSpace(strings.Join(lines, "\n"))
 	return safeTruncate(clean, maxLen)
 }
-
-// extractCoverageGap extracts what's missing from coverage reports
-// Looks for uncovered lines, branches, or functions
-func extractCoverageGap(message string) string {
-	if message == "" {
-		return ""
-	}
-
-	lower := strings.ToLower(message)
-	lines := strings.Split(message, "\n")
-
-	// Look for uncovered/missing patterns
-	for _, line := range lines {
-		lineLower := strings.ToLower(line)
-		line = strings.TrimSpace(line)
-
-		// Common patterns for uncovered code
-		if strings.Contains(lineLower, "uncovered") ||
-			strings.Contains(lineLower, "not covered") ||
-			strings.Contains(lineLower, "missing coverage") ||
-			strings.Contains(lineLower, "lines not covered") {
-			if len(line) > 100 {
-				return line[:97] + "..."
-			}
-			return line
-		}
-
-		// Look for specific file:line patterns in coverage reports
-		if strings.Contains(lineLower, "branch") && strings.Contains(lineLower, "not taken") {
-			if len(line) > 100 {
-				return line[:97] + "..."
-			}
-			return line
-		}
-	}
-
-	// Look for function names that aren't covered
-	if strings.Contains(lower, "function") && strings.Contains(lower, "0%") {
-		for _, line := range lines {
-			if strings.Contains(strings.ToLower(line), "0%") && strings.Contains(line, "function") {
-				line = strings.TrimSpace(line)
-				if len(line) > 100 {
-					return line[:97] + "..."
-				}
-				return line
-			}
-		}
-	}
-
-	return ""
-}
-
-// extractErrorDetail extracts meaningful error context from task output
-// Returns the most relevant error information up to maxLen characters
-func extractErrorDetail(message string, maxLen int) string {
-	if message == "" || maxLen <= 0 {
-		return ""
-	}
-
-	lines := strings.Split(message, "\n")
-	var errorLines []string
-
-	// Look for error-related lines
-	for _, line := range lines {
-		line = strings.TrimSpace(line)
-		if line == "" {
-			continue
-		}
-
-		lower := strings.ToLower(line)
-
-		// Skip noise lines
-		if strings.HasPrefix(line, "at ") && strings.Contains(line, "(") {
-			// Stack trace line - only keep first one
-			if len(errorLines) > 0 && strings.HasPrefix(strings.ToLower(errorLines[len(errorLines)-1]), "at ") {
-				continue
-			}
-		}
-
-		// Prioritize error/fail lines
-		if strings.Contains(lower, "error") ||
-			strings.Contains(lower, "fail") ||
-			strings.Contains(lower, "exception") ||
-			strings.Contains(lower, "assert") ||
-			strings.Contains(lower, "expected") ||
-			strings.Contains(lower, "timeout") ||
-			strings.Contains(lower, "not found") ||
-			strings.Contains(lower, "cannot") ||
-			strings.Contains(lower, "undefined") ||
-			strings.HasPrefix(line, "FAIL") ||
-			strings.HasPrefix(line, "●") {
-			errorLines = append(errorLines, line)
-		}
-	}
-
-	if len(errorLines) == 0 {
-		// No specific error lines found, take last few lines
-		start := len(lines) - 5
-		if start < 0 {
-			start = 0
-		}
-		for _, line := range lines[start:] {
-			line = strings.TrimSpace(line)
-			if line != "" {
-				errorLines = append(errorLines, line)
-			}
-		}
-	}
-
-	// Join and truncate
-	result := strings.Join(errorLines, " | ")
-	return safeTruncate(result, maxLen)
-}
--- a/codeagent-wrapper/internal/app/utils_test.go
+++ b/codeagent-wrapper/internal/app/utils_test.go
@@ -1,4 +1,4 @@
-package main
+package wrapper

 import (
 	"fmt"
--- a/codeagent-wrapper/internal/app/wrapper_name.go
+++ b/codeagent-wrapper/internal/app/wrapper_name.go
@@ -0,0 +1,9 @@
+package wrapper
+
+import ilogger "codeagent-wrapper/internal/logger"
+
+const wrapperName = ilogger.WrapperName
+
+func currentWrapperName() string { return ilogger.CurrentWrapperName() }
+
+func primaryLogPrefix() string { return ilogger.PrimaryLogPrefix() }
--- a/codeagent-wrapper/internal/backend/backend.go
+++ b/codeagent-wrapper/internal/backend/backend.go
@@ -0,0 +1,33 @@
+package backend
+
+import config "codeagent-wrapper/internal/config"
+
+// Backend defines the contract for invoking different AI CLI backends.
+// Each backend is responsible for supplying the executable command and
+// building the argument list based on the wrapper config.
+type Backend interface {
+	Name() string
+	BuildArgs(cfg *config.Config, targetArg string) []string
+	Command() string
+	Env(baseURL, apiKey string) map[string]string
+}
+
+var (
+	logWarnFn  = func(string) {}
+	logErrorFn = func(string) {}
+)
+
+// SetLogFuncs configures optional logging hooks used by some backends.
+// Callers can safely pass nil to disable the hook.
+func SetLogFuncs(warnFn, errorFn func(string)) {
+	if warnFn != nil {
+		logWarnFn = warnFn
+	} else {
+		logWarnFn = func(string) {}
+	}
+	if errorFn != nil {
+		logErrorFn = errorFn
+	} else {
+		logErrorFn = func(string) {}
+	}
+}
--- a/codeagent-wrapper/internal/backend/backend_test.go
+++ b/codeagent-wrapper/internal/backend/backend_test.go
@@ -1,4 +1,4 @@
-package main
+package backend

 import (
 	"bytes"
@@ -6,6 +6,8 @@ import (
 	"path/filepath"
 	"reflect"
 	"testing"
+
+	config "codeagent-wrapper/internal/config"
 )

 func TestClaudeBuildArgs_ModesAndPermissions(t *testing.T) {
@@ -13,7 +15,7 @@ func TestClaudeBuildArgs_ModesAndPermissions(t *testing.T) {

 	t.Run("new mode omits skip-permissions when env disabled", func(t *testing.T) {
 		t.Setenv("CODEAGENT_SKIP_PERMISSIONS", "false")
-		cfg := &Config{Mode: "new", WorkDir: "/repo"}
+		cfg := &config.Config{Mode: "new", WorkDir: "/repo"}
 		got := backend.BuildArgs(cfg, "todo")
 		want := []string{"-p", "--setting-sources", "", "--output-format", "stream-json", "--verbose", "todo"}
 		if !reflect.DeepEqual(got, want) {
@@ -22,7 +24,7 @@ func TestClaudeBuildArgs_ModesAndPermissions(t *testing.T) {
 	})

 	t.Run("new mode includes skip-permissions by default", func(t *testing.T) {
-		cfg := &Config{Mode: "new", SkipPermissions: false}
+		cfg := &config.Config{Mode: "new", SkipPermissions: false}
 		got := backend.BuildArgs(cfg, "-")
 		want := []string{"-p", "--dangerously-skip-permissions", "--setting-sources", "", "--output-format", "stream-json", "--verbose", "-"}
 		if !reflect.DeepEqual(got, want) {
@@ -32,7 +34,7 @@ func TestClaudeBuildArgs_ModesAndPermissions(t *testing.T) {

 	t.Run("resume mode includes session id", func(t *testing.T) {
 		t.Setenv("CODEAGENT_SKIP_PERMISSIONS", "false")
-		cfg := &Config{Mode: "resume", SessionID: "sid-123", WorkDir: "/ignored"}
+		cfg := &config.Config{Mode: "resume", SessionID: "sid-123", WorkDir: "/ignored"}
 		got := backend.BuildArgs(cfg, "resume-task")
 		want := []string{"-p", "--setting-sources", "", "-r", "sid-123", "--output-format", "stream-json", "--verbose", "resume-task"}
 		if !reflect.DeepEqual(got, want) {
@@ -42,7 +44,7 @@ func TestClaudeBuildArgs_ModesAndPermissions(t *testing.T) {

 	t.Run("resume mode without session still returns base flags", func(t *testing.T) {
 		t.Setenv("CODEAGENT_SKIP_PERMISSIONS", "false")
-		cfg := &Config{Mode: "resume", WorkDir: "/ignored"}
+		cfg := &config.Config{Mode: "resume", WorkDir: "/ignored"}
 		got := backend.BuildArgs(cfg, "follow-up")
 		want := []string{"-p", "--setting-sources", "", "--output-format", "stream-json", "--verbose", "follow-up"}
 		if !reflect.DeepEqual(got, want) {
@@ -51,7 +53,7 @@ func TestClaudeBuildArgs_ModesAndPermissions(t *testing.T) {
 	})

 	t.Run("resume mode can opt-in skip permissions", func(t *testing.T) {
-		cfg := &Config{Mode: "resume", SessionID: "sid-123", SkipPermissions: true}
+		cfg := &config.Config{Mode: "resume", SessionID: "sid-123", SkipPermissions: true}
 		got := backend.BuildArgs(cfg, "resume-task")
 		want := []string{"-p", "--dangerously-skip-permissions", "--setting-sources", "", "-r", "sid-123", "--output-format", "stream-json", "--verbose", "resume-task"}
 		if !reflect.DeepEqual(got, want) {
@@ -70,7 +72,7 @@ func TestBackendBuildArgs_Model(t *testing.T) {
 	t.Run("claude includes --model when set", func(t *testing.T) {
 		t.Setenv("CODEAGENT_SKIP_PERMISSIONS", "false")
 		backend := ClaudeBackend{}
-		cfg := &Config{Mode: "new", Model: "opus"}
+		cfg := &config.Config{Mode: "new", Model: "opus"}
 		got := backend.BuildArgs(cfg, "todo")
 		want := []string{"-p", "--setting-sources", "", "--model", "opus", "--output-format", "stream-json", "--verbose", "todo"}
 		if !reflect.DeepEqual(got, want) {
@@ -80,7 +82,7 @@ func TestBackendBuildArgs_Model(t *testing.T) {

 	t.Run("gemini includes -m when set", func(t *testing.T) {
 		backend := GeminiBackend{}
-		cfg := &Config{Mode: "new", Model: "gemini-3-pro-preview"}
+		cfg := &config.Config{Mode: "new", Model: "gemini-3-pro-preview"}
 		got := backend.BuildArgs(cfg, "task")
 		want := []string{"-o", "stream-json", "-y", "-m", "gemini-3-pro-preview", "task"}
 		if !reflect.DeepEqual(got, want) {
@@ -93,7 +95,7 @@ func TestBackendBuildArgs_Model(t *testing.T) {
 		t.Setenv(key, "false")

 		backend := CodexBackend{}
-		cfg := &Config{Mode: "new", WorkDir: "/tmp", Model: "o3"}
+		cfg := &config.Config{Mode: "new", WorkDir: "/tmp", Model: "o3"}
 		got := backend.BuildArgs(cfg, "task")
 		want := []string{"e", "--model", "o3", "--skip-git-repo-check", "-C", "/tmp", "--json", "task"}
 		if !reflect.DeepEqual(got, want) {
@@ -105,7 +107,7 @@ func TestBackendBuildArgs_Model(t *testing.T) {
 func TestClaudeBuildArgs_GeminiAndCodexModes(t *testing.T) {
 	t.Run("gemini new mode defaults workdir", func(t *testing.T) {
 		backend := GeminiBackend{}
-		cfg := &Config{Mode: "new", WorkDir: "/workspace"}
+		cfg := &config.Config{Mode: "new", WorkDir: "/workspace"}
 		got := backend.BuildArgs(cfg, "task")
 		want := []string{"-o", "stream-json", "-y", "task"}
 		if !reflect.DeepEqual(got, want) {
@@ -115,7 +117,7 @@ func TestClaudeBuildArgs_GeminiAndCodexModes(t *testing.T) {

 	t.Run("gemini resume mode uses session id", func(t *testing.T) {
 		backend := GeminiBackend{}
-		cfg := &Config{Mode: "resume", SessionID: "sid-999"}
+		cfg := &config.Config{Mode: "resume", SessionID: "sid-999"}
 		got := backend.BuildArgs(cfg, "resume")
 		want := []string{"-o", "stream-json", "-y", "-r", "sid-999", "resume"}
 		if !reflect.DeepEqual(got, want) {
@@ -125,7 +127,7 @@ func TestClaudeBuildArgs_GeminiAndCodexModes(t *testing.T) {

 	t.Run("gemini resume mode without session omits identifier", func(t *testing.T) {
 		backend := GeminiBackend{}
-		cfg := &Config{Mode: "resume"}
+		cfg := &config.Config{Mode: "resume"}
 		got := backend.BuildArgs(cfg, "resume")
 		want := []string{"-o", "stream-json", "-y", "resume"}
 		if !reflect.DeepEqual(got, want) {
@@ -142,7 +144,7 @@ func TestClaudeBuildArgs_GeminiAndCodexModes(t *testing.T) {

 	t.Run("gemini stdin mode uses -p flag", func(t *testing.T) {
 		backend := GeminiBackend{}
-		cfg := &Config{Mode: "new"}
+		cfg := &config.Config{Mode: "new"}
 		got := backend.BuildArgs(cfg, "-")
 		want := []string{"-o", "stream-json", "-y", "-p", "-"}
 		if !reflect.DeepEqual(got, want) {
@@ -155,7 +157,7 @@ func TestClaudeBuildArgs_GeminiAndCodexModes(t *testing.T) {
 		t.Setenv(key, "false")

 		backend := CodexBackend{}
-		cfg := &Config{Mode: "new", WorkDir: "/tmp"}
+		cfg := &config.Config{Mode: "new", WorkDir: "/tmp"}
 		got := backend.BuildArgs(cfg, "task")
 		want := []string{"e", "--skip-git-repo-check", "-C", "/tmp", "--json", "task"}
 		if !reflect.DeepEqual(got, want) {
@@ -168,7 +170,7 @@ func TestClaudeBuildArgs_GeminiAndCodexModes(t *testing.T) {
 		t.Setenv(key, "true")

 		backend := CodexBackend{}
-		cfg := &Config{Mode: "new", WorkDir: "/tmp"}
+		cfg := &config.Config{Mode: "new", WorkDir: "/tmp"}
 		got := backend.BuildArgs(cfg, "task")
 		want := []string{"e", "--dangerously-bypass-approvals-and-sandbox", "--skip-git-repo-check", "-C", "/tmp", "--json", "task"}
 		if !reflect.DeepEqual(got, want) {
@@ -204,7 +206,7 @@ func TestLoadMinimalEnvSettings(t *testing.T) {
 	t.Setenv("USERPROFILE", home)

 	t.Run("missing file returns empty", func(t *testing.T) {
-		if got := loadMinimalEnvSettings(); len(got) != 0 {
+		if got := LoadMinimalEnvSettings(); len(got) != 0 {
 			t.Fatalf("got %v, want empty", got)
 		}
 	})
@@ -220,7 +222,7 @@ func TestLoadMinimalEnvSettings(t *testing.T) {
 			t.Fatalf("WriteFile: %v", err)
 		}

-		got := loadMinimalEnvSettings()
+		got := LoadMinimalEnvSettings()
 		if got["ANTHROPIC_API_KEY"] != "secret" || got["FOO"] != "bar" {
 			t.Fatalf("got %v, want keys present", got)
 		}
@@ -234,7 +236,7 @@ func TestLoadMinimalEnvSettings(t *testing.T) {
 			t.Fatalf("WriteFile: %v", err)
 		}

-		got := loadMinimalEnvSettings()
+		got := LoadMinimalEnvSettings()
 		if got["GOOD"] != "ok" {
 			t.Fatalf("got %v, want GOOD=ok", got)
 		}
@@ -249,12 +251,72 @@ func TestLoadMinimalEnvSettings(t *testing.T) {
 	t.Run("oversized file returns empty", func(t *testing.T) {
 		dir := filepath.Join(home, ".claude")
 		path := filepath.Join(dir, "settings.json")
-		data := bytes.Repeat([]byte("a"), maxClaudeSettingsBytes+1)
+		data := bytes.Repeat([]byte("a"), MaxClaudeSettingsBytes+1)
 		if err := os.WriteFile(path, data, 0o600); err != nil {
 			t.Fatalf("WriteFile: %v", err)
 		}
-		if got := loadMinimalEnvSettings(); len(got) != 0 {
+		if got := LoadMinimalEnvSettings(); len(got) != 0 {
 			t.Fatalf("got %v, want empty", got)
 		}
 	})
 }
+
+func TestOpencodeBackend_BuildArgs(t *testing.T) {
+	backend := OpencodeBackend{}
+
+	t.Run("basic", func(t *testing.T) {
+		cfg := &config.Config{Mode: "new"}
+		got := backend.BuildArgs(cfg, "hello")
+		want := []string{"run", "--format", "json", "hello"}
+		if !reflect.DeepEqual(got, want) {
+			t.Errorf("got %v, want %v", got, want)
+		}
+	})
+
+	t.Run("with model", func(t *testing.T) {
+		cfg := &config.Config{Mode: "new", Model: "opencode/grok-code"}
+		got := backend.BuildArgs(cfg, "task")
+		want := []string{"run", "-m", "opencode/grok-code", "--format", "json", "task"}
+		if !reflect.DeepEqual(got, want) {
+			t.Errorf("got %v, want %v", got, want)
+		}
+	})
+
+	t.Run("resume mode", func(t *testing.T) {
+		cfg := &config.Config{Mode: "resume", SessionID: "ses_123", Model: "opencode/grok-code"}
+		got := backend.BuildArgs(cfg, "follow-up")
+		want := []string{"run", "-m", "opencode/grok-code", "-s", "ses_123", "--format", "json", "follow-up"}
+		if !reflect.DeepEqual(got, want) {
+			t.Errorf("got %v, want %v", got, want)
+		}
+	})
+
+	t.Run("resume without session", func(t *testing.T) {
+		cfg := &config.Config{Mode: "resume"}
+		got := backend.BuildArgs(cfg, "task")
+		want := []string{"run", "--format", "json", "task"}
+		if !reflect.DeepEqual(got, want) {
+			t.Errorf("got %v, want %v", got, want)
+		}
+	})
+
+	t.Run("stdin mode omits dash", func(t *testing.T) {
+		cfg := &config.Config{Mode: "new"}
+		got := backend.BuildArgs(cfg, "-")
+		want := []string{"run", "--format", "json"}
+		if !reflect.DeepEqual(got, want) {
+			t.Errorf("got %v, want %v", got, want)
+		}
+	})
+}
+
+func TestOpencodeBackend_Interface(t *testing.T) {
+	backend := OpencodeBackend{}
+
+	if backend.Name() != "opencode" {
+		t.Errorf("Name() = %q, want %q", backend.Name(), "opencode")
+	}
+	if backend.Command() != "opencode" {
+		t.Errorf("Command() = %q, want %q", backend.Command(), "opencode")
+	}
+}
--- a/codeagent-wrapper/internal/backend/claude.go
+++ b/codeagent-wrapper/internal/backend/claude.go
@@ -0,0 +1,139 @@
+package backend
+
+import (
+	"os"
+	"path/filepath"
+	"strings"
+
+	config "codeagent-wrapper/internal/config"
+
+	"github.com/goccy/go-json"
+)
+
+type ClaudeBackend struct{}
+
+func (ClaudeBackend) Name() string    { return "claude" }
+func (ClaudeBackend) Command() string { return "claude" }
+func (ClaudeBackend) Env(baseURL, apiKey string) map[string]string {
+	baseURL = strings.TrimSpace(baseURL)
+	apiKey = strings.TrimSpace(apiKey)
+	if baseURL == "" && apiKey == "" {
+		return nil
+	}
+	env := make(map[string]string, 2)
+	if baseURL != "" {
+		env["ANTHROPIC_BASE_URL"] = baseURL
+	}
+	if apiKey != "" {
+		env["ANTHROPIC_AUTH_TOKEN"] = apiKey
+	}
+	return env
+}
+func (ClaudeBackend) BuildArgs(cfg *config.Config, targetArg string) []string {
+	return buildClaudeArgs(cfg, targetArg)
+}
+
+const MaxClaudeSettingsBytes = 1 << 20 // 1MB
+
+type MinimalClaudeSettings struct {
+	Env   map[string]string
+	Model string
+}
+
+// LoadMinimalClaudeSettings 从 ~/.claude/settings.json 只提取安全的最小子集：
+// - env: 只接受字符串类型的值
+// - model: 只接受字符串类型的值
+// 文件缺失/解析失败/超限都返回空。
+func LoadMinimalClaudeSettings() MinimalClaudeSettings {
+	home, err := os.UserHomeDir()
+	if err != nil || home == "" {
+		return MinimalClaudeSettings{}
+	}
+
+	claudeDir := filepath.Clean(filepath.Join(home, ".claude"))
+	settingPath := filepath.Clean(filepath.Join(claudeDir, "settings.json"))
+	rel, err := filepath.Rel(claudeDir, settingPath)
+	if err != nil || rel == ".." || strings.HasPrefix(rel, ".."+string(os.PathSeparator)) {
+		return MinimalClaudeSettings{}
+	}
+
+	info, err := os.Stat(settingPath)
+	if err != nil || info.Size() > MaxClaudeSettingsBytes {
+		return MinimalClaudeSettings{}
+	}
+
+	data, err := os.ReadFile(settingPath) // #nosec G304 -- path is fixed under user home and validated to stay within claudeDir
+	if err != nil {
+		return MinimalClaudeSettings{}
+	}
+
+	var cfg struct {
+		Env   map[string]any `json:"env"`
+		Model any            `json:"model"`
+	}
+	if err := json.Unmarshal(data, &cfg); err != nil {
+		return MinimalClaudeSettings{}
+	}
+
+	out := MinimalClaudeSettings{}
+
+	if model, ok := cfg.Model.(string); ok {
+		out.Model = strings.TrimSpace(model)
+	}
+
+	if len(cfg.Env) == 0 {
+		return out
+	}
+
+	env := make(map[string]string, len(cfg.Env))
+	for k, v := range cfg.Env {
+		s, ok := v.(string)
+		if !ok {
+			continue
+		}
+		env[k] = s
+	}
+	if len(env) == 0 {
+		return out
+	}
+	out.Env = env
+	return out
+}
+
+func LoadMinimalEnvSettings() map[string]string {
+	settings := LoadMinimalClaudeSettings()
+	if len(settings.Env) == 0 {
+		return nil
+	}
+	return settings.Env
+}
+
+func buildClaudeArgs(cfg *config.Config, targetArg string) []string {
+	if cfg == nil {
+		return nil
+	}
+	args := []string{"-p"}
+	// Default to skip permissions unless CODEAGENT_SKIP_PERMISSIONS=false
+	if cfg.SkipPermissions || cfg.Yolo || config.EnvFlagDefaultTrue("CODEAGENT_SKIP_PERMISSIONS") {
+		args = append(args, "--dangerously-skip-permissions")
+	}
+
+	// Prevent infinite recursion: disable all setting sources (user, project, local)
+	// This ensures a clean execution environment without CLAUDE.md or skills that would trigger codeagent
+	args = append(args, "--setting-sources", "")
+
+	if model := strings.TrimSpace(cfg.Model); model != "" {
+		args = append(args, "--model", model)
+	}
+
+	if cfg.Mode == "resume" {
+		if cfg.SessionID != "" {
+			// Claude CLI uses -r <session_id> for resume.
+			args = append(args, "-r", cfg.SessionID)
+		}
+	}
+
+	args = append(args, "--output-format", "stream-json", "--verbose", targetArg)
+
+	return args
+}
--- a/codeagent-wrapper/internal/backend/codex.go
+++ b/codeagent-wrapper/internal/backend/codex.go
@@ -0,0 +1,79 @@
+package backend
+
+import (
+	"strings"
+
+	config "codeagent-wrapper/internal/config"
+)
+
+type CodexBackend struct{}
+
+func (CodexBackend) Name() string    { return "codex" }
+func (CodexBackend) Command() string { return "codex" }
+func (CodexBackend) Env(baseURL, apiKey string) map[string]string {
+	baseURL = strings.TrimSpace(baseURL)
+	apiKey = strings.TrimSpace(apiKey)
+	if baseURL == "" && apiKey == "" {
+		return nil
+	}
+	env := make(map[string]string, 2)
+	if baseURL != "" {
+		env["OPENAI_BASE_URL"] = baseURL
+	}
+	if apiKey != "" {
+		env["OPENAI_API_KEY"] = apiKey
+	}
+	return env
+}
+func (CodexBackend) BuildArgs(cfg *config.Config, targetArg string) []string {
+	return BuildCodexArgs(cfg, targetArg)
+}
+
+func BuildCodexArgs(cfg *config.Config, targetArg string) []string {
+	if cfg == nil {
+		panic("buildCodexArgs: nil config")
+	}
+
+	var resumeSessionID string
+	isResume := cfg.Mode == "resume"
+	if isResume {
+		resumeSessionID = strings.TrimSpace(cfg.SessionID)
+		if resumeSessionID == "" {
+			logErrorFn("invalid config: resume mode requires non-empty session_id")
+			isResume = false
+		}
+	}
+
+	args := []string{"e"}
+
+	// Default to bypass sandbox unless CODEX_BYPASS_SANDBOX=false
+	if cfg.Yolo || config.EnvFlagDefaultTrue("CODEX_BYPASS_SANDBOX") {
+		logWarnFn("YOLO mode or CODEX_BYPASS_SANDBOX enabled: running without approval/sandbox protection")
+		args = append(args, "--dangerously-bypass-approvals-and-sandbox")
+	}
+
+	if model := strings.TrimSpace(cfg.Model); model != "" {
+		args = append(args, "--model", model)
+	}
+
+	if reasoningEffort := strings.TrimSpace(cfg.ReasoningEffort); reasoningEffort != "" {
+		args = append(args, "-c", "model_reasoning_effort="+reasoningEffort)
+	}
+
+	args = append(args, "--skip-git-repo-check")
+
+	if isResume {
+		return append(args,
+			"--json",
+			"resume",
+			resumeSessionID,
+			targetArg,
+		)
+	}
+
+	return append(args,
+		"-C", cfg.WorkDir,
+		"--json",
+		targetArg,
+	)
+}
--- a/codeagent-wrapper/internal/backend/gemini.go
+++ b/codeagent-wrapper/internal/backend/gemini.go
@@ -0,0 +1,110 @@
+package backend
+
+import (
+	"os"
+	"path/filepath"
+	"strings"
+
+	config "codeagent-wrapper/internal/config"
+)
+
+type GeminiBackend struct{}
+
+func (GeminiBackend) Name() string    { return "gemini" }
+func (GeminiBackend) Command() string { return "gemini" }
+func (GeminiBackend) Env(baseURL, apiKey string) map[string]string {
+	baseURL = strings.TrimSpace(baseURL)
+	apiKey = strings.TrimSpace(apiKey)
+	if baseURL == "" && apiKey == "" {
+		return nil
+	}
+	env := make(map[string]string, 2)
+	if baseURL != "" {
+		env["GOOGLE_GEMINI_BASE_URL"] = baseURL
+	}
+	if apiKey != "" {
+		env["GEMINI_API_KEY"] = apiKey
+	}
+	return env
+}
+func (GeminiBackend) BuildArgs(cfg *config.Config, targetArg string) []string {
+	return buildGeminiArgs(cfg, targetArg)
+}
+
+// LoadGeminiEnv loads environment variables from ~/.gemini/.env
+// Supports GEMINI_API_KEY, GEMINI_MODEL, GOOGLE_GEMINI_BASE_URL
+// Also sets GEMINI_API_KEY_AUTH_MECHANISM=bearer for third-party API compatibility
+func LoadGeminiEnv() map[string]string {
+	home, err := os.UserHomeDir()
+	if err != nil || home == "" {
+		return nil
+	}
+
+	envDir := filepath.Clean(filepath.Join(home, ".gemini"))
+	envPath := filepath.Clean(filepath.Join(envDir, ".env"))
+	rel, err := filepath.Rel(envDir, envPath)
+	if err != nil || rel == ".." || strings.HasPrefix(rel, ".."+string(os.PathSeparator)) {
+		return nil
+	}
+
+	data, err := os.ReadFile(envPath) // #nosec G304 -- path is fixed under user home and validated to stay within envDir
+	if err != nil {
+		return nil
+	}
+
+	env := make(map[string]string)
+	for _, line := range strings.Split(string(data), "\n") {
+		line = strings.TrimSpace(line)
+		if line == "" || strings.HasPrefix(line, "#") {
+			continue
+		}
+		idx := strings.IndexByte(line, '=')
+		if idx <= 0 {
+			continue
+		}
+		key := strings.TrimSpace(line[:idx])
+		value := strings.TrimSpace(line[idx+1:])
+		if key != "" && value != "" {
+			env[key] = value
+		}
+	}
+
+	// Set bearer auth mechanism for third-party API compatibility
+	if _, ok := env["GEMINI_API_KEY"]; ok {
+		if _, hasAuth := env["GEMINI_API_KEY_AUTH_MECHANISM"]; !hasAuth {
+			env["GEMINI_API_KEY_AUTH_MECHANISM"] = "bearer"
+		}
+	}
+
+	if len(env) == 0 {
+		return nil
+	}
+	return env
+}
+
+func buildGeminiArgs(cfg *config.Config, targetArg string) []string {
+	if cfg == nil {
+		return nil
+	}
+	args := []string{"-o", "stream-json", "-y"}
+
+	if model := strings.TrimSpace(cfg.Model); model != "" {
+		args = append(args, "-m", model)
+	}
+
+	if cfg.Mode == "resume" {
+		if cfg.SessionID != "" {
+			args = append(args, "-r", cfg.SessionID)
+		}
+	}
+
+	// Use positional argument instead of deprecated -p flag.
+	// For stdin mode ("-"), use -p to read from stdin.
+	if targetArg == "-" {
+		args = append(args, "-p", targetArg)
+	} else {
+		args = append(args, targetArg)
+	}
+
+	return args
+}
--- a/codeagent-wrapper/internal/backend/opencode.go
+++ b/codeagent-wrapper/internal/backend/opencode.go
@@ -0,0 +1,29 @@
+package backend
+
+import (
+	"strings"
+
+	config "codeagent-wrapper/internal/config"
+)
+
+type OpencodeBackend struct{}
+
+func (OpencodeBackend) Name() string                                 { return "opencode" }
+func (OpencodeBackend) Command() string                              { return "opencode" }
+func (OpencodeBackend) Env(baseURL, apiKey string) map[string]string { return nil }
+func (OpencodeBackend) BuildArgs(cfg *config.Config, targetArg string) []string {
+	args := []string{"run"}
+	if cfg != nil {
+		if model := strings.TrimSpace(cfg.Model); model != "" {
+			args = append(args, "-m", model)
+		}
+		if cfg.Mode == "resume" && cfg.SessionID != "" {
+			args = append(args, "-s", cfg.SessionID)
+		}
+	}
+	args = append(args, "--format", "json")
+	if targetArg != "-" {
+		args = append(args, targetArg)
+	}
+	return args
+}
--- a/codeagent-wrapper/internal/backend/registry.go
+++ b/codeagent-wrapper/internal/backend/registry.go
@@ -0,0 +1,29 @@
+package backend
+
+import (
+	"fmt"
+	"strings"
+)
+
+var registry = map[string]Backend{
+	"codex":    CodexBackend{},
+	"claude":   ClaudeBackend{},
+	"gemini":   GeminiBackend{},
+	"opencode": OpencodeBackend{},
+}
+
+// Registry exposes the available backends. Intended for internal inspection/tests.
+func Registry() map[string]Backend {
+	return registry
+}
+
+func Select(name string) (Backend, error) {
+	key := strings.ToLower(strings.TrimSpace(name))
+	if key == "" {
+		key = "codex"
+	}
+	if backend, ok := registry[key]; ok {
+		return backend, nil
+	}
+	return nil, fmt.Errorf("unsupported backend %q", name)
+}
--- a/codeagent-wrapper/internal/config/agent.go
+++ b/codeagent-wrapper/internal/config/agent.go
@@ -0,0 +1,220 @@
+package config
+
+import (
+	"fmt"
+	"os"
+	"path/filepath"
+	"strings"
+	"sync"
+
+	ilogger "codeagent-wrapper/internal/logger"
+
+	"github.com/goccy/go-json"
+)
+
+type BackendConfig struct {
+	BaseURL string `json:"base_url,omitempty"`
+	APIKey  string `json:"api_key,omitempty"`
+}
+
+type AgentModelConfig struct {
+	Backend     string `json:"backend"`
+	Model       string `json:"model"`
+	PromptFile  string `json:"prompt_file,omitempty"`
+	Description string `json:"description,omitempty"`
+	Yolo        bool   `json:"yolo,omitempty"`
+	Reasoning   string `json:"reasoning,omitempty"`
+	BaseURL     string `json:"base_url,omitempty"`
+	APIKey      string `json:"api_key,omitempty"`
+}
+
+type ModelsConfig struct {
+	DefaultBackend string                      `json:"default_backend"`
+	DefaultModel   string                      `json:"default_model"`
+	Agents         map[string]AgentModelConfig `json:"agents"`
+	Backends       map[string]BackendConfig    `json:"backends,omitempty"`
+}
+
+var defaultModelsConfig = ModelsConfig{
+	DefaultBackend: "opencode",
+	DefaultModel:   "opencode/grok-code",
+	Agents: map[string]AgentModelConfig{
+		"oracle":                  {Backend: "claude", Model: "claude-opus-4-5-20251101", PromptFile: "~/.claude/skills/omo/references/oracle.md", Description: "Technical advisor"},
+		"librarian":               {Backend: "claude", Model: "claude-sonnet-4-5-20250929", PromptFile: "~/.claude/skills/omo/references/librarian.md", Description: "Researcher"},
+		"explore":                 {Backend: "opencode", Model: "opencode/grok-code", PromptFile: "~/.claude/skills/omo/references/explore.md", Description: "Code search"},
+		"develop":                 {Backend: "codex", Model: "", PromptFile: "~/.claude/skills/omo/references/develop.md", Description: "Code development"},
+		"frontend-ui-ux-engineer": {Backend: "gemini", Model: "", PromptFile: "~/.claude/skills/omo/references/frontend-ui-ux-engineer.md", Description: "Frontend engineer"},
+		"document-writer":         {Backend: "gemini", Model: "", PromptFile: "~/.claude/skills/omo/references/document-writer.md", Description: "Documentation"},
+	},
+}
+
+var (
+	modelsConfigOnce   sync.Once
+	modelsConfigCached *ModelsConfig
+)
+
+func modelsConfig() *ModelsConfig {
+	modelsConfigOnce.Do(func() {
+		modelsConfigCached = loadModelsConfig()
+	})
+	if modelsConfigCached == nil {
+		return &defaultModelsConfig
+	}
+	return modelsConfigCached
+}
+
+func loadModelsConfig() *ModelsConfig {
+	home, err := os.UserHomeDir()
+	if err != nil {
+		ilogger.LogWarn(fmt.Sprintf("Failed to resolve home directory for models config: %v; using defaults", err))
+		return &defaultModelsConfig
+	}
+
+	configDir := filepath.Clean(filepath.Join(home, ".codeagent"))
+	configPath := filepath.Clean(filepath.Join(configDir, "models.json"))
+	rel, err := filepath.Rel(configDir, configPath)
+	if err != nil || rel == ".." || strings.HasPrefix(rel, ".."+string(os.PathSeparator)) {
+		return &defaultModelsConfig
+	}
+
+	data, err := os.ReadFile(configPath) // #nosec G304 -- path is fixed under user home and validated to stay within configDir
+	if err != nil {
+		if !os.IsNotExist(err) {
+			ilogger.LogWarn(fmt.Sprintf("Failed to read models config %s: %v; using defaults", configPath, err))
+		}
+		return &defaultModelsConfig
+	}
+
+	var cfg ModelsConfig
+	if err := json.Unmarshal(data, &cfg); err != nil {
+		ilogger.LogWarn(fmt.Sprintf("Failed to parse models config %s: %v; using defaults", configPath, err))
+		return &defaultModelsConfig
+	}
+
+	cfg.DefaultBackend = strings.TrimSpace(cfg.DefaultBackend)
+	if cfg.DefaultBackend == "" {
+		cfg.DefaultBackend = defaultModelsConfig.DefaultBackend
+	}
+	cfg.DefaultModel = strings.TrimSpace(cfg.DefaultModel)
+	if cfg.DefaultModel == "" {
+		cfg.DefaultModel = defaultModelsConfig.DefaultModel
+	}
+
+	// Merge with defaults
+	for name, agent := range defaultModelsConfig.Agents {
+		if _, exists := cfg.Agents[name]; !exists {
+			if cfg.Agents == nil {
+				cfg.Agents = make(map[string]AgentModelConfig)
+			}
+			cfg.Agents[name] = agent
+		}
+	}
+
+	// Normalize backend keys so lookups can be case-insensitive.
+	if len(cfg.Backends) > 0 {
+		normalized := make(map[string]BackendConfig, len(cfg.Backends))
+		for k, v := range cfg.Backends {
+			key := strings.ToLower(strings.TrimSpace(k))
+			if key == "" {
+				continue
+			}
+			normalized[key] = v
+		}
+		if len(normalized) > 0 {
+			cfg.Backends = normalized
+		} else {
+			cfg.Backends = nil
+		}
+	}
+
+	return &cfg
+}
+
+func LoadDynamicAgent(name string) (AgentModelConfig, bool) {
+	if err := ValidateAgentName(name); err != nil {
+		return AgentModelConfig{}, false
+	}
+
+	home, err := os.UserHomeDir()
+	if err != nil || strings.TrimSpace(home) == "" {
+		return AgentModelConfig{}, false
+	}
+
+	absPath := filepath.Join(home, ".codeagent", "agents", name+".md")
+	info, err := os.Stat(absPath)
+	if err != nil || info.IsDir() {
+		return AgentModelConfig{}, false
+	}
+
+	return AgentModelConfig{PromptFile: "~/.codeagent/agents/" + name + ".md"}, true
+}
+
+func ResolveBackendConfig(backendName string) (baseURL, apiKey string) {
+	cfg := modelsConfig()
+	resolved := resolveBackendConfig(cfg, backendName)
+	return strings.TrimSpace(resolved.BaseURL), strings.TrimSpace(resolved.APIKey)
+}
+
+func resolveBackendConfig(cfg *ModelsConfig, backendName string) BackendConfig {
+	if cfg == nil || len(cfg.Backends) == 0 {
+		return BackendConfig{}
+	}
+	key := strings.ToLower(strings.TrimSpace(backendName))
+	if key == "" {
+		key = strings.ToLower(strings.TrimSpace(cfg.DefaultBackend))
+	}
+	if key == "" {
+		return BackendConfig{}
+	}
+	if backend, ok := cfg.Backends[key]; ok {
+		return backend
+	}
+	return BackendConfig{}
+}
+
+func resolveAgentConfig(agentName string) (backend, model, promptFile, reasoning, baseURL, apiKey string, yolo bool) {
+	cfg := modelsConfig()
+	if agent, ok := cfg.Agents[agentName]; ok {
+		backend = strings.TrimSpace(agent.Backend)
+		if backend == "" {
+			backend = cfg.DefaultBackend
+		}
+		backendCfg := resolveBackendConfig(cfg, backend)
+
+		baseURL = strings.TrimSpace(agent.BaseURL)
+		if baseURL == "" {
+			baseURL = strings.TrimSpace(backendCfg.BaseURL)
+		}
+		apiKey = strings.TrimSpace(agent.APIKey)
+		if apiKey == "" {
+			apiKey = strings.TrimSpace(backendCfg.APIKey)
+		}
+
+		return backend, strings.TrimSpace(agent.Model), agent.PromptFile, agent.Reasoning, baseURL, apiKey, agent.Yolo
+	}
+
+	if dynamic, ok := LoadDynamicAgent(agentName); ok {
+		backend = cfg.DefaultBackend
+		model = cfg.DefaultModel
+		backendCfg := resolveBackendConfig(cfg, backend)
+		baseURL = strings.TrimSpace(backendCfg.BaseURL)
+		apiKey = strings.TrimSpace(backendCfg.APIKey)
+		return backend, model, dynamic.PromptFile, "", baseURL, apiKey, false
+	}
+
+	backend = cfg.DefaultBackend
+	model = cfg.DefaultModel
+	backendCfg := resolveBackendConfig(cfg, backend)
+	baseURL = strings.TrimSpace(backendCfg.BaseURL)
+	apiKey = strings.TrimSpace(backendCfg.APIKey)
+	return backend, model, "", "", baseURL, apiKey, false
+}
+
+func ResolveAgentConfig(agentName string) (backend, model, promptFile, reasoning, baseURL, apiKey string, yolo bool) {
+	return resolveAgentConfig(agentName)
+}
+
+func ResetModelsConfigCacheForTest() {
+	modelsConfigCached = nil
+	modelsConfigOnce = sync.Once{}
+}
--- a/codeagent-wrapper/internal/config/agent_config_test.go
+++ b/codeagent-wrapper/internal/config/agent_config_test.go
@@ -1,9 +1,8 @@
-package main
+package config

 import (
 	"os"
 	"path/filepath"
-	"reflect"
 	"testing"
 )

@@ -11,6 +10,8 @@ func TestResolveAgentConfig_Defaults(t *testing.T) {
 	home := t.TempDir()
 	t.Setenv("HOME", home)
 	t.Setenv("USERPROFILE", home)
+	t.Cleanup(ResetModelsConfigCacheForTest)
+	ResetModelsConfigCacheForTest()

 	// Test that default agents resolve correctly without config file
 	tests := []struct {
@@ -19,16 +20,16 @@ func TestResolveAgentConfig_Defaults(t *testing.T) {
 		wantModel      string
 		wantPromptFile string
 	}{
-			{"oracle", "claude", "claude-sonnet-4-20250514", "~/.claude/skills/omo/references/oracle.md"},
-			{"librarian", "claude", "claude-sonnet-4-5-20250514", "~/.claude/skills/omo/references/librarian.md"},
-			{"explore", "opencode", "opencode/grok-code", "~/.claude/skills/omo/references/explore.md"},
-			{"frontend-ui-ux-engineer", "gemini", "", "~/.claude/skills/omo/references/frontend-ui-ux-engineer.md"},
-			{"document-writer", "gemini", "", "~/.claude/skills/omo/references/document-writer.md"},
-		}
+		{"oracle", "claude", "claude-opus-4-5-20251101", "~/.claude/skills/omo/references/oracle.md"},
+		{"librarian", "claude", "claude-sonnet-4-5-20250929", "~/.claude/skills/omo/references/librarian.md"},
+		{"explore", "opencode", "opencode/grok-code", "~/.claude/skills/omo/references/explore.md"},
+		{"frontend-ui-ux-engineer", "gemini", "", "~/.claude/skills/omo/references/frontend-ui-ux-engineer.md"},
+		{"document-writer", "gemini", "", "~/.claude/skills/omo/references/document-writer.md"},
+	}

 	for _, tt := range tests {
 		t.Run(tt.agent, func(t *testing.T) {
-			backend, model, promptFile, _, _ := resolveAgentConfig(tt.agent)
+			backend, model, promptFile, _, _, _, _ := resolveAgentConfig(tt.agent)
 			if backend != tt.wantBackend {
 				t.Errorf("backend = %q, want %q", backend, tt.wantBackend)
 			}
@@ -46,8 +47,10 @@ func TestResolveAgentConfig_UnknownAgent(t *testing.T) {
 	home := t.TempDir()
 	t.Setenv("HOME", home)
 	t.Setenv("USERPROFILE", home)
+	t.Cleanup(ResetModelsConfigCacheForTest)
+	ResetModelsConfigCacheForTest()

-	backend, model, promptFile, _, _ := resolveAgentConfig("unknown-agent")
+	backend, model, promptFile, _, _, _, _ := resolveAgentConfig("unknown-agent")
 	if backend != "opencode" {
 		t.Errorf("unknown agent backend = %q, want %q", backend, "opencode")
 	}
@@ -63,6 +66,8 @@ func TestLoadModelsConfig_NoFile(t *testing.T) {
 	home := "/nonexistent/path/that/does/not/exist"
 	t.Setenv("HOME", home)
 	t.Setenv("USERPROFILE", home)
+	t.Cleanup(ResetModelsConfigCacheForTest)
+	ResetModelsConfigCacheForTest()

 	cfg := loadModelsConfig()
 	if cfg.DefaultBackend != "opencode" {
@@ -84,11 +89,23 @@ func TestLoadModelsConfig_WithFile(t *testing.T) {
 	configContent := `{
 		"default_backend": "claude",
 		"default_model": "claude-opus-4",
+		"backends": {
+			"Claude": {
+				"base_url": "https://backend.example",
+				"api_key": "backend-key"
+			},
+			"codex": {
+				"base_url": "https://openai.example",
+				"api_key": "openai-key"
+			}
+		},
 		"agents": {
 			"custom-agent": {
 				"backend": "codex",
 				"model": "gpt-4o",
-				"description": "Custom agent"
+				"description": "Custom agent",
+				"base_url": "https://agent.example",
+				"api_key": "agent-key"
 			}
 		}
 	}`
@@ -99,6 +116,8 @@ func TestLoadModelsConfig_WithFile(t *testing.T) {

 	t.Setenv("HOME", tmpDir)
 	t.Setenv("USERPROFILE", tmpDir)
+	t.Cleanup(ResetModelsConfigCacheForTest)
+	ResetModelsConfigCacheForTest()

 	cfg := loadModelsConfig()

@@ -125,6 +144,55 @@ func TestLoadModelsConfig_WithFile(t *testing.T) {
 	if _, ok := cfg.Agents["oracle"]; !ok {
 		t.Error("default agent oracle should be merged")
 	}
+
+	baseURL, apiKey := ResolveBackendConfig("claude")
+	if baseURL != "https://backend.example" {
+		t.Errorf("ResolveBackendConfig(baseURL) = %q, want %q", baseURL, "https://backend.example")
+	}
+	if apiKey != "backend-key" {
+		t.Errorf("ResolveBackendConfig(apiKey) = %q, want %q", apiKey, "backend-key")
+	}
+
+	backend, model, _, _, agentBaseURL, agentAPIKey, _ := ResolveAgentConfig("custom-agent")
+	if backend != "codex" {
+		t.Errorf("ResolveAgentConfig(backend) = %q, want %q", backend, "codex")
+	}
+	if model != "gpt-4o" {
+		t.Errorf("ResolveAgentConfig(model) = %q, want %q", model, "gpt-4o")
+	}
+	if agentBaseURL != "https://agent.example" {
+		t.Errorf("ResolveAgentConfig(baseURL) = %q, want %q", agentBaseURL, "https://agent.example")
+	}
+	if agentAPIKey != "agent-key" {
+		t.Errorf("ResolveAgentConfig(apiKey) = %q, want %q", agentAPIKey, "agent-key")
+	}
+}
+
+func TestResolveAgentConfig_DynamicAgent(t *testing.T) {
+	home := t.TempDir()
+	t.Setenv("HOME", home)
+	t.Setenv("USERPROFILE", home)
+	t.Cleanup(ResetModelsConfigCacheForTest)
+	ResetModelsConfigCacheForTest()
+
+	agentDir := filepath.Join(home, ".codeagent", "agents")
+	if err := os.MkdirAll(agentDir, 0o755); err != nil {
+		t.Fatalf("MkdirAll: %v", err)
+	}
+	if err := os.WriteFile(filepath.Join(agentDir, "sarsh.md"), []byte("prompt\n"), 0o644); err != nil {
+		t.Fatalf("WriteFile: %v", err)
+	}
+
+	backend, model, promptFile, _, _, _, _ := resolveAgentConfig("sarsh")
+	if backend != "opencode" {
+		t.Errorf("backend = %q, want %q", backend, "opencode")
+	}
+	if model != "opencode/grok-code" {
+		t.Errorf("model = %q, want %q", model, "opencode/grok-code")
+	}
+	if promptFile != "~/.codeagent/agents/sarsh.md" {
+		t.Errorf("promptFile = %q, want %q", promptFile, "~/.codeagent/agents/sarsh.md")
+	}
 }

 func TestLoadModelsConfig_InvalidJSON(t *testing.T) {
@@ -142,6 +210,8 @@ func TestLoadModelsConfig_InvalidJSON(t *testing.T) {

 	t.Setenv("HOME", tmpDir)
 	t.Setenv("USERPROFILE", tmpDir)
+	t.Cleanup(ResetModelsConfigCacheForTest)
+	ResetModelsConfigCacheForTest()

 	cfg := loadModelsConfig()
 	// Should fall back to defaults
@@ -149,60 +219,3 @@ func TestLoadModelsConfig_InvalidJSON(t *testing.T) {
 		t.Errorf("invalid JSON should fallback, got DefaultBackend = %q", cfg.DefaultBackend)
 	}
 }
-
-func TestOpencodeBackend_BuildArgs(t *testing.T) {
-	backend := OpencodeBackend{}
-
-	t.Run("basic", func(t *testing.T) {
-		cfg := &Config{Mode: "new"}
-		got := backend.BuildArgs(cfg, "hello")
-		want := []string{"run", "--format", "json", "hello"}
-		if !reflect.DeepEqual(got, want) {
-			t.Errorf("got %v, want %v", got, want)
-		}
-	})
-
-	t.Run("with model", func(t *testing.T) {
-		cfg := &Config{Mode: "new", Model: "opencode/grok-code"}
-		got := backend.BuildArgs(cfg, "task")
-		want := []string{"run", "-m", "opencode/grok-code", "--format", "json", "task"}
-		if !reflect.DeepEqual(got, want) {
-			t.Errorf("got %v, want %v", got, want)
-		}
-	})
-
-	t.Run("resume mode", func(t *testing.T) {
-		cfg := &Config{Mode: "resume", SessionID: "ses_123", Model: "opencode/grok-code"}
-		got := backend.BuildArgs(cfg, "follow-up")
-		want := []string{"run", "-m", "opencode/grok-code", "-s", "ses_123", "--format", "json", "follow-up"}
-		if !reflect.DeepEqual(got, want) {
-			t.Errorf("got %v, want %v", got, want)
-		}
-	})
-
-	t.Run("resume without session", func(t *testing.T) {
-		cfg := &Config{Mode: "resume"}
-		got := backend.BuildArgs(cfg, "task")
-		want := []string{"run", "--format", "json", "task"}
-		if !reflect.DeepEqual(got, want) {
-			t.Errorf("got %v, want %v", got, want)
-		}
-	})
-}
-
-func TestOpencodeBackend_Interface(t *testing.T) {
-	backend := OpencodeBackend{}
-
-	if backend.Name() != "opencode" {
-		t.Errorf("Name() = %q, want %q", backend.Name(), "opencode")
-	}
-	if backend.Command() != "opencode" {
-		t.Errorf("Command() = %q, want %q", backend.Command(), "opencode")
-	}
-}
-
-func TestBackendRegistry_IncludesOpencode(t *testing.T) {
-	if _, ok := backendRegistry["opencode"]; !ok {
-		t.Error("backendRegistry should include opencode")
-	}
-}
--- a/codeagent-wrapper/internal/config/config.go
+++ b/codeagent-wrapper/internal/config/config.go
@@ -0,0 +1,102 @@
+package config
+
+import (
+	"fmt"
+	"os"
+	"strconv"
+	"strings"
+)
+
+// Config holds CLI configuration.
+type Config struct {
+	Mode               string // "new" or "resume"
+	Task               string
+	SessionID          string
+	WorkDir            string
+	Model              string
+	ReasoningEffort    string
+	ExplicitStdin      bool
+	Timeout            int
+	Backend            string
+	Agent              string
+	PromptFile         string
+	PromptFileExplicit bool
+	SkipPermissions    bool
+	Yolo               bool
+	MaxParallelWorkers int
+}
+
+// EnvFlagEnabled returns true when the environment variable exists and is not
+// explicitly set to a falsey value ("0/false/no/off").
+func EnvFlagEnabled(key string) bool {
+	val, ok := os.LookupEnv(key)
+	if !ok {
+		return false
+	}
+	val = strings.TrimSpace(strings.ToLower(val))
+	switch val {
+	case "", "0", "false", "no", "off":
+		return false
+	default:
+		return true
+	}
+}
+
+func ParseBoolFlag(val string, defaultValue bool) bool {
+	val = strings.TrimSpace(strings.ToLower(val))
+	switch val {
+	case "1", "true", "yes", "on":
+		return true
+	case "0", "false", "no", "off":
+		return false
+	default:
+		return defaultValue
+	}
+}
+
+// EnvFlagDefaultTrue returns true unless the env var is explicitly set to
+// false/0/no/off.
+func EnvFlagDefaultTrue(key string) bool {
+	val, ok := os.LookupEnv(key)
+	if !ok {
+		return true
+	}
+	return ParseBoolFlag(val, true)
+}
+
+func ValidateAgentName(name string) error {
+	if strings.TrimSpace(name) == "" {
+		return fmt.Errorf("agent name is empty")
+	}
+	for _, r := range name {
+		switch {
+		case r >= 'a' && r <= 'z':
+		case r >= 'A' && r <= 'Z':
+		case r >= '0' && r <= '9':
+		case r == '-', r == '_':
+		default:
+			return fmt.Errorf("agent name %q contains invalid character %q", name, r)
+		}
+	}
+	return nil
+}
+
+const maxParallelWorkersLimit = 100
+
+// ResolveMaxParallelWorkers reads CODEAGENT_MAX_PARALLEL_WORKERS. It returns 0
+// for "unlimited".
+func ResolveMaxParallelWorkers() int {
+	raw := strings.TrimSpace(os.Getenv("CODEAGENT_MAX_PARALLEL_WORKERS"))
+	if raw == "" {
+		return 0
+	}
+
+	value, err := strconv.Atoi(raw)
+	if err != nil || value < 0 {
+		return 0
+	}
+	if value > maxParallelWorkersLimit {
+		return maxParallelWorkersLimit
+	}
+	return value
+}
--- a/codeagent-wrapper/internal/config/viper.go
+++ b/codeagent-wrapper/internal/config/viper.go
@@ -0,0 +1,47 @@
+package config
+
+import (
+	"errors"
+	"os"
+	"path/filepath"
+	"strings"
+
+	"github.com/spf13/viper"
+)
+
+// NewViper returns a viper instance configured for CODEAGENT_* environment
+// variables and an optional config file.
+//
+// Search order when configFile is empty:
+//   - $HOME/.codeagent/config.(yaml|yml|json|toml|...)
+func NewViper(configFile string) (*viper.Viper, error) {
+	v := viper.New()
+	v.SetEnvPrefix("CODEAGENT")
+	v.SetEnvKeyReplacer(strings.NewReplacer("-", "_"))
+	v.AutomaticEnv()
+
+	if strings.TrimSpace(configFile) != "" {
+		v.SetConfigFile(configFile)
+		if err := v.ReadInConfig(); err != nil {
+			return nil, err
+		}
+		return v, nil
+	}
+
+	home, err := os.UserHomeDir()
+	if err != nil || strings.TrimSpace(home) == "" {
+		return v, nil
+	}
+
+	v.SetConfigName("config")
+	v.AddConfigPath(filepath.Join(home, ".codeagent"))
+	if err := v.ReadInConfig(); err != nil {
+		var notFound viper.ConfigFileNotFoundError
+		if errors.As(err, &notFound) {
+			return v, nil
+		}
+		return nil, err
+	}
+
+	return v, nil
+}
--- a/codeagent-wrapper/internal/executor/env_inject_test.go
+++ b/codeagent-wrapper/internal/executor/env_inject_test.go
@@ -0,0 +1,193 @@
+package executor
+
+import (
+	"os"
+	"path/filepath"
+	"strings"
+	"testing"
+
+	backend "codeagent-wrapper/internal/backend"
+	config "codeagent-wrapper/internal/config"
+)
+
+// TestEnvInjectionWithAgent tests the full flow of env injection with agent config
+func TestEnvInjectionWithAgent(t *testing.T) {
+	// Setup temp config
+	tmpDir := t.TempDir()
+	configDir := filepath.Join(tmpDir, ".codeagent")
+	if err := os.MkdirAll(configDir, 0755); err != nil {
+		t.Fatal(err)
+	}
+
+	// Write test config with agent that has base_url and api_key
+	configContent := `{
+		"default_backend": "codex",
+		"agents": {
+			"test-agent": {
+				"backend": "claude",
+				"model": "test-model",
+				"base_url": "https://test.api.com",
+				"api_key": "test-api-key-12345678"
+			}
+		}
+	}`
+	configPath := filepath.Join(configDir, "models.json")
+	if err := os.WriteFile(configPath, []byte(configContent), 0644); err != nil {
+		t.Fatal(err)
+	}
+
+	// Override HOME to use temp dir
+	oldHome := os.Getenv("HOME")
+	os.Setenv("HOME", tmpDir)
+	defer os.Setenv("HOME", oldHome)
+
+	// Reset config cache
+	config.ResetModelsConfigCacheForTest()
+	defer config.ResetModelsConfigCacheForTest()
+
+	// Test ResolveAgentConfig
+	agentBackend, model, _, _, baseURL, apiKey, _ := config.ResolveAgentConfig("test-agent")
+	t.Logf("ResolveAgentConfig: backend=%q, model=%q, baseURL=%q, apiKey=%q",
+		agentBackend, model, baseURL, apiKey)
+
+	if agentBackend != "claude" {
+		t.Errorf("expected backend 'claude', got %q", agentBackend)
+	}
+	if baseURL != "https://test.api.com" {
+		t.Errorf("expected baseURL 'https://test.api.com', got %q", baseURL)
+	}
+	if apiKey != "test-api-key-12345678" {
+		t.Errorf("expected apiKey 'test-api-key-12345678', got %q", apiKey)
+	}
+
+	// Test Backend.Env
+	b := backend.ClaudeBackend{}
+	env := b.Env(baseURL, apiKey)
+	t.Logf("Backend.Env: %v", env)
+
+	if env == nil {
+		t.Fatal("expected non-nil env from Backend.Env")
+	}
+	if env["ANTHROPIC_BASE_URL"] != baseURL {
+		t.Errorf("expected ANTHROPIC_BASE_URL=%q, got %q", baseURL, env["ANTHROPIC_BASE_URL"])
+	}
+	if env["ANTHROPIC_AUTH_TOKEN"] != apiKey {
+		t.Errorf("expected ANTHROPIC_AUTH_TOKEN=%q, got %q", apiKey, env["ANTHROPIC_AUTH_TOKEN"])
+	}
+}
+
+// TestEnvInjectionLogic tests the exact logic used in executor
+func TestEnvInjectionLogic(t *testing.T) {
+	// Setup temp config
+	tmpDir := t.TempDir()
+	configDir := filepath.Join(tmpDir, ".codeagent")
+	if err := os.MkdirAll(configDir, 0755); err != nil {
+		t.Fatal(err)
+	}
+
+	configContent := `{
+		"default_backend": "codex",
+		"agents": {
+			"explore": {
+				"backend": "claude",
+				"model": "MiniMax-M2.1",
+				"base_url": "https://api.minimaxi.com/anthropic",
+				"api_key": "eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCJ9.test"
+			}
+		}
+	}`
+	configPath := filepath.Join(configDir, "models.json")
+	if err := os.WriteFile(configPath, []byte(configContent), 0644); err != nil {
+		t.Fatal(err)
+	}
+
+	oldHome := os.Getenv("HOME")
+	os.Setenv("HOME", tmpDir)
+	defer os.Setenv("HOME", oldHome)
+
+	config.ResetModelsConfigCacheForTest()
+	defer config.ResetModelsConfigCacheForTest()
+
+	// Simulate the executor logic
+	cfgBackend := "claude" // This should come from taskSpec.Backend
+	agentName := "explore"
+
+	// Step 1: Get backend config (usually empty for claude without global config)
+	baseURL, apiKey := config.ResolveBackendConfig(cfgBackend)
+	t.Logf("Step 1 - ResolveBackendConfig(%q): baseURL=%q, apiKey=%q", cfgBackend, baseURL, apiKey)
+
+	// Step 2: If agent specified, get agent config
+	if agentName != "" {
+		agentBackend, _, _, _, agentBaseURL, agentAPIKey, _ := config.ResolveAgentConfig(agentName)
+		t.Logf("Step 2 - ResolveAgentConfig(%q): backend=%q, baseURL=%q, apiKey=%q",
+			agentName, agentBackend, agentBaseURL, agentAPIKey)
+
+		// Step 3: Check if agent backend matches cfg backend
+		if strings.EqualFold(strings.TrimSpace(agentBackend), strings.TrimSpace(cfgBackend)) {
+			baseURL, apiKey = agentBaseURL, agentAPIKey
+			t.Logf("Step 3 - Backend match! Using agent config: baseURL=%q, apiKey=%q", baseURL, apiKey)
+		} else {
+			t.Logf("Step 3 - Backend mismatch: agent=%q, cfg=%q", agentBackend, cfgBackend)
+		}
+	}
+
+	// Step 4: Get env vars from backend
+	b := backend.ClaudeBackend{}
+	injected := b.Env(baseURL, apiKey)
+	t.Logf("Step 4 - Backend.Env: %v", injected)
+
+	// Verify
+	if len(injected) == 0 {
+		t.Fatal("Expected env vars to be injected, got none")
+	}
+
+	expectedURL := "https://api.minimaxi.com/anthropic"
+	if injected["ANTHROPIC_BASE_URL"] != expectedURL {
+		t.Errorf("ANTHROPIC_BASE_URL: expected %q, got %q", expectedURL, injected["ANTHROPIC_BASE_URL"])
+	}
+
+	if _, ok := injected["ANTHROPIC_AUTH_TOKEN"]; !ok {
+		t.Error("ANTHROPIC_AUTH_TOKEN not set")
+	}
+
+	// Step 5: Test masking
+	for k, v := range injected {
+		masked := maskSensitiveValue(k, v)
+		t.Logf("Step 5 - Env log: %s=%s", k, masked)
+	}
+}
+
+// TestTaskSpecBackendPropagation tests that taskSpec.Backend is properly used
+func TestTaskSpecBackendPropagation(t *testing.T) {
+	// Simulate what happens in RunCodexTaskWithContext
+	taskSpec := TaskSpec{
+		ID:      "test",
+		Task:    "hello",
+		Backend: "claude",
+		Agent:   "explore",
+	}
+
+	// This is the logic from executor.go lines 889-916
+	cfg := &config.Config{
+		Mode:    "new",
+		Task:    taskSpec.Task,
+		Backend: "codex", // default
+	}
+
+	var backend Backend = nil // nil in single mode
+	commandName := "codex"    // default
+
+	if backend != nil {
+		cfg.Backend = backend.Name()
+	} else if taskSpec.Backend != "" {
+		cfg.Backend = taskSpec.Backend
+	} else if commandName != "" {
+		cfg.Backend = commandName
+	}
+
+	t.Logf("taskSpec.Backend=%q, cfg.Backend=%q", taskSpec.Backend, cfg.Backend)
+
+	if cfg.Backend != "claude" {
+		t.Errorf("expected cfg.Backend='claude', got %q", cfg.Backend)
+	}
+}
--- a/codeagent-wrapper/internal/executor/env_logging_test.go
+++ b/codeagent-wrapper/internal/executor/env_logging_test.go
@@ -0,0 +1,333 @@
+package executor
+
+import (
+	"strings"
+	"testing"
+
+	backend "codeagent-wrapper/internal/backend"
+)
+
+func TestMaskSensitiveValue(t *testing.T) {
+	tests := []struct {
+		name     string
+		key      string
+		value    string
+		expected string
+	}{
+		{
+			name:     "API_KEY with long value",
+			key:      "ANTHROPIC_AUTH_TOKEN",
+			value:    "sk-ant-api03-xxxxxxxxxxxxxxxxxxxxxxxxxxxx",
+			expected: "sk-a****xxxx",
+		},
+		{
+			name:     "api_key lowercase",
+			key:      "api_key",
+			value:    "abcdefghijklmnop",
+			expected: "abcd****mnop",
+		},
+		{
+			name:     "AUTH_TOKEN",
+			key:      "AUTH_TOKEN",
+			value:    "eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCJ9",
+			expected: "eyJh****VCJ9",
+		},
+		{
+			name:     "SECRET",
+			key:      "MY_SECRET",
+			value:    "super-secret-value-12345",
+			expected: "supe****2345",
+		},
+		{
+			name:     "short key value (8 chars)",
+			key:      "API_KEY",
+			value:    "12345678",
+			expected: "****",
+		},
+		{
+			name:     "very short key value",
+			key:      "API_KEY",
+			value:    "abc",
+			expected: "****",
+		},
+		{
+			name:     "empty key value",
+			key:      "API_KEY",
+			value:    "",
+			expected: "",
+		},
+		{
+			name:     "non-sensitive BASE_URL",
+			key:      "ANTHROPIC_BASE_URL",
+			value:    "https://api.anthropic.com",
+			expected: "https://api.anthropic.com",
+		},
+		{
+			name:     "non-sensitive MODEL",
+			key:      "MODEL",
+			value:    "claude-3-opus",
+			expected: "claude-3-opus",
+		},
+		{
+			name:     "case insensitive - Key",
+			key:      "My_Key",
+			value:    "1234567890abcdef",
+			expected: "1234****cdef",
+		},
+		{
+			name:     "case insensitive - TOKEN",
+			key:      "ACCESS_TOKEN",
+			value:    "access123456789",
+			expected: "acce****6789",
+		},
+		{
+			name:     "partial match - apikey",
+			key:      "MYAPIKEY",
+			value:    "1234567890",
+			expected: "1234****7890",
+		},
+		{
+			name:     "partial match - secretvalue",
+			key:      "SECRETVALUE",
+			value:    "abcdefghij",
+			expected: "abcd****ghij",
+		},
+		{
+			name:     "9 char value (just above threshold)",
+			key:      "API_KEY",
+			value:    "123456789",
+			expected: "1234****6789",
+		},
+		{
+			name:     "exactly 8 char value (at threshold)",
+			key:      "API_KEY",
+			value:    "12345678",
+			expected: "****",
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			result := maskSensitiveValue(tt.key, tt.value)
+			if result != tt.expected {
+				t.Errorf("maskSensitiveValue(%q, %q) = %q, want %q", tt.key, tt.value, result, tt.expected)
+			}
+		})
+	}
+}
+
+func TestMaskSensitiveValue_NoLeakage(t *testing.T) {
+	// Ensure sensitive values are never fully exposed
+	sensitiveKeys := []string{"API_KEY", "api_key", "AUTH_TOKEN", "SECRET", "access_token", "MYAPIKEY"}
+	longValue := "this-is-a-very-long-secret-value-that-should-be-masked"
+
+	for _, key := range sensitiveKeys {
+		t.Run(key, func(t *testing.T) {
+			masked := maskSensitiveValue(key, longValue)
+			// Should not contain the full value
+			if masked == longValue {
+				t.Errorf("key %q: value was not masked", key)
+			}
+			// Should contain mask marker
+			if !strings.Contains(masked, "****") {
+				t.Errorf("key %q: masked value %q does not contain ****", key, masked)
+			}
+			// First 4 chars should be visible
+			if !strings.HasPrefix(masked, longValue[:4]) {
+				t.Errorf("key %q: masked value should start with first 4 chars", key)
+			}
+			// Last 4 chars should be visible
+			if !strings.HasSuffix(masked, longValue[len(longValue)-4:]) {
+				t.Errorf("key %q: masked value should end with last 4 chars", key)
+			}
+		})
+	}
+}
+
+func TestMaskSensitiveValue_NonSensitivePassthrough(t *testing.T) {
+	// Non-sensitive keys should pass through unchanged
+	nonSensitiveKeys := []string{
+		"ANTHROPIC_BASE_URL",
+		"BASE_URL",
+		"MODEL",
+		"BACKEND",
+		"WORKDIR",
+		"HOME",
+		"PATH",
+	}
+	value := "any-value-here-12345"
+
+	for _, key := range nonSensitiveKeys {
+		t.Run(key, func(t *testing.T) {
+			result := maskSensitiveValue(key, value)
+			if result != value {
+				t.Errorf("key %q: expected passthrough but got %q", key, result)
+			}
+		})
+	}
+}
+
+// TestClaudeBackendEnv tests that ClaudeBackend.Env returns correct env vars
+func TestClaudeBackendEnv(t *testing.T) {
+	tests := []struct {
+		name       string
+		baseURL    string
+		apiKey     string
+		expectKeys []string
+		expectNil  bool
+	}{
+		{
+			name:       "both base_url and api_key",
+			baseURL:    "https://api.custom.com",
+			apiKey:     "sk-test-key-12345",
+			expectKeys: []string{"ANTHROPIC_BASE_URL", "ANTHROPIC_AUTH_TOKEN"},
+		},
+		{
+			name:       "only base_url",
+			baseURL:    "https://api.custom.com",
+			apiKey:     "",
+			expectKeys: []string{"ANTHROPIC_BASE_URL"},
+		},
+		{
+			name:       "only api_key",
+			baseURL:    "",
+			apiKey:     "sk-test-key-12345",
+			expectKeys: []string{"ANTHROPIC_AUTH_TOKEN"},
+		},
+		{
+			name:      "both empty",
+			baseURL:   "",
+			apiKey:    "",
+			expectNil: true,
+		},
+		{
+			name:      "whitespace only",
+			baseURL:   "   ",
+			apiKey:    "  ",
+			expectNil: true,
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			b := backend.ClaudeBackend{}
+			env := b.Env(tt.baseURL, tt.apiKey)
+
+			if tt.expectNil {
+				if env != nil {
+					t.Errorf("expected nil env, got %v", env)
+				}
+				return
+			}
+
+			if env == nil {
+				t.Fatal("expected non-nil env")
+			}
+
+			for _, key := range tt.expectKeys {
+				if _, ok := env[key]; !ok {
+					t.Errorf("expected key %q in env", key)
+				}
+			}
+
+			// Verify values are correct
+			if tt.baseURL != "" && strings.TrimSpace(tt.baseURL) != "" {
+				if env["ANTHROPIC_BASE_URL"] != strings.TrimSpace(tt.baseURL) {
+					t.Errorf("ANTHROPIC_BASE_URL = %q, want %q", env["ANTHROPIC_BASE_URL"], strings.TrimSpace(tt.baseURL))
+				}
+			}
+			if tt.apiKey != "" && strings.TrimSpace(tt.apiKey) != "" {
+				if env["ANTHROPIC_AUTH_TOKEN"] != strings.TrimSpace(tt.apiKey) {
+					t.Errorf("ANTHROPIC_AUTH_TOKEN = %q, want %q", env["ANTHROPIC_AUTH_TOKEN"], strings.TrimSpace(tt.apiKey))
+				}
+			}
+		})
+	}
+}
+
+// TestEnvLoggingIntegration tests that env vars are properly masked in logs
+func TestEnvLoggingIntegration(t *testing.T) {
+	b := backend.ClaudeBackend{}
+	baseURL := "https://api.minimaxi.com/anthropic"
+	apiKey := "eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCJ9.longjwttoken"
+
+	env := b.Env(baseURL, apiKey)
+	if env == nil {
+		t.Fatal("expected non-nil env")
+	}
+
+	// Verify that when we log these values, sensitive ones are masked
+	for k, v := range env {
+		masked := maskSensitiveValue(k, v)
+
+		if k == "ANTHROPIC_BASE_URL" {
+			// URL should not be masked
+			if masked != v {
+				t.Errorf("BASE_URL should not be masked: got %q, want %q", masked, v)
+			}
+		}
+
+		if k == "ANTHROPIC_AUTH_TOKEN" {
+			// API key should be masked
+			if masked == v {
+				t.Errorf("API_KEY should be masked, but got original value")
+			}
+			if !strings.Contains(masked, "****") {
+				t.Errorf("masked API_KEY should contain ****: got %q", masked)
+			}
+			// Should still show first 4 and last 4 chars
+			if !strings.HasPrefix(masked, v[:4]) {
+				t.Errorf("masked value should start with first 4 chars of original")
+			}
+			if !strings.HasSuffix(masked, v[len(v)-4:]) {
+				t.Errorf("masked value should end with last 4 chars of original")
+			}
+		}
+	}
+}
+
+// TestGeminiBackendEnv tests GeminiBackend.Env for comparison
+func TestGeminiBackendEnv(t *testing.T) {
+	b := backend.GeminiBackend{}
+	env := b.Env("https://custom.api", "gemini-api-key-12345")
+
+	if env == nil {
+		t.Fatal("expected non-nil env")
+	}
+
+	// Check that GEMINI env vars are set
+	if _, ok := env["GOOGLE_GEMINI_BASE_URL"]; !ok {
+		t.Error("expected GOOGLE_GEMINI_BASE_URL in env")
+	}
+	if _, ok := env["GEMINI_API_KEY"]; !ok {
+		t.Error("expected GEMINI_API_KEY in env")
+	}
+
+	// Verify masking works for Gemini keys too
+	for k, v := range env {
+		masked := maskSensitiveValue(k, v)
+		if strings.Contains(strings.ToLower(k), "key") {
+			if masked == v && len(v) > 0 {
+				t.Errorf("key %q should be masked", k)
+			}
+		}
+	}
+}
+
+// TestCodexBackendEnv tests CodexBackend.Env
+func TestCodexBackendEnv(t *testing.T) {
+	b := backend.CodexBackend{}
+	env := b.Env("https://custom.api", "codex-api-key-12345")
+
+	if env == nil {
+		t.Fatal("expected non-nil env for codex")
+	}
+
+	// Check for OPENAI env vars
+	if _, ok := env["OPENAI_BASE_URL"]; !ok {
+		t.Error("expected OPENAI_BASE_URL in env")
+	}
+	if _, ok := env["OPENAI_API_KEY"]; !ok {
+		t.Error("expected OPENAI_API_KEY in env")
+	}
+}
--- a/codeagent-wrapper/internal/executor/env_stderr_test.go
+++ b/codeagent-wrapper/internal/executor/env_stderr_test.go
@@ -0,0 +1,133 @@
+package executor
+
+import (
+	"context"
+	"errors"
+	"io"
+	"os"
+	"path/filepath"
+	"strings"
+	"testing"
+
+	config "codeagent-wrapper/internal/config"
+)
+
+type fakeCmd struct {
+	env map[string]string
+}
+
+func (f *fakeCmd) Start() error { return nil }
+func (f *fakeCmd) Wait() error  { return nil }
+func (f *fakeCmd) StdoutPipe() (io.ReadCloser, error) {
+	return io.NopCloser(strings.NewReader("")), nil
+}
+func (f *fakeCmd) StderrPipe() (io.ReadCloser, error) {
+	return nil, errors.New("fake stderr pipe error")
+}
+func (f *fakeCmd) StdinPipe() (io.WriteCloser, error) {
+	return nil, errors.New("fake stdin pipe error")
+}
+func (f *fakeCmd) SetStderr(io.Writer) {}
+func (f *fakeCmd) SetDir(string)       {}
+func (f *fakeCmd) SetEnv(env map[string]string) {
+	if len(env) == 0 {
+		return
+	}
+	if f.env == nil {
+		f.env = make(map[string]string, len(env))
+	}
+	for k, v := range env {
+		f.env[k] = v
+	}
+}
+func (f *fakeCmd) Process() processHandle { return nil }
+
+func TestEnvInjection_LogsToStderrAndMasksKey(t *testing.T) {
+	// Arrange ~/.codeagent/models.json via HOME override.
+	tmpDir := t.TempDir()
+	configDir := filepath.Join(tmpDir, ".codeagent")
+	if err := os.MkdirAll(configDir, 0o755); err != nil {
+		t.Fatal(err)
+	}
+	const baseURL = "https://api.minimaxi.com/anthropic"
+	const apiKey = "eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCJ9.test"
+	models := `{
+  "agents": {
+    "explore": {
+      "backend": "claude",
+      "model": "MiniMax-M2.1",
+      "base_url": "` + baseURL + `",
+      "api_key": "` + apiKey + `"
+    }
+  }
+}`
+	if err := os.WriteFile(filepath.Join(configDir, "models.json"), []byte(models), 0o644); err != nil {
+		t.Fatal(err)
+	}
+
+	oldHome := os.Getenv("HOME")
+	if err := os.Setenv("HOME", tmpDir); err != nil {
+		t.Fatal(err)
+	}
+	defer func() { _ = os.Setenv("HOME", oldHome) }()
+
+	config.ResetModelsConfigCacheForTest()
+	defer config.ResetModelsConfigCacheForTest()
+
+	// Capture stderr (RunCodexTaskWithContext prints env injection lines there).
+	r, w, err := os.Pipe()
+	if err != nil {
+		t.Fatal(err)
+	}
+	oldStderr := os.Stderr
+	os.Stderr = w
+	defer func() { os.Stderr = oldStderr }()
+
+	readDone := make(chan string, 1)
+	go func() {
+		defer r.Close()
+		b, _ := io.ReadAll(r)
+		readDone <- string(b)
+	}()
+
+	var cmd *fakeCmd
+	restoreRunner := SetNewCommandRunner(func(ctx context.Context, name string, args ...string) CommandRunner {
+		cmd = &fakeCmd{}
+		return cmd
+	})
+	defer restoreRunner()
+
+	// Act: force an early return right after env injection by making StderrPipe fail.
+	_ = RunCodexTaskWithContext(
+		context.Background(),
+		TaskSpec{Task: "hi", WorkDir: ".", Backend: "claude", Agent: "explore"},
+		nil,
+		"claude",
+		nil,
+		nil,
+		false,
+		false,
+		1,
+	)
+
+	_ = w.Close()
+	got := <-readDone
+
+	// Assert: env was injected into the command and logging is present with masking.
+	if cmd == nil || cmd.env == nil {
+		t.Fatalf("expected cmd env to be set, got cmd=%v env=%v", cmd, nil)
+	}
+	if cmd.env["ANTHROPIC_BASE_URL"] != baseURL {
+		t.Fatalf("ANTHROPIC_BASE_URL=%q, want %q", cmd.env["ANTHROPIC_BASE_URL"], baseURL)
+	}
+	if cmd.env["ANTHROPIC_AUTH_TOKEN"] != apiKey {
+		t.Fatalf("ANTHROPIC_AUTH_TOKEN=%q, want %q", cmd.env["ANTHROPIC_AUTH_TOKEN"], apiKey)
+	}
+
+	if !strings.Contains(got, "Env: ANTHROPIC_BASE_URL="+baseURL) {
+		t.Fatalf("stderr missing base URL env log; stderr=%q", got)
+	}
+	if !strings.Contains(got, "Env: ANTHROPIC_AUTH_TOKEN=eyJh****test") {
+		t.Fatalf("stderr missing masked API key log; stderr=%q", got)
+	}
+}
--- a/codeagent-wrapper/internal/executor/executor.go
+++ b/codeagent-wrapper/internal/executor/executor.go
@@ -1,4 +1,4 @@
-package main
+package executor

 import (
 	"context"
@@ -14,11 +14,92 @@ import (
 	"sync/atomic"
 	"syscall"
 	"time"
+
+	backend "codeagent-wrapper/internal/backend"
+	config "codeagent-wrapper/internal/config"
+	ilogger "codeagent-wrapper/internal/logger"
+	parser "codeagent-wrapper/internal/parser"
+	utils "codeagent-wrapper/internal/utils"
 )

 const postMessageTerminateDelay = 1 * time.Second
 const forceKillWaitTimeout = 5 * time.Second

+// Defaults duplicated from wrapper for module decoupling.
+const (
+	defaultWorkdir        = "."
+	defaultCoverageTarget = 90.0
+	defaultBackendName    = "codex"
+
+	codexLogLineLimit  = 1000
+	stderrCaptureLimit = 4 * 1024
+)
+
+const (
+	// stdout close reasons
+	stdoutCloseReasonWait  = "wait-done"
+	stdoutCloseReasonDrain = "drain-timeout"
+	stdoutCloseReasonCtx   = "context-cancel"
+	stdoutDrainTimeout     = 100 * time.Millisecond
+)
+
+// Hook points (tests can override inside this package).
+var (
+	selectBackendFn    = backend.Select
+	commandContext     = exec.CommandContext
+	terminateCommandFn = terminateCommand
+)
+
+var forceKillDelay atomic.Int32
+
+func init() {
+	forceKillDelay.Store(5) // seconds - default value
+}
+
+type (
+	Backend = backend.Backend
+	Config  = config.Config
+	Logger  = ilogger.Logger
+)
+
+type minimalClaudeSettings = backend.MinimalClaudeSettings
+
+func loadMinimalClaudeSettings() minimalClaudeSettings { return backend.LoadMinimalClaudeSettings() }
+
+func loadGeminiEnv() map[string]string { return backend.LoadGeminiEnv() }
+
+func NewLogger() (*Logger, error) { return ilogger.NewLogger() }
+
+func NewLoggerWithSuffix(suffix string) (*Logger, error) { return ilogger.NewLoggerWithSuffix(suffix) }
+
+func setLogger(l *Logger) { ilogger.SetLogger(l) }
+
+func closeLogger() error { return ilogger.CloseLogger() }
+
+func activeLogger() *Logger { return ilogger.ActiveLogger() }
+
+func logInfo(msg string) { ilogger.LogInfo(msg) }
+
+func logWarn(msg string) { ilogger.LogWarn(msg) }
+
+func logError(msg string) { ilogger.LogError(msg) }
+
+func logConcurrencyPlanning(limit, total int) { ilogger.LogConcurrencyPlanning(limit, total) }
+
+func logConcurrencyState(event, taskID string, active, limit int) {
+	ilogger.LogConcurrencyState(event, taskID, active, limit)
+}
+
+func parseJSONStreamInternal(r io.Reader, warnFn func(string), infoFn func(string), onMessage func(), onComplete func()) (message, threadID string) {
+	return parser.ParseJSONStreamInternal(r, warnFn, infoFn, onMessage, onComplete)
+}
+
+func sanitizeOutput(s string) string { return utils.SanitizeOutput(s) }
+
+func safeTruncate(s string, maxLen int) string { return utils.SafeTruncate(s, maxLen) }
+
+func min(a, b int) int { return utils.Min(a, b) }
+
 // commandRunner abstracts exec.Cmd for testability
 type commandRunner interface {
 	Start() error
@@ -230,7 +311,7 @@ func newTaskLoggerHandle(taskID string) taskLoggerHandle {
 }

 // defaultRunCodexTaskFn is the default implementation of runCodexTaskFn (exposed for test reset)
-func defaultRunCodexTaskFn(task TaskSpec, timeout int) TaskResult {
+func DefaultRunCodexTaskFn(task TaskSpec, timeout int) TaskResult {
 	if task.WorkDir == "" {
 		task.WorkDir = defaultWorkdir
 	}
@@ -238,13 +319,13 @@ func defaultRunCodexTaskFn(task TaskSpec, timeout int) TaskResult {
 		task.Mode = "new"
 	}
 	if strings.TrimSpace(task.PromptFile) != "" {
-		prompt, err := readAgentPromptFile(task.PromptFile, false)
+		prompt, err := ReadAgentPromptFile(task.PromptFile, false)
 		if err != nil {
 			return TaskResult{TaskID: task.ID, ExitCode: 1, Error: "failed to read prompt file: " + err.Error()}
 		}
-		task.Task = wrapTaskWithAgentPrompt(prompt, task.Task)
+		task.Task = WrapTaskWithAgentPrompt(prompt, task.Task)
 	}
-	if task.UseStdin || shouldUseStdin(task.Task, false) {
+	if task.UseStdin || ShouldUseStdin(task.Task, false) {
 		task.UseStdin = true
 	}

@@ -263,12 +344,10 @@ func defaultRunCodexTaskFn(task TaskSpec, timeout int) TaskResult {
 	if parentCtx == nil {
 		parentCtx = context.Background()
 	}
-	return runCodexTaskWithContext(parentCtx, task, backend, nil, false, true, timeout)
+	return RunCodexTaskWithContext(parentCtx, task, backend, "", nil, nil, false, true, timeout)
 }

-var runCodexTaskFn = defaultRunCodexTaskFn
-
-func topologicalSort(tasks []TaskSpec) ([][]TaskSpec, error) {
+func TopologicalSort(tasks []TaskSpec) ([][]TaskSpec, error) {
 	idToTask := make(map[string]TaskSpec, len(tasks))
 	indegree := make(map[string]int, len(tasks))
 	adj := make(map[string][]string, len(tasks))
@@ -334,12 +413,16 @@ func topologicalSort(tasks []TaskSpec) ([][]TaskSpec, error) {
 	return layers, nil
 }

-func executeConcurrent(layers [][]TaskSpec, timeout int) []TaskResult {
-	maxWorkers := resolveMaxParallelWorkers()
-	return executeConcurrentWithContext(context.Background(), layers, timeout, maxWorkers)
+func ExecuteConcurrent(layers [][]TaskSpec, timeout int, runTask func(TaskSpec, int) TaskResult) []TaskResult {
+	maxWorkers := config.ResolveMaxParallelWorkers()
+	return ExecuteConcurrentWithContext(context.Background(), layers, timeout, maxWorkers, runTask)
 }

-func executeConcurrentWithContext(parentCtx context.Context, layers [][]TaskSpec, timeout int, maxWorkers int) []TaskResult {
+func ExecuteConcurrentWithContext(parentCtx context.Context, layers [][]TaskSpec, timeout int, maxWorkers int, runTask func(TaskSpec, int) TaskResult) []TaskResult {
+	if runTask == nil {
+		runTask = DefaultRunCodexTaskFn
+	}
+
 	totalTasks := 0
 	for _, layer := range layers {
 		totalTasks += len(layer)
@@ -470,7 +553,7 @@ func executeConcurrentWithContext(parentCtx context.Context, layers [][]TaskSpec

 				printTaskStart(ts.ID, taskLogPath, handle.shared)

-				res := runCodexTaskFn(ts, timeout)
+				res := runTask(ts, timeout)
 				if taskLogPath != "" {
 					if res.LogPath == "" || (handle.shared && handle.logger != nil && res.LogPath == handle.logger.Path()) {
 						res.LogPath = taskLogPath
@@ -535,14 +618,14 @@ func getStatusSymbols() (success, warning, failed string) {
 	return "✓", "⚠️", "✗"
 }

-func generateFinalOutput(results []TaskResult) string {
-	return generateFinalOutputWithMode(results, true) // default to summary mode
+func GenerateFinalOutput(results []TaskResult) string {
+	return GenerateFinalOutputWithMode(results, true) // default to summary mode
 }

 // generateFinalOutputWithMode generates output based on mode
 // summaryOnly=true: structured report - every token has value
 // summaryOnly=false: full output with complete messages (legacy behavior)
-func generateFinalOutputWithMode(results []TaskResult, summaryOnly bool) string {
+func GenerateFinalOutputWithMode(results []TaskResult, summaryOnly bool) string {
 	var sb strings.Builder
 	successSymbol, warningSymbol, failedSymbol := getStatusSymbols()

@@ -756,7 +839,7 @@ func buildCodexArgs(cfg *Config, targetArg string) []string {
 	args := []string{"e"}

 	// Default to bypass sandbox unless CODEX_BYPASS_SANDBOX=false
-	if cfg.Yolo || envFlagDefaultTrue("CODEX_BYPASS_SANDBOX") {
+	if cfg.Yolo || config.EnvFlagDefaultTrue("CODEX_BYPASS_SANDBOX") {
 		logWarn("YOLO mode or CODEX_BYPASS_SANDBOX enabled: running without approval/sandbox protection")
 		args = append(args, "--dangerously-bypass-approvals-and-sandbox")
 	}
@@ -766,7 +849,7 @@ func buildCodexArgs(cfg *Config, targetArg string) []string {
 	}

 	if reasoningEffort := strings.TrimSpace(cfg.ReasoningEffort); reasoningEffort != "" {
-		args = append(args, "--reasoning-effort", reasoningEffort)
+		args = append(args, "-c", "model_reasoning_effort="+reasoningEffort)
 	}

 	args = append(args, "--skip-git-repo-check")
@@ -787,25 +870,20 @@ func buildCodexArgs(cfg *Config, targetArg string) []string {
 	)
 }

-func runCodexTask(taskSpec TaskSpec, silent bool, timeoutSec int) TaskResult {
-	return runCodexTaskWithContext(context.Background(), taskSpec, nil, nil, false, silent, timeoutSec)
-}
-
-func runCodexProcess(parentCtx context.Context, codexArgs []string, taskText string, useStdin bool, timeoutSec int) (message, threadID string, exitCode int) {
-	res := runCodexTaskWithContext(parentCtx, TaskSpec{Task: taskText, WorkDir: defaultWorkdir, Mode: "new", UseStdin: useStdin}, nil, codexArgs, true, false, timeoutSec)
-	return res.Message, res.SessionID, res.ExitCode
-}
-
-func runCodexTaskWithContext(parentCtx context.Context, taskSpec TaskSpec, backend Backend, customArgs []string, useCustomArgs bool, silent bool, timeoutSec int) TaskResult {
+func RunCodexTaskWithContext(parentCtx context.Context, taskSpec TaskSpec, backend Backend, defaultCommandName string, defaultArgsBuilder func(*Config, string) []string, customArgs []string, useCustomArgs bool, silent bool, timeoutSec int) TaskResult {
+	taskCtx := taskSpec.Context
 	if parentCtx == nil {
-		parentCtx = taskSpec.Context
+		parentCtx = taskCtx
 	}
 	if parentCtx == nil {
 		parentCtx = context.Background()
 	}

 	result := TaskResult{TaskID: taskSpec.ID}
-	injectedLogger := taskLoggerFromContext(parentCtx)
+	injectedLogger := taskLoggerFromContext(taskCtx)
+	if injectedLogger == nil {
+		injectedLogger = taskLoggerFromContext(parentCtx)
+	}
 	logger := injectedLogger

 	cfg := &Config{
@@ -819,8 +897,14 @@ func runCodexTaskWithContext(parentCtx context.Context, taskSpec TaskSpec, backe
 		Backend:         defaultBackendName,
 	}

-	commandName := codexCommand
-	argsBuilder := buildCodexArgsFn
+	commandName := strings.TrimSpace(defaultCommandName)
+	if commandName == "" {
+		commandName = defaultBackendName
+	}
+	argsBuilder := defaultArgsBuilder
+	if argsBuilder == nil {
+		argsBuilder = buildCodexArgs
+	}
 	if backend != nil {
 		commandName = backend.Command()
 		argsBuilder = backend.BuildArgs
@@ -844,19 +928,23 @@ func runCodexTaskWithContext(parentCtx context.Context, taskSpec TaskSpec, backe
 		return result
 	}

-	var claudeEnv map[string]string
+	var fileEnv map[string]string
 	if cfg.Backend == "claude" {
 		settings := loadMinimalClaudeSettings()
-		claudeEnv = settings.Env
+		fileEnv = settings.Env
 		if cfg.Mode != "resume" && strings.TrimSpace(cfg.Model) == "" && settings.Model != "" {
 			cfg.Model = settings.Model
 		}
 	}

 	// Load gemini env from ~/.gemini/.env if exists
-	var geminiEnv map[string]string
 	if cfg.Backend == "gemini" {
-		geminiEnv = loadGeminiEnv()
+		fileEnv = loadGeminiEnv()
+		if cfg.Mode != "resume" && strings.TrimSpace(cfg.Model) == "" {
+			if model := fileEnv["GEMINI_MODEL"]; model != "" {
+				cfg.Model = model
+			}
+		}
 	}

 	useStdin := taskSpec.UseStdin
@@ -958,11 +1046,34 @@ func runCodexTaskWithContext(parentCtx context.Context, taskSpec TaskSpec, backe

 	cmd := newCommandRunner(ctx, commandName, codexArgs...)

-	if cfg.Backend == "claude" && len(claudeEnv) > 0 {
-		cmd.SetEnv(claudeEnv)
+	if len(fileEnv) > 0 {
+		cmd.SetEnv(fileEnv)
 	}
-	if cfg.Backend == "gemini" && len(geminiEnv) > 0 {
-		cmd.SetEnv(geminiEnv)
+
+	envBackend := backend
+	if envBackend == nil && cfg.Backend != "" {
+		if b, err := selectBackendFn(cfg.Backend); err == nil {
+			envBackend = b
+		}
+	}
+
+	if envBackend != nil {
+		baseURL, apiKey := config.ResolveBackendConfig(cfg.Backend)
+		if agentName := strings.TrimSpace(taskSpec.Agent); agentName != "" {
+			agentBackend, _, _, _, agentBaseURL, agentAPIKey, _ := config.ResolveAgentConfig(agentName)
+			if strings.EqualFold(strings.TrimSpace(agentBackend), strings.TrimSpace(cfg.Backend)) {
+				baseURL, apiKey = agentBaseURL, agentAPIKey
+			}
+		}
+		if injected := envBackend.Env(baseURL, apiKey); len(injected) > 0 {
+			cmd.SetEnv(injected)
+			// Log injected env vars with masked API keys (to file and stderr)
+			for k, v := range injected {
+				msg := fmt.Sprintf("Env: %s=%s", k, maskSensitiveValue(k, v))
+				logInfoFn(msg)
+				fmt.Fprintln(os.Stderr, "  "+msg)
+			}
+		}
 	}

 	// For backends that don't support -C flag (claude, gemini), set working directory via cmd.Dir
@@ -983,6 +1094,9 @@ func runCodexTaskWithContext(parentCtx context.Context, taskSpec TaskSpec, backe
 		if cfg.Backend == "gemini" {
 			stderrFilter = newFilteringWriter(os.Stderr, geminiNoisePatterns)
 			stderrOut = stderrFilter
+		} else if cfg.Backend == "codex" {
+			stderrFilter = newFilteringWriter(os.Stderr, codexNoisePatterns)
+			stderrOut = stderrFilter
 		}
 		stderrWriters = append([]io.Writer{stderrOut}, stderrWriters...)
 	}
@@ -1199,11 +1313,9 @@ waitLoop:
 		case parsed = <-parseCh:
 			closeWithReason(stdout, stdoutCloseReasonWait)
 		case <-messageSeen:
-			messageSeenObserved = true
 			closeWithReason(stdout, stdoutCloseReasonWait)
 			parsed = <-parseCh
 		case <-completeSeen:
-			completeSeenObserved = true
 			closeWithReason(stdout, stdoutCloseReasonWait)
 			parsed = <-parseCh
 		case <-drainTimer.C:
@@ -1273,44 +1385,13 @@ waitLoop:
 	return result
 }

-func forwardSignals(ctx context.Context, cmd commandRunner, logErrorFn func(string)) {
-	notify := signalNotifyFn
-	stop := signalStopFn
-	if notify == nil {
-		notify = signal.Notify
-	}
-	if stop == nil {
-		stop = signal.Stop
-	}
-
-	sigCh := make(chan os.Signal, 1)
-	notify(sigCh, syscall.SIGINT, syscall.SIGTERM)
-
-	go func() {
-		defer stop(sigCh)
-		select {
-		case sig := <-sigCh:
-			logErrorFn(fmt.Sprintf("Received signal: %v", sig))
-			if proc := cmd.Process(); proc != nil {
-				_ = sendTermSignal(proc)
-				time.AfterFunc(time.Duration(forceKillDelay.Load())*time.Second, func() {
-					if p := cmd.Process(); p != nil {
-						_ = p.Kill()
-					}
-				})
-			}
-		case <-ctx.Done():
-		}
-	}()
-}
-
 func cancelReason(commandName string, ctx context.Context) string {
 	if ctx == nil {
 		return "Context cancelled"
 	}

 	if commandName == "" {
-		commandName = codexCommand
+		commandName = defaultBackendName
 	}

 	if errors.Is(ctx.Err(), context.DeadlineExceeded) {
@@ -1375,20 +1456,18 @@ func terminateCommand(cmd commandRunner) *forceKillTimer {
 	return &forceKillTimer{timer: timer, done: done}
 }

-func terminateProcess(cmd commandRunner) *time.Timer {
-	if cmd == nil {
-		return nil
-	}
-	proc := cmd.Process()
-	if proc == nil {
-		return nil
-	}
-
-	_ = sendTermSignal(proc)
-
-	return time.AfterFunc(time.Duration(forceKillDelay.Load())*time.Second, func() {
-		if p := cmd.Process(); p != nil {
-			_ = p.Kill()
+// maskSensitiveValue masks sensitive values like API keys for logging.
+// Values containing "key", "token", or "secret" (case-insensitive) are masked.
+// For values longer than 8 chars: shows first 4 + **** + last 4.
+// For shorter values: shows only ****.
+func maskSensitiveValue(key, value string) string {
+	keyLower := strings.ToLower(key)
+	if strings.Contains(keyLower, "key") || strings.Contains(keyLower, "token") || strings.Contains(keyLower, "secret") {
+		if len(value) > 8 {
+			return value[:4] + "****" + value[len(value)-4:]
+		} else if len(value) > 0 {
+			return "****"
 		}
-	})
+	}
+	return value
 }
--- a/codeagent-wrapper/internal/executor/filter.go
+++ b/codeagent-wrapper/internal/executor/filter.go
@@ -1,4 +1,4 @@
-package main
+package executor

 import (
 	"bytes"
@@ -18,6 +18,12 @@ var geminiNoisePatterns = []string{
 	"YOLO mode is enabled",
 }

+// codexNoisePatterns contains stderr patterns to filter for codex backend
+var codexNoisePatterns = []string{
+	"ERROR codex_core::codex: needs_follow_up:",
+	"ERROR codex_core::skills::loader:",
+}
+
 // filteringWriter wraps an io.Writer and filters out lines matching patterns
 type filteringWriter struct {
 	w        io.Writer
@@ -39,7 +45,7 @@ func (f *filteringWriter) Write(p []byte) (n int, err error) {
 			break
 		}
 		if !f.shouldFilter(line) {
-			f.w.Write([]byte(line))
+			_, _ = f.w.Write([]byte(line))
 		}
 	}
 	return len(p), nil
@@ -59,7 +65,7 @@ func (f *filteringWriter) Flush() {
 	if f.buf.Len() > 0 {
 		remaining := f.buf.String()
 		if !f.shouldFilter(remaining) {
-			f.w.Write([]byte(remaining))
+			_, _ = f.w.Write([]byte(remaining))
 		}
 		f.buf.Reset()
 	}
--- a/codeagent-wrapper/internal/executor/filter_test.go
+++ b/codeagent-wrapper/internal/executor/filter_test.go
@@ -1,4 +1,4 @@
-package main
+package executor

 import (
 	"bytes"
@@ -48,7 +48,7 @@ func TestFilteringWriter(t *testing.T) {
 		t.Run(tt.name, func(t *testing.T) {
 			var buf bytes.Buffer
 			fw := newFilteringWriter(&buf, tt.patterns)
-			fw.Write([]byte(tt.input))
+			_, _ = fw.Write([]byte(tt.input))
 			fw.Flush()

 			if got := buf.String(); got != tt.want {
@@ -63,8 +63,8 @@ func TestFilteringWriterPartialLines(t *testing.T) {
 	fw := newFilteringWriter(&buf, geminiNoisePatterns)

 	// Write partial line
-	fw.Write([]byte("Hello "))
-	fw.Write([]byte("World\n"))
+	_, _ = fw.Write([]byte("Hello "))
+	_, _ = fw.Write([]byte("World\n"))
 	fw.Flush()

 	if got := buf.String(); got != "Hello World\n" {
--- a/codeagent-wrapper/internal/executor/log_helpers.go
+++ b/codeagent-wrapper/internal/executor/log_helpers.go
@@ -0,0 +1,124 @@
+package executor
+
+import "bytes"
+
+type logWriter struct {
+	prefix  string
+	maxLen  int
+	buf     bytes.Buffer
+	dropped bool
+}
+
+func newLogWriter(prefix string, maxLen int) *logWriter {
+	if maxLen <= 0 {
+		maxLen = codexLogLineLimit
+	}
+	return &logWriter{prefix: prefix, maxLen: maxLen}
+}
+
+func (lw *logWriter) Write(p []byte) (int, error) {
+	if lw == nil {
+		return len(p), nil
+	}
+	total := len(p)
+	for len(p) > 0 {
+		if idx := bytes.IndexByte(p, '\n'); idx >= 0 {
+			lw.writeLimited(p[:idx])
+			lw.logLine(true)
+			p = p[idx+1:]
+			continue
+		}
+		lw.writeLimited(p)
+		break
+	}
+	return total, nil
+}
+
+func (lw *logWriter) Flush() {
+	if lw == nil || lw.buf.Len() == 0 {
+		return
+	}
+	lw.logLine(false)
+}
+
+func (lw *logWriter) logLine(force bool) {
+	if lw == nil {
+		return
+	}
+	line := lw.buf.String()
+	dropped := lw.dropped
+	lw.dropped = false
+	lw.buf.Reset()
+	if line == "" && !force {
+		return
+	}
+	if lw.maxLen > 0 {
+		if dropped {
+			if lw.maxLen > 3 {
+				line = line[:min(len(line), lw.maxLen-3)] + "..."
+			} else {
+				line = line[:min(len(line), lw.maxLen)]
+			}
+		} else if len(line) > lw.maxLen {
+			cutoff := lw.maxLen
+			if cutoff > 3 {
+				line = line[:cutoff-3] + "..."
+			} else {
+				line = line[:cutoff]
+			}
+		}
+	}
+	logInfo(lw.prefix + line)
+}
+
+func (lw *logWriter) writeLimited(p []byte) {
+	if lw == nil || len(p) == 0 {
+		return
+	}
+	if lw.maxLen <= 0 {
+		lw.buf.Write(p)
+		return
+	}
+
+	remaining := lw.maxLen - lw.buf.Len()
+	if remaining <= 0 {
+		lw.dropped = true
+		return
+	}
+	if len(p) <= remaining {
+		lw.buf.Write(p)
+		return
+	}
+	lw.buf.Write(p[:remaining])
+	lw.dropped = true
+}
+
+type tailBuffer struct {
+	limit int
+	data  []byte
+}
+
+func (b *tailBuffer) Write(p []byte) (int, error) {
+	if b.limit <= 0 {
+		return len(p), nil
+	}
+
+	if len(p) >= b.limit {
+		b.data = append(b.data[:0], p[len(p)-b.limit:]...)
+		return len(p), nil
+	}
+
+	total := len(b.data) + len(p)
+	if total <= b.limit {
+		b.data = append(b.data, p...)
+		return len(p), nil
+	}
+
+	overflow := total - b.limit
+	b.data = append(b.data[overflow:], p...)
+	return len(p), nil
+}
+
+func (b *tailBuffer) String() string {
+	return string(b.data)
+}
--- a/codeagent-wrapper/internal/executor/log_writer_limit_test.go
+++ b/codeagent-wrapper/internal/executor/log_writer_limit_test.go
@@ -1,4 +1,4 @@
-package main
+package executor

 import (
 	"os"
@@ -7,14 +7,12 @@ import (
 )

 func TestLogWriterWriteLimitsBuffer(t *testing.T) {
-	defer resetTestHooks()
-
 	logger, err := NewLogger()
 	if err != nil {
 		t.Fatalf("NewLogger error: %v", err)
 	}
 	setLogger(logger)
-	defer closeLogger()
+	t.Cleanup(func() { _ = closeLogger() })

 	lw := newLogWriter("P:", 10)
 	_, _ = lw.Write([]byte(strings.Repeat("a", 100)))
--- a/codeagent-wrapper/internal/executor/parallel_config.go
+++ b/codeagent-wrapper/internal/executor/parallel_config.go
@@ -0,0 +1,135 @@
+package executor
+
+import (
+	"bytes"
+	"fmt"
+	"strings"
+
+	config "codeagent-wrapper/internal/config"
+)
+
+func ParseParallelConfig(data []byte) (*ParallelConfig, error) {
+	trimmed := bytes.TrimSpace(data)
+	if len(trimmed) == 0 {
+		return nil, fmt.Errorf("parallel config is empty")
+	}
+
+	tasks := strings.Split(string(trimmed), "---TASK---")
+	var cfg ParallelConfig
+	seen := make(map[string]struct{})
+
+	taskIndex := 0
+	for _, taskBlock := range tasks {
+		taskBlock = strings.TrimSpace(taskBlock)
+		if taskBlock == "" {
+			continue
+		}
+		taskIndex++
+
+		parts := strings.SplitN(taskBlock, "---CONTENT---", 2)
+		if len(parts) != 2 {
+			return nil, fmt.Errorf("task block #%d missing ---CONTENT--- separator", taskIndex)
+		}
+
+		meta := strings.TrimSpace(parts[0])
+		content := strings.TrimSpace(parts[1])
+
+		task := TaskSpec{WorkDir: defaultWorkdir}
+		agentSpecified := false
+		for _, line := range strings.Split(meta, "\n") {
+			line = strings.TrimSpace(line)
+			if line == "" {
+				continue
+			}
+			kv := strings.SplitN(line, ":", 2)
+			if len(kv) != 2 {
+				continue
+			}
+			key := strings.TrimSpace(kv[0])
+			value := strings.TrimSpace(kv[1])
+
+			switch key {
+			case "id":
+				task.ID = value
+			case "workdir":
+				// Validate workdir: "-" is not a valid directory
+				if value == "-" {
+					return nil, fmt.Errorf("task block #%d has invalid workdir: '-' is not a valid directory path", taskIndex)
+				}
+				task.WorkDir = value
+			case "session_id":
+				task.SessionID = value
+				task.Mode = "resume"
+			case "backend":
+				task.Backend = value
+			case "model":
+				task.Model = value
+			case "reasoning_effort":
+				task.ReasoningEffort = value
+			case "agent":
+				agentSpecified = true
+				task.Agent = value
+			case "skip_permissions", "skip-permissions":
+				if value == "" {
+					task.SkipPermissions = true
+					continue
+				}
+				task.SkipPermissions = config.ParseBoolFlag(value, false)
+			case "dependencies":
+				for _, dep := range strings.Split(value, ",") {
+					dep = strings.TrimSpace(dep)
+					if dep != "" {
+						task.Dependencies = append(task.Dependencies, dep)
+					}
+				}
+			}
+		}
+
+		if task.Mode == "" {
+			task.Mode = "new"
+		}
+
+		if agentSpecified {
+			if strings.TrimSpace(task.Agent) == "" {
+				return nil, fmt.Errorf("task block #%d has empty agent field", taskIndex)
+			}
+			if err := config.ValidateAgentName(task.Agent); err != nil {
+				return nil, fmt.Errorf("task block #%d invalid agent name: %w", taskIndex, err)
+			}
+			backend, model, promptFile, reasoning, _, _, _ := config.ResolveAgentConfig(task.Agent)
+			if task.Backend == "" {
+				task.Backend = backend
+			}
+			if task.Model == "" {
+				task.Model = model
+			}
+			if task.ReasoningEffort == "" {
+				task.ReasoningEffort = reasoning
+			}
+			task.PromptFile = promptFile
+		}
+
+		if task.ID == "" {
+			return nil, fmt.Errorf("task block #%d missing id field", taskIndex)
+		}
+		if content == "" {
+			return nil, fmt.Errorf("task block #%d (%q) missing content", taskIndex, task.ID)
+		}
+		if task.Mode == "resume" && strings.TrimSpace(task.SessionID) == "" {
+			return nil, fmt.Errorf("task block #%d (%q) has empty session_id", taskIndex, task.ID)
+		}
+		if _, exists := seen[task.ID]; exists {
+			return nil, fmt.Errorf("task block #%d has duplicate id: %s", taskIndex, task.ID)
+		}
+
+		task.Task = content
+		cfg.Tasks = append(cfg.Tasks, task)
+		seen[task.ID] = struct{}{}
+	}
+
+	if len(cfg.Tasks) == 0 {
+		return nil, fmt.Errorf("no tasks found")
+	}
+
+	return &cfg, nil
+}
--- a/codeagent-wrapper/internal/executor/prompt.go
+++ b/codeagent-wrapper/internal/executor/prompt.go
@@ -0,0 +1,130 @@
+package executor
+
+import (
+	"fmt"
+	"os"
+	"path/filepath"
+	"strings"
+)
+
+func ReadAgentPromptFile(path string, allowOutsideClaudeDir bool) (string, error) {
+	raw := strings.TrimSpace(path)
+	if raw == "" {
+		return "", nil
+	}
+
+	expanded := raw
+	if raw == "~" || strings.HasPrefix(raw, "~/") || strings.HasPrefix(raw, "~\\") {
+		home, err := os.UserHomeDir()
+		if err != nil {
+			return "", err
+		}
+		if raw == "~" {
+			expanded = home
+		} else {
+			expanded = home + raw[1:]
+		}
+	}
+
+	absPath, err := filepath.Abs(expanded)
+	if err != nil {
+		return "", err
+	}
+	absPath = filepath.Clean(absPath)
+
+	home, err := os.UserHomeDir()
+	if err != nil {
+		if !allowOutsideClaudeDir {
+			return "", err
+		}
+		logWarn(fmt.Sprintf("Failed to resolve home directory for prompt file validation: %v; proceeding without restriction", err))
+	} else {
+		allowedDirs := []string{
+			filepath.Clean(filepath.Join(home, ".claude")),
+			filepath.Clean(filepath.Join(home, ".codeagent", "agents")),
+		}
+		for i := range allowedDirs {
+			allowedAbs, err := filepath.Abs(allowedDirs[i])
+			if err == nil {
+				allowedDirs[i] = filepath.Clean(allowedAbs)
+			}
+		}
+
+		isWithinDir := func(path, dir string) bool {
+			rel, err := filepath.Rel(dir, path)
+			if err != nil {
+				return false
+			}
+			rel = filepath.Clean(rel)
+			if rel == "." {
+				return true
+			}
+			if rel == ".." {
+				return false
+			}
+			prefix := ".." + string(os.PathSeparator)
+			return !strings.HasPrefix(rel, prefix)
+		}
+
+		if !allowOutsideClaudeDir {
+			withinAllowed := false
+			for _, dir := range allowedDirs {
+				if isWithinDir(absPath, dir) {
+					withinAllowed = true
+					break
+				}
+			}
+			if !withinAllowed {
+				logWarn(fmt.Sprintf("Refusing to read prompt file outside allowed dirs (%s): %s", strings.Join(allowedDirs, ", "), absPath))
+				return "", fmt.Errorf("prompt file must be under ~/.claude or ~/.codeagent/agents")
+			}
+
+			resolvedPath, errPath := filepath.EvalSymlinks(absPath)
+			if errPath == nil {
+				resolvedPath = filepath.Clean(resolvedPath)
+				resolvedAllowed := make([]string, 0, len(allowedDirs))
+				for _, dir := range allowedDirs {
+					resolvedBase, errBase := filepath.EvalSymlinks(dir)
+					if errBase != nil {
+						continue
+					}
+					resolvedAllowed = append(resolvedAllowed, filepath.Clean(resolvedBase))
+				}
+				if len(resolvedAllowed) > 0 {
+					withinResolved := false
+					for _, dir := range resolvedAllowed {
+						if isWithinDir(resolvedPath, dir) {
+							withinResolved = true
+							break
+						}
+					}
+					if !withinResolved {
+						logWarn(fmt.Sprintf("Refusing to read prompt file outside allowed dirs (%s) (resolved): %s", strings.Join(resolvedAllowed, ", "), resolvedPath))
+						return "", fmt.Errorf("prompt file must be under ~/.claude or ~/.codeagent/agents")
+					}
+				}
+			}
+		} else {
+			withinAllowed := false
+			for _, dir := range allowedDirs {
+				if isWithinDir(absPath, dir) {
+					withinAllowed = true
+					break
+				}
+			}
+			if !withinAllowed {
+				logWarn(fmt.Sprintf("Reading prompt file outside allowed dirs (%s): %s", strings.Join(allowedDirs, ", "), absPath))
+			}
+		}
+	}
+
+	data, err := os.ReadFile(absPath)
+	if err != nil {
+		return "", err
+	}
+	return strings.TrimRight(string(data), "\r\n"), nil
+}
+
+func WrapTaskWithAgentPrompt(prompt string, task string) string {
+	return "<agent-prompt>\n" + prompt + "\n</agent-prompt>\n\n" + task
+}
--- a/codeagent-wrapper/internal/executor/prompt_file_test.go
+++ b/codeagent-wrapper/internal/executor/prompt_file_test.go
@@ -1,4 +1,4 @@
-package main
+package executor

 import (
 	"os"
@@ -9,7 +9,7 @@ import (
 )

 func TestWrapTaskWithAgentPrompt(t *testing.T) {
-	got := wrapTaskWithAgentPrompt("P", "do")
+	got := WrapTaskWithAgentPrompt("P", "do")
 	want := "<agent-prompt>\nP\n</agent-prompt>\n\ndo"
 	if got != want {
 		t.Fatalf("wrapTaskWithAgentPrompt mismatch:\n got=%q\nwant=%q", got, want)
@@ -18,7 +18,7 @@ func TestWrapTaskWithAgentPrompt(t *testing.T) {

 func TestReadAgentPromptFile_EmptyPath(t *testing.T) {
 	for _, allowOutside := range []bool{false, true} {
-		got, err := readAgentPromptFile("   ", allowOutside)
+		got, err := ReadAgentPromptFile("   ", allowOutside)
 		if err != nil {
 			t.Fatalf("unexpected error (allowOutside=%v): %v", allowOutside, err)
 		}
@@ -35,7 +35,7 @@ func TestReadAgentPromptFile_ExplicitAbsolutePath(t *testing.T) {
 		t.Fatalf("WriteFile: %v", err)
 	}

-	got, err := readAgentPromptFile(path, true)
+	got, err := ReadAgentPromptFile(path, true)
 	if err != nil {
 		t.Fatalf("readAgentPromptFile error: %v", err)
 	}
@@ -54,7 +54,7 @@ func TestReadAgentPromptFile_ExplicitTildeExpansion(t *testing.T) {
 		t.Fatalf("WriteFile: %v", err)
 	}

-	got, err := readAgentPromptFile("~/prompt.md", true)
+	got, err := ReadAgentPromptFile("~/prompt.md", true)
 	if err != nil {
 		t.Fatalf("readAgentPromptFile error: %v", err)
 	}
@@ -77,7 +77,30 @@ func TestReadAgentPromptFile_RestrictedAllowsClaudeDir(t *testing.T) {
 		t.Fatalf("WriteFile: %v", err)
 	}

-	got, err := readAgentPromptFile("~/.claude/prompt.md", false)
+	got, err := ReadAgentPromptFile("~/.claude/prompt.md", false)
+	if err != nil {
+		t.Fatalf("readAgentPromptFile error: %v", err)
+	}
+	if got != "OK" {
+		t.Fatalf("got %q, want %q", got, "OK")
+	}
+}
+
+func TestReadAgentPromptFile_RestrictedAllowsCodeagentAgentsDir(t *testing.T) {
+	home := t.TempDir()
+	t.Setenv("HOME", home)
+	t.Setenv("USERPROFILE", home)
+
+	agentDir := filepath.Join(home, ".codeagent", "agents")
+	if err := os.MkdirAll(agentDir, 0o755); err != nil {
+		t.Fatalf("MkdirAll: %v", err)
+	}
+	path := filepath.Join(agentDir, "sarsh.md")
+	if err := os.WriteFile(path, []byte("OK\n"), 0o644); err != nil {
+		t.Fatalf("WriteFile: %v", err)
+	}
+
+	got, err := ReadAgentPromptFile("~/.codeagent/agents/sarsh.md", false)
 	if err != nil {
 		t.Fatalf("readAgentPromptFile error: %v", err)
 	}
@@ -96,7 +119,7 @@ func TestReadAgentPromptFile_RestrictedRejectsOutsideClaudeDir(t *testing.T) {
 		t.Fatalf("WriteFile: %v", err)
 	}

-	if _, err := readAgentPromptFile("~/prompt.md", false); err == nil {
+	if _, err := ReadAgentPromptFile("~/prompt.md", false); err == nil {
 		t.Fatalf("expected error for prompt file outside ~/.claude, got nil")
 	}
 }
@@ -111,7 +134,7 @@ func TestReadAgentPromptFile_RestrictedRejectsTraversal(t *testing.T) {
 		t.Fatalf("WriteFile: %v", err)
 	}

-	if _, err := readAgentPromptFile("~/.claude/../secret.md", false); err == nil {
+	if _, err := ReadAgentPromptFile("~/.claude/../secret.md", false); err == nil {
 		t.Fatalf("expected traversal to be rejected, got nil")
 	}
 }
@@ -126,7 +149,7 @@ func TestReadAgentPromptFile_NotFound(t *testing.T) {
 		t.Fatalf("MkdirAll: %v", err)
 	}

-	_, err := readAgentPromptFile("~/.claude/missing.md", false)
+	_, err := ReadAgentPromptFile("~/.claude/missing.md", false)
 	if err == nil || !os.IsNotExist(err) {
 		t.Fatalf("expected not-exist error, got %v", err)
 	}
@@ -153,7 +176,7 @@ func TestReadAgentPromptFile_PermissionDenied(t *testing.T) {
 		t.Fatalf("Chmod: %v", err)
 	}

-	_, err := readAgentPromptFile("~/.claude/private.md", false)
+	_, err := ReadAgentPromptFile("~/.claude/private.md", false)
 	if err == nil {
 		t.Fatalf("expected permission error, got nil")
 	}
--- a/codeagent-wrapper/internal/executor/report_helpers.go
+++ b/codeagent-wrapper/internal/executor/report_helpers.go
@@ -0,0 +1,104 @@
+package executor
+
+import "strings"
+
+// extractCoverageGap extracts what's missing from coverage reports.
+func extractCoverageGap(message string) string {
+	if message == "" {
+		return ""
+	}
+
+	lower := strings.ToLower(message)
+	lines := strings.Split(message, "\n")
+
+	for _, line := range lines {
+		lineLower := strings.ToLower(line)
+		line = strings.TrimSpace(line)
+
+		if strings.Contains(lineLower, "uncovered") ||
+			strings.Contains(lineLower, "not covered") ||
+			strings.Contains(lineLower, "missing coverage") ||
+			strings.Contains(lineLower, "lines not covered") {
+			if len(line) > 100 {
+				return line[:97] + "..."
+			}
+			return line
+		}
+
+		if strings.Contains(lineLower, "branch") && strings.Contains(lineLower, "not taken") {
+			if len(line) > 100 {
+				return line[:97] + "..."
+			}
+			return line
+		}
+	}
+
+	if strings.Contains(lower, "function") && strings.Contains(lower, "0%") {
+		for _, line := range lines {
+			if strings.Contains(strings.ToLower(line), "0%") && strings.Contains(line, "function") {
+				line = strings.TrimSpace(line)
+				if len(line) > 100 {
+					return line[:97] + "..."
+				}
+				return line
+			}
+		}
+	}
+
+	return ""
+}
+
+// extractErrorDetail extracts meaningful error context from task output.
+func extractErrorDetail(message string, maxLen int) string {
+	if message == "" || maxLen <= 0 {
+		return ""
+	}
+
+	lines := strings.Split(message, "\n")
+	var errorLines []string
+
+	for _, line := range lines {
+		line = strings.TrimSpace(line)
+		if line == "" {
+			continue
+		}
+
+		lower := strings.ToLower(line)
+
+		if strings.HasPrefix(line, "at ") && strings.Contains(line, "(") {
+			if len(errorLines) > 0 && strings.HasPrefix(strings.ToLower(errorLines[len(errorLines)-1]), "at ") {
+				continue
+			}
+		}
+
+		if strings.Contains(lower, "error") ||
+			strings.Contains(lower, "fail") ||
+			strings.Contains(lower, "exception") ||
+			strings.Contains(lower, "assert") ||
+			strings.Contains(lower, "expected") ||
+			strings.Contains(lower, "timeout") ||
+			strings.Contains(lower, "not found") ||
+			strings.Contains(lower, "cannot") ||
+			strings.Contains(lower, "undefined") ||
+			strings.HasPrefix(line, "FAIL") ||
+			strings.HasPrefix(line, "●") {
+			errorLines = append(errorLines, line)
+		}
+	}
+
+	if len(errorLines) == 0 {
+		start := len(lines) - 5
+		if start < 0 {
+			start = 0
+		}
+		for _, line := range lines[start:] {
+			line = strings.TrimSpace(line)
+			if line != "" {
+				errorLines = append(errorLines, line)
+			}
+		}
+	}
+
+	result := strings.Join(errorLines, " | ")
+	return safeTruncate(result, maxLen)
+}
--- a/codeagent-wrapper/internal/executor/signal_unix.go
+++ b/codeagent-wrapper/internal/executor/signal_unix.go
@@ -1,7 +1,7 @@
 //go:build unix || darwin || linux
 // +build unix darwin linux

-package main
+package executor

 import (
 	"syscall"
--- a/codeagent-wrapper/internal/executor/signal_windows.go
+++ b/codeagent-wrapper/internal/executor/signal_windows.go
@@ -1,7 +1,7 @@
 //go:build windows
 // +build windows

-package main
+package executor

 import (
 	"io"
--- a/codeagent-wrapper/internal/executor/stdin.go
+++ b/codeagent-wrapper/internal/executor/stdin.go
@@ -0,0 +1,15 @@
+package executor
+
+import "strings"
+
+const stdinSpecialChars = "\n\\\"'`$"
+
+func ShouldUseStdin(taskText string, piped bool) bool {
+	if piped {
+		return true
+	}
+	if len(taskText) > 800 {
+		return true
+	}
+	return strings.ContainsAny(taskText, stdinSpecialChars)
+}
--- a/codeagent-wrapper/internal/executor/task_types.go
+++ b/codeagent-wrapper/internal/executor/task_types.go
@@ -0,0 +1,46 @@
+package executor
+
+import "context"
+
+// ParallelConfig defines the JSON schema for parallel execution.
+type ParallelConfig struct {
+	Tasks         []TaskSpec `json:"tasks"`
+	GlobalBackend string     `json:"backend,omitempty"`
+}
+
+// TaskSpec describes an individual task entry in the parallel config.
+type TaskSpec struct {
+	ID              string          `json:"id"`
+	Task            string          `json:"task"`
+	WorkDir         string          `json:"workdir,omitempty"`
+	Dependencies    []string        `json:"dependencies,omitempty"`
+	SessionID       string          `json:"session_id,omitempty"`
+	Backend         string          `json:"backend,omitempty"`
+	Model           string          `json:"model,omitempty"`
+	ReasoningEffort string          `json:"reasoning_effort,omitempty"`
+	Agent           string          `json:"agent,omitempty"`
+	PromptFile      string          `json:"prompt_file,omitempty"`
+	SkipPermissions bool            `json:"skip_permissions,omitempty"`
+	Mode            string          `json:"-"`
+	UseStdin        bool            `json:"-"`
+	Context         context.Context `json:"-"`
+}
+
+// TaskResult captures the execution outcome of a task.
+type TaskResult struct {
+	TaskID    string `json:"task_id"`
+	ExitCode  int    `json:"exit_code"`
+	Message   string `json:"message"`
+	SessionID string `json:"session_id"`
+	Error     string `json:"error"`
+	LogPath   string `json:"log_path"`
+	// Structured report fields
+	Coverage       string   `json:"coverage,omitempty"`        // extracted coverage percentage (e.g., "92%")
+	CoverageNum    float64  `json:"coverage_num,omitempty"`    // numeric coverage for comparison
+	CoverageTarget float64  `json:"coverage_target,omitempty"` // target coverage (default 90)
+	FilesChanged   []string `json:"files_changed,omitempty"`   // list of changed files
+	KeyOutput      string   `json:"key_output,omitempty"`      // brief summary of what was done
+	TestsPassed    int      `json:"tests_passed,omitempty"`    // number of tests passed
+	TestsFailed    int      `json:"tests_failed,omitempty"`    // number of tests failed
+	sharedLog      bool
+}
--- a/codeagent-wrapper/internal/executor/testhooks.go
+++ b/codeagent-wrapper/internal/executor/testhooks.go
@@ -0,0 +1,57 @@
+package executor
+
+import (
+	"context"
+	"os/exec"
+
+	backend "codeagent-wrapper/internal/backend"
+)
+
+type CommandRunner = commandRunner
+type ProcessHandle = processHandle
+
+func SetForceKillDelay(seconds int32) (restore func()) {
+	prev := forceKillDelay.Load()
+	forceKillDelay.Store(seconds)
+	return func() { forceKillDelay.Store(prev) }
+}
+
+func SetSelectBackendFn(fn func(string) (Backend, error)) (restore func()) {
+	prev := selectBackendFn
+	if fn != nil {
+		selectBackendFn = fn
+	} else {
+		selectBackendFn = backend.Select
+	}
+	return func() { selectBackendFn = prev }
+}
+
+func SetCommandContextFn(fn func(context.Context, string, ...string) *exec.Cmd) (restore func()) {
+	prev := commandContext
+	if fn != nil {
+		commandContext = fn
+	} else {
+		commandContext = exec.CommandContext
+	}
+	return func() { commandContext = prev }
+}
+
+func SetNewCommandRunner(fn func(context.Context, string, ...string) CommandRunner) (restore func()) {
+	prev := newCommandRunner
+	if fn != nil {
+		newCommandRunner = fn
+	} else {
+		newCommandRunner = func(ctx context.Context, name string, args ...string) commandRunner {
+			return &realCmd{cmd: commandContext(ctx, name, args...)}
+		}
+	}
+	return func() { newCommandRunner = prev }
+}
+
+func WithTaskLogger(ctx context.Context, logger *Logger) context.Context {
+	return withTaskLogger(ctx, logger)
+}
+
+func TaskLoggerFromContext(ctx context.Context) *Logger {
+	return taskLoggerFromContext(ctx)
+}
--- a/codeagent-wrapper/internal/logger/active.go
+++ b/codeagent-wrapper/internal/logger/active.go
@@ -0,0 +1,59 @@
+package logger
+
+import "sync/atomic"
+
+var loggerPtr atomic.Pointer[Logger]
+
+func setLogger(l *Logger) {
+	loggerPtr.Store(l)
+}
+
+func closeLogger() error {
+	logger := loggerPtr.Swap(nil)
+	if logger == nil {
+		return nil
+	}
+	return logger.Close()
+}
+
+func activeLogger() *Logger {
+	return loggerPtr.Load()
+}
+
+func logDebug(msg string) {
+	if logger := activeLogger(); logger != nil {
+		logger.Debug(msg)
+	}
+}
+
+func logInfo(msg string) {
+	if logger := activeLogger(); logger != nil {
+		logger.Info(msg)
+	}
+}
+
+func logWarn(msg string) {
+	if logger := activeLogger(); logger != nil {
+		logger.Warn(msg)
+	}
+}
+
+func logError(msg string) {
+	if logger := activeLogger(); logger != nil {
+		logger.Error(msg)
+	}
+}
+
+func SetLogger(l *Logger) { setLogger(l) }
+
+func CloseLogger() error { return closeLogger() }
+
+func ActiveLogger() *Logger { return activeLogger() }
+
+func LogInfo(msg string) { logInfo(msg) }
+
+func LogDebug(msg string) { logDebug(msg) }
+
+func LogWarn(msg string) { logWarn(msg) }
+
+func LogError(msg string) { logError(msg) }
--- a/codeagent-wrapper/internal/logger/logger.go
+++ b/codeagent-wrapper/internal/logger/logger.go
@@ -1,4 +1,4 @@
-package main
+package logger

 import (
 	"bufio"
@@ -13,6 +13,8 @@ import (
 	"sync"
 	"sync/atomic"
 	"time"
+
+	"github.com/rs/zerolog"
 )

 // Logger writes log messages asynchronously to a temp file.
@@ -22,6 +24,7 @@ type Logger struct {
 	path         string
 	file         *os.File
 	writer       *bufio.Writer
+	zlogger      zerolog.Logger
 	ch           chan logEntry
 	flushReq     chan chan struct{}
 	done         chan struct{}
@@ -37,6 +40,7 @@ type Logger struct {

 type logEntry struct {
 	msg     string
+	level   zerolog.Level
 	isError bool // true for ERROR or WARN levels
 }

@@ -73,7 +77,7 @@ func NewLogger() (*Logger, error) {
 // Useful for tests that need isolated log files within the same process.
 func NewLoggerWithSuffix(suffix string) (*Logger, error) {
 	pid := os.Getpid()
-	filename := fmt.Sprintf("%s-%d", primaryLogPrefix(), pid)
+	filename := fmt.Sprintf("%s-%d", PrimaryLogPrefix(), pid)
 	var safeSuffix string
 	if suffix != "" {
 		safeSuffix = sanitizeLogSuffix(suffix)
@@ -103,6 +107,8 @@ func NewLoggerWithSuffix(suffix string) (*Logger, error) {
 		done:     make(chan struct{}),
 	}

+	l.zlogger = zerolog.New(l.writer).With().Timestamp().Logger()
+
 	l.workerWG.Add(1)
 	go l.run()

@@ -184,17 +190,24 @@ func (l *Logger) Path() string {
 	return l.path
 }

+func (l *Logger) IsClosed() bool {
+	if l == nil {
+		return true
+	}
+	return l.closed.Load()
+}
+
 // Info logs at INFO level.
-func (l *Logger) Info(msg string) { l.log("INFO", msg) }
+func (l *Logger) Info(msg string) { l.logWithLevel(zerolog.InfoLevel, msg) }

 // Warn logs at WARN level.
-func (l *Logger) Warn(msg string) { l.log("WARN", msg) }
+func (l *Logger) Warn(msg string) { l.logWithLevel(zerolog.WarnLevel, msg) }

 // Debug logs at DEBUG level.
-func (l *Logger) Debug(msg string) { l.log("DEBUG", msg) }
+func (l *Logger) Debug(msg string) { l.logWithLevel(zerolog.DebugLevel, msg) }

 // Error logs at ERROR level.
-func (l *Logger) Error(msg string) { l.log("ERROR", msg) }
+func (l *Logger) Error(msg string) { l.logWithLevel(zerolog.ErrorLevel, msg) }

 // Close signals the worker to flush and close the log file.
 // The log file is NOT removed, allowing inspection after program exit.
@@ -335,7 +348,7 @@ func (l *Logger) Flush() {
 	}
 }

-func (l *Logger) log(level, msg string) {
+func (l *Logger) logWithLevel(entryLevel zerolog.Level, msg string) {
 	if l == nil {
 		return
 	}
@@ -343,8 +356,8 @@ func (l *Logger) log(level, msg string) {
 		return
 	}

-	isError := level == "WARN" || level == "ERROR"
-	entry := logEntry{msg: msg, isError: isError}
+	isError := entryLevel == zerolog.WarnLevel || entryLevel == zerolog.ErrorLevel
+	entry := logEntry{msg: msg, level: entryLevel, isError: isError}
 	l.flushMu.Lock()
 	l.pendingWG.Add(1)
 	l.flushMu.Unlock()
@@ -366,8 +379,7 @@ func (l *Logger) run() {
 	defer ticker.Stop()

 	writeEntry := func(entry logEntry) {
-		timestamp := time.Now().Format("2006-01-02 15:04:05.000")
-		fmt.Fprintf(l.writer, "[%s] %s\n", timestamp, entry.msg)
+		l.zlogger.WithLevel(entry.level).Msg(entry.msg)

 		// Cache error/warn entries in memory for fast extraction
 		if entry.isError {
@@ -439,10 +451,7 @@ func cleanupOldLogs() (CleanupStats, error) {
 	var stats CleanupStats
 	tempDir := os.TempDir()

-	prefixes := logPrefixes()
-	if len(prefixes) == 0 {
-		prefixes = []string{defaultWrapperName}
-	}
+	prefixes := LogPrefixes()

 	seen := make(map[string]struct{})
 	var matches []string
@@ -473,7 +482,8 @@ func cleanupOldLogs() (CleanupStats, error) {
 			stats.Kept++
 			stats.KeptFiles = append(stats.KeptFiles, filename)
 			if reason != "" {
-				logWarn(fmt.Sprintf("cleanupOldLogs: skipping %s: %s", filename, reason))
+				// Use Debug level to avoid polluting Recent Errors with cleanup noise
+				logDebug(fmt.Sprintf("cleanupOldLogs: skipping %s: %s", filename, reason))
 			}
 			continue
 		}
@@ -591,10 +601,7 @@ func isPIDReused(logPath string, pid int) bool {
 	if procStartTime.IsZero() {
 		// Can't determine process start time
 		// Check if file is very old (>7 days), likely from a dead process
-		if time.Since(fileModTime) > 7*24*time.Hour {
-			return true // File is old enough to be from a different process
-		}
-		return false // Be conservative for recent files
+		return time.Since(fileModTime) > 7*24*time.Hour
 	}

 	// If the log file was modified before the process started, PID was reused
@@ -604,10 +611,7 @@ func isPIDReused(logPath string, pid int) bool {

 func parsePIDFromLog(path string) (int, bool) {
 	name := filepath.Base(path)
-	prefixes := logPrefixes()
-	if len(prefixes) == 0 {
-		prefixes = []string{defaultWrapperName}
-	}
+	prefixes := LogPrefixes()

 	for _, prefix := range prefixes {
 		prefixWithDash := fmt.Sprintf("%s-", prefix)
@@ -661,3 +665,19 @@ func renderWorkerLimit(limit int) string {
 	}
 	return strconv.Itoa(limit)
 }
+
+func CleanupOldLogs() (CleanupStats, error) { return cleanupOldLogs() }
+
+func IsUnsafeFile(path string, tempDir string) (bool, string) { return isUnsafeFile(path, tempDir) }
+
+func IsPIDReused(logPath string, pid int) bool { return isPIDReused(logPath, pid) }
+
+func ParsePIDFromLog(path string) (int, bool) { return parsePIDFromLog(path) }
+
+func LogConcurrencyPlanning(limit, total int) { logConcurrencyPlanning(limit, total) }
+
+func LogConcurrencyState(event, taskID string, active, limit int) {
+	logConcurrencyState(event, taskID, active, limit)
+}
+
+func SanitizeLogSuffix(raw string) string { return sanitizeLogSuffix(raw) }
--- a/codeagent-wrapper/internal/logger/logger_additional_coverage_test.go
+++ b/codeagent-wrapper/internal/logger/logger_additional_coverage_test.go
@@ -1,4 +1,4 @@
-package main
+package logger

 import (
 	"fmt"
@@ -28,7 +28,7 @@ func TestLoggerConcurrencyLogHelpers(t *testing.T) {
 		t.Fatalf("NewLoggerWithSuffix error: %v", err)
 	}
 	setLogger(logger)
-	defer closeLogger()
+	defer func() { _ = closeLogger() }()

 	logConcurrencyPlanning(0, 2)
 	logConcurrencyPlanning(3, 2)
@@ -64,8 +64,8 @@ func TestLoggerConcurrencyLogHelpersNoopWithoutActiveLogger(t *testing.T) {
 func TestLoggerCleanupOldLogsSkipsUnsafeAndHandlesAlreadyDeleted(t *testing.T) {
 	tempDir := setTempDirEnv(t, t.TempDir())

-	unsafePath := createTempLog(t, tempDir, fmt.Sprintf("%s-%d.log", primaryLogPrefix(), 222))
-	orphanPath := createTempLog(t, tempDir, fmt.Sprintf("%s-%d.log", primaryLogPrefix(), 111))
+	unsafePath := createTempLog(t, tempDir, fmt.Sprintf("%s-%d.log", PrimaryLogPrefix(), 222))
+	orphanPath := createTempLog(t, tempDir, fmt.Sprintf("%s-%d.log", PrimaryLogPrefix(), 111))

 	stubFileStat(t, func(path string) (os.FileInfo, error) {
 		if path == unsafePath {
--- a/codeagent-wrapper/internal/logger/logger_suffix_test.go
+++ b/codeagent-wrapper/internal/logger/logger_suffix_test.go
@@ -1,4 +1,4 @@
-package main
+package logger

 import (
 	"fmt"
@@ -26,12 +26,12 @@ func TestLoggerWithSuffixNamingAndIsolation(t *testing.T) {
 	}
 	defer loggerB.Close()

-	wantA := filepath.Join(tempDir, fmt.Sprintf("%s-%d-%s.log", primaryLogPrefix(), os.Getpid(), taskA))
+	wantA := filepath.Join(tempDir, fmt.Sprintf("%s-%d-%s.log", PrimaryLogPrefix(), os.Getpid(), taskA))
 	if loggerA.Path() != wantA {
 		t.Fatalf("loggerA path = %q, want %q", loggerA.Path(), wantA)
 	}

-	wantB := filepath.Join(tempDir, fmt.Sprintf("%s-%d-%s.log", primaryLogPrefix(), os.Getpid(), taskB))
+	wantB := filepath.Join(tempDir, fmt.Sprintf("%s-%d-%s.log", PrimaryLogPrefix(), os.Getpid(), taskB))
 	if loggerB.Path() != wantB {
 		t.Fatalf("loggerB path = %q, want %q", loggerB.Path(), wantB)
 	}
@@ -105,7 +105,7 @@ func TestLoggerWithSuffixSanitizesUnsafeSuffix(t *testing.T) {
 		_ = os.Remove(logger.Path())
 	})

-	wantBase := fmt.Sprintf("%s-%d-%s.log", primaryLogPrefix(), os.Getpid(), safe)
+	wantBase := fmt.Sprintf("%s-%d-%s.log", PrimaryLogPrefix(), os.Getpid(), safe)
 	if gotBase := filepath.Base(logger.Path()); gotBase != wantBase {
 		t.Fatalf("log filename = %q, want %q", gotBase, wantBase)
 	}
--- a/codeagent-wrapper/internal/logger/logger_test.go
+++ b/codeagent-wrapper/internal/logger/logger_test.go
@@ -1,4 +1,4 @@
-package main
+package logger

 import (
 	"bufio"
@@ -6,7 +6,6 @@ import (
 	"fmt"
 	"math"
 	"os"
-	"os/exec"
 	"path/filepath"
 	"strconv"
 	"strings"
@@ -77,30 +76,6 @@ func TestLoggerWritesLevels(t *testing.T) {
 	}
 }

-func TestLoggerDefaultIsTerminalCoverage(t *testing.T) {
-	oldStdin := os.Stdin
-	t.Cleanup(func() { os.Stdin = oldStdin })
-
-	f, err := os.CreateTemp(t.TempDir(), "stdin-*")
-	if err != nil {
-		t.Fatalf("os.CreateTemp() error = %v", err)
-	}
-	defer os.Remove(f.Name())
-
-	os.Stdin = f
-	if got := defaultIsTerminal(); got {
-		t.Fatalf("defaultIsTerminal() = %v, want false for regular file", got)
-	}
-
-	if err := f.Close(); err != nil {
-		t.Fatalf("Close() error = %v", err)
-	}
-	os.Stdin = f
-	if got := defaultIsTerminal(); !got {
-		t.Fatalf("defaultIsTerminal() = %v, want true when Stat fails", got)
-	}
-}
-
 func TestLoggerCloseStopsWorkerAndKeepsFile(t *testing.T) {
 	tempDir := t.TempDir()
 	t.Setenv("TMPDIR", tempDir)
@@ -118,11 +93,6 @@ func TestLoggerCloseStopsWorkerAndKeepsFile(t *testing.T) {
 	if err := logger.Close(); err != nil {
 		t.Fatalf("Close() returned error: %v", err)
 	}
-	if logger.file != nil {
-		if _, err := logger.file.Write([]byte("x")); err == nil {
-			t.Fatalf("expected file to be closed after Close()")
-		}
-	}

 	// After recent changes, log file is kept for debugging - NOT removed
 	if _, err := os.Stat(logPath); os.IsNotExist(err) {
@@ -131,18 +101,6 @@ func TestLoggerCloseStopsWorkerAndKeepsFile(t *testing.T) {

 	// Clean up manually for test
 	defer os.Remove(logPath)
-
-	done := make(chan struct{})
-	go func() {
-		logger.workerWG.Wait()
-		close(done)
-	}()
-
-	select {
-	case <-done:
-	case <-time.After(200 * time.Millisecond):
-		t.Fatalf("worker goroutine did not exit after Close")
-	}
 }

 func TestLoggerConcurrentWritesSafe(t *testing.T) {
@@ -194,50 +152,13 @@ func TestLoggerConcurrentWritesSafe(t *testing.T) {
 	}
 }

-func TestLoggerTerminateProcessActive(t *testing.T) {
-	cmd := exec.Command("sleep", "5")
-	if err := cmd.Start(); err != nil {
-		t.Skipf("cannot start sleep command: %v", err)
-	}
-
-	timer := terminateProcess(&realCmd{cmd: cmd})
-	if timer == nil {
-		t.Fatalf("terminateProcess returned nil timer for active process")
-	}
-	defer timer.Stop()
-
-	done := make(chan error, 1)
-	go func() {
-		done <- cmd.Wait()
-	}()
-
-	select {
-	case <-time.After(500 * time.Millisecond):
-		t.Fatalf("process not terminated promptly")
-	case <-done:
-	}
-
-	// Force the timer callback to run immediately to cover the kill branch.
-	timer.Reset(0)
-	time.Sleep(10 * time.Millisecond)
-}
-
-func TestLoggerTerminateProcessNil(t *testing.T) {
-	if timer := terminateProcess(nil); timer != nil {
-		t.Fatalf("terminateProcess(nil) should return nil timer")
-	}
-	if timer := terminateProcess(&realCmd{cmd: &exec.Cmd{}}); timer != nil {
-		t.Fatalf("terminateProcess with nil process should return nil timer")
-	}
-}
-
 func TestLoggerCleanupOldLogsRemovesOrphans(t *testing.T) {
 	tempDir := setTempDirEnv(t, t.TempDir())

-	orphan1 := createTempLog(t, tempDir, "codex-wrapper-111.log")
-	orphan2 := createTempLog(t, tempDir, "codex-wrapper-222-suffix.log")
-	running1 := createTempLog(t, tempDir, "codex-wrapper-333.log")
-	running2 := createTempLog(t, tempDir, "codex-wrapper-444-extra-info.log")
+	orphan1 := createTempLog(t, tempDir, "codeagent-wrapper-111.log")
+	orphan2 := createTempLog(t, tempDir, "codeagent-wrapper-222-suffix.log")
+	running1 := createTempLog(t, tempDir, "codeagent-wrapper-333.log")
+	running2 := createTempLog(t, tempDir, "codeagent-wrapper-444-extra-info.log")
 	untouched := createTempLog(t, tempDir, "unrelated.log")

 	runningPIDs := map[int]bool{333: true, 444: true}
@@ -285,15 +206,15 @@ func TestLoggerCleanupOldLogsHandlesInvalidNamesAndErrors(t *testing.T) {
 	tempDir := setTempDirEnv(t, t.TempDir())

 	invalid := []string{
-		"codex-wrapper-.log",
-		"codex-wrapper.log",
-		"codex-wrapper-foo-bar.txt",
+		"codeagent-wrapper-.log",
+		"codeagent-wrapper.log",
+		"codeagent-wrapper-foo-bar.txt",
 		"not-a-codex.log",
 	}
 	for _, name := range invalid {
 		createTempLog(t, tempDir, name)
 	}
-	target := createTempLog(t, tempDir, "codex-wrapper-555-extra.log")
+	target := createTempLog(t, tempDir, "codeagent-wrapper-555-extra.log")

 	var checked []int
 	stubProcessRunning(t, func(pid int) bool {
@@ -389,8 +310,8 @@ func TestLoggerCleanupOldLogsHandlesTempDirPermissionErrors(t *testing.T) {
 	tempDir := setTempDirEnv(t, t.TempDir())

 	paths := []string{
-		createTempLog(t, tempDir, "codex-wrapper-6100.log"),
-		createTempLog(t, tempDir, "codex-wrapper-6101.log"),
+		createTempLog(t, tempDir, "codeagent-wrapper-6100.log"),
+		createTempLog(t, tempDir, "codeagent-wrapper-6101.log"),
 	}

 	stubProcessRunning(t, func(int) bool { return false })
@@ -428,8 +349,8 @@ func TestLoggerCleanupOldLogsHandlesTempDirPermissionErrors(t *testing.T) {
 func TestLoggerCleanupOldLogsHandlesPermissionDeniedFile(t *testing.T) {
 	tempDir := setTempDirEnv(t, t.TempDir())

-	protected := createTempLog(t, tempDir, "codex-wrapper-6200.log")
-	deletable := createTempLog(t, tempDir, "codex-wrapper-6201.log")
+	protected := createTempLog(t, tempDir, "codeagent-wrapper-6200.log")
+	deletable := createTempLog(t, tempDir, "codeagent-wrapper-6201.log")

 	stubProcessRunning(t, func(int) bool { return false })
 	stubProcessStartTime(t, func(int) time.Time { return time.Time{} })
@@ -468,7 +389,7 @@ func TestLoggerCleanupOldLogsPerformanceBound(t *testing.T) {
 	const fileCount = 400
 	fakePaths := make([]string, fileCount)
 	for i := 0; i < fileCount; i++ {
-		name := fmt.Sprintf("codex-wrapper-%d.log", 10000+i)
+		name := fmt.Sprintf("codeagent-wrapper-%d.log", 10000+i)
 		fakePaths[i] = createTempLog(t, tempDir, name)
 	}

@@ -505,102 +426,11 @@ func TestLoggerCleanupOldLogsPerformanceBound(t *testing.T) {
 	}
 }

-func TestLoggerCleanupOldLogsCoverageSuite(t *testing.T) {
-	TestBackendParseJSONStream_CoverageSuite(t)
-}
-
-// Reuse the existing coverage suite so the focused TestLogger run still exercises
-// the rest of the codebase and keeps coverage high.
-func TestLoggerCoverageSuite(t *testing.T) {
-	suite := []struct {
-		name string
-		fn   func(*testing.T)
-	}{
-		{"TestBackendParseJSONStream_CoverageSuite", TestBackendParseJSONStream_CoverageSuite},
-		{"TestVersionCoverageFullRun", TestVersionCoverageFullRun},
-		{"TestVersionMainWrapper", TestVersionMainWrapper},
-
-		{"TestExecutorHelperCoverage", TestExecutorHelperCoverage},
-		{"TestExecutorRunCodexTaskWithContext", TestExecutorRunCodexTaskWithContext},
-		{"TestExecutorParallelLogIsolation", TestExecutorParallelLogIsolation},
-		{"TestExecutorTaskLoggerContext", TestExecutorTaskLoggerContext},
-		{"TestExecutorExecuteConcurrentWithContextBranches", TestExecutorExecuteConcurrentWithContextBranches},
-		{"TestExecutorSignalAndTermination", TestExecutorSignalAndTermination},
-		{"TestExecutorCancelReasonAndCloseWithReason", TestExecutorCancelReasonAndCloseWithReason},
-		{"TestExecutorForceKillTimerStop", TestExecutorForceKillTimerStop},
-		{"TestExecutorForwardSignalsDefaults", TestExecutorForwardSignalsDefaults},
-
-		{"TestBackendParseArgs_NewMode", TestBackendParseArgs_NewMode},
-		{"TestBackendParseArgs_ResumeMode", TestBackendParseArgs_ResumeMode},
-		{"TestBackendParseArgs_BackendFlag", TestBackendParseArgs_BackendFlag},
-		{"TestBackendParseArgs_SkipPermissions", TestBackendParseArgs_SkipPermissions},
-		{"TestBackendParseBoolFlag", TestBackendParseBoolFlag},
-		{"TestBackendEnvFlagEnabled", TestBackendEnvFlagEnabled},
-		{"TestRunResolveTimeout", TestRunResolveTimeout},
-		{"TestRunIsTerminal", TestRunIsTerminal},
-		{"TestRunReadPipedTask", TestRunReadPipedTask},
-		{"TestTailBufferWrite", TestTailBufferWrite},
-		{"TestLogWriterWriteLimitsBuffer", TestLogWriterWriteLimitsBuffer},
-		{"TestLogWriterLogLine", TestLogWriterLogLine},
-		{"TestNewLogWriterDefaultMaxLen", TestNewLogWriterDefaultMaxLen},
-		{"TestNewLogWriterDefaultLimit", TestNewLogWriterDefaultLimit},
-		{"TestRunHello", TestRunHello},
-		{"TestRunGreet", TestRunGreet},
-		{"TestRunFarewell", TestRunFarewell},
-		{"TestRunFarewellEmpty", TestRunFarewellEmpty},
-
-		{"TestParallelParseConfig_Success", TestParallelParseConfig_Success},
-		{"TestParallelParseConfig_Backend", TestParallelParseConfig_Backend},
-		{"TestParallelParseConfig_InvalidFormat", TestParallelParseConfig_InvalidFormat},
-		{"TestParallelParseConfig_EmptyTasks", TestParallelParseConfig_EmptyTasks},
-		{"TestParallelParseConfig_MissingID", TestParallelParseConfig_MissingID},
-		{"TestParallelParseConfig_MissingTask", TestParallelParseConfig_MissingTask},
-		{"TestParallelParseConfig_DuplicateID", TestParallelParseConfig_DuplicateID},
-		{"TestParallelParseConfig_DelimiterFormat", TestParallelParseConfig_DelimiterFormat},
-
-		{"TestBackendSelectBackend", TestBackendSelectBackend},
-		{"TestBackendSelectBackend_Invalid", TestBackendSelectBackend_Invalid},
-		{"TestBackendSelectBackend_DefaultOnEmpty", TestBackendSelectBackend_DefaultOnEmpty},
-		{"TestBackendBuildArgs_CodexBackend", TestBackendBuildArgs_CodexBackend},
-		{"TestBackendBuildArgs_ClaudeBackend", TestBackendBuildArgs_ClaudeBackend},
-		{"TestClaudeBackendBuildArgs_OutputValidation", TestClaudeBackendBuildArgs_OutputValidation},
-		{"TestBackendBuildArgs_GeminiBackend", TestBackendBuildArgs_GeminiBackend},
-		{"TestGeminiBackendBuildArgs_OutputValidation", TestGeminiBackendBuildArgs_OutputValidation},
-		{"TestBackendNamesAndCommands", TestBackendNamesAndCommands},
-
-		{"TestBackendParseJSONStream", TestBackendParseJSONStream},
-		{"TestBackendParseJSONStream_ClaudeEvents", TestBackendParseJSONStream_ClaudeEvents},
-		{"TestBackendParseJSONStream_GeminiEvents", TestBackendParseJSONStream_GeminiEvents},
-		{"TestBackendParseJSONStreamWithWarn_InvalidLine", TestBackendParseJSONStreamWithWarn_InvalidLine},
-		{"TestBackendParseJSONStream_OnMessage", TestBackendParseJSONStream_OnMessage},
-		{"TestBackendParseJSONStream_ScannerError", TestBackendParseJSONStream_ScannerError},
-		{"TestBackendDiscardInvalidJSON", TestBackendDiscardInvalidJSON},
-		{"TestBackendDiscardInvalidJSONBuffer", TestBackendDiscardInvalidJSONBuffer},
-
-		{"TestCurrentWrapperNameFallsBackToExecutable", TestCurrentWrapperNameFallsBackToExecutable},
-		{"TestCurrentWrapperNameDetectsLegacyAliasSymlink", TestCurrentWrapperNameDetectsLegacyAliasSymlink},
-
-		{"TestIsProcessRunning", TestIsProcessRunning},
-		{"TestGetProcessStartTimeReadsProcStat", TestGetProcessStartTimeReadsProcStat},
-		{"TestGetProcessStartTimeInvalidData", TestGetProcessStartTimeInvalidData},
-		{"TestGetBootTimeParsesBtime", TestGetBootTimeParsesBtime},
-		{"TestGetBootTimeInvalidData", TestGetBootTimeInvalidData},
-
-		{"TestClaudeBuildArgs_ModesAndPermissions", TestClaudeBuildArgs_ModesAndPermissions},
-		{"TestClaudeBuildArgs_GeminiAndCodexModes", TestClaudeBuildArgs_GeminiAndCodexModes},
-		{"TestClaudeBuildArgs_BackendMetadata", TestClaudeBuildArgs_BackendMetadata},
-	}
-
-	for _, tc := range suite {
-		t.Run(tc.name, tc.fn)
-	}
-}
-
 func TestLoggerCleanupOldLogsKeepsCurrentProcessLog(t *testing.T) {
 	tempDir := setTempDirEnv(t, t.TempDir())

 	currentPID := os.Getpid()
-	currentLog := createTempLog(t, tempDir, fmt.Sprintf("codex-wrapper-%d.log", currentPID))
+	currentLog := createTempLog(t, tempDir, fmt.Sprintf("codeagent-wrapper-%d.log", currentPID))

 	stubProcessRunning(t, func(pid int) bool {
 		if pid != currentPID {
@@ -676,7 +506,7 @@ func TestLoggerIsUnsafeFileSecurityChecks(t *testing.T) {
 		stubEvalSymlinks(t, func(path string) (string, error) {
 			return filepath.Join(absTempDir, filepath.Base(path)), nil
 		})
-		unsafe, reason := isUnsafeFile(filepath.Join(absTempDir, "codex-wrapper-1.log"), tempDir)
+		unsafe, reason := isUnsafeFile(filepath.Join(absTempDir, "codeagent-wrapper-1.log"), tempDir)
 		if !unsafe || reason != "refusing to delete symlink" {
 			t.Fatalf("expected symlink to be rejected, got unsafe=%v reason=%q", unsafe, reason)
 		}
@@ -702,9 +532,9 @@ func TestLoggerIsUnsafeFileSecurityChecks(t *testing.T) {
 		})
 		otherDir := t.TempDir()
 		stubEvalSymlinks(t, func(string) (string, error) {
-			return filepath.Join(otherDir, "codex-wrapper-9.log"), nil
+			return filepath.Join(otherDir, "codeagent-wrapper-9.log"), nil
 		})
-		unsafe, reason := isUnsafeFile(filepath.Join(otherDir, "codex-wrapper-9.log"), tempDir)
+		unsafe, reason := isUnsafeFile(filepath.Join(otherDir, "codeagent-wrapper-9.log"), tempDir)
 		if !unsafe || reason != "file is outside tempDir" {
 			t.Fatalf("expected outside file to be rejected, got unsafe=%v reason=%q", unsafe, reason)
 		}
@@ -713,15 +543,21 @@ func TestLoggerIsUnsafeFileSecurityChecks(t *testing.T) {

 func TestLoggerPathAndRemove(t *testing.T) {
 	tempDir := t.TempDir()
-	path := filepath.Join(tempDir, "sample.log")
-	if err := os.WriteFile(path, []byte("test"), 0o644); err != nil {
-		t.Fatalf("failed to create temp file: %v", err)
+	t.Setenv("TMPDIR", tempDir)
+
+	logger, err := NewLoggerWithSuffix("sample")
+	if err != nil {
+		t.Fatalf("NewLoggerWithSuffix() error = %v", err)
+	}
+	path := logger.Path()
+	if path == "" {
+		_ = logger.Close()
+		t.Fatalf("logger.Path() returned empty path")
+	}
+	if err := logger.Close(); err != nil {
+		t.Fatalf("Close() error = %v", err)
 	}

-	logger := &Logger{path: path}
-	if got := logger.Path(); got != path {
-		t.Fatalf("Path() = %q, want %q", got, path)
-	}
 	if err := logger.RemoveLogFile(); err != nil {
 		t.Fatalf("RemoveLogFile() error = %v", err)
 	}
@@ -738,43 +574,6 @@ func TestLoggerPathAndRemove(t *testing.T) {
 	}
 }

-func TestLoggerTruncateBytesCoverage(t *testing.T) {
-	if got := truncateBytes([]byte("abc"), 3); got != "abc" {
-		t.Fatalf("truncateBytes() = %q, want %q", got, "abc")
-	}
-	if got := truncateBytes([]byte("abcd"), 3); got != "abc..." {
-		t.Fatalf("truncateBytes() = %q, want %q", got, "abc...")
-	}
-	if got := truncateBytes([]byte("abcd"), -1); got != "" {
-		t.Fatalf("truncateBytes() = %q, want empty string", got)
-	}
-}
-
-func TestLoggerInternalLog(t *testing.T) {
-	logger := &Logger{
-		ch:        make(chan logEntry, 1),
-		done:      make(chan struct{}),
-		pendingWG: sync.WaitGroup{},
-	}
-
-	done := make(chan logEntry, 1)
-	go func() {
-		entry := <-logger.ch
-		logger.pendingWG.Done()
-		done <- entry
-	}()
-
-	logger.log("INFO", "hello")
-	entry := <-done
-	if entry.msg != "hello" {
-		t.Fatalf("unexpected entry %+v", entry)
-	}
-
-	logger.closed.Store(true)
-	logger.log("INFO", "ignored")
-	close(logger.done)
-}
-
 func TestLoggerParsePIDFromLog(t *testing.T) {
 	hugePID := strconv.FormatInt(math.MaxInt64, 10) + "0"
 	tests := []struct {
@@ -782,13 +581,13 @@ func TestLoggerParsePIDFromLog(t *testing.T) {
 		pid  int
 		ok   bool
 	}{
-		{"codex-wrapper-123.log", 123, true},
-		{"codex-wrapper-999-extra.log", 999, true},
-		{"codex-wrapper-.log", 0, false},
+		{"codeagent-wrapper-123.log", 123, true},
+		{"codeagent-wrapper-999-extra.log", 999, true},
+		{"codeagent-wrapper-.log", 0, false},
 		{"invalid-name.log", 0, false},
-		{"codex-wrapper--5.log", 0, false},
-		{"codex-wrapper-0.log", 0, false},
-		{fmt.Sprintf("codex-wrapper-%s.log", hugePID), 0, false},
+		{"codeagent-wrapper--5.log", 0, false},
+		{"codeagent-wrapper-0.log", 0, false},
+		{fmt.Sprintf("codeagent-wrapper-%s.log", hugePID), 0, false},
 	}

 	for _, tt := range tests {
@@ -827,56 +626,32 @@ func setTempDirEnv(t *testing.T, dir string) string {

 func stubProcessRunning(t *testing.T, fn func(int) bool) {
 	t.Helper()
-	original := processRunningCheck
-	processRunningCheck = fn
-	t.Cleanup(func() {
-		processRunningCheck = original
-	})
+	t.Cleanup(SetProcessRunningCheck(fn))
 }

 func stubProcessStartTime(t *testing.T, fn func(int) time.Time) {
 	t.Helper()
-	original := processStartTimeFn
-	processStartTimeFn = fn
-	t.Cleanup(func() {
-		processStartTimeFn = original
-	})
+	t.Cleanup(SetProcessStartTimeFn(fn))
 }

 func stubRemoveLogFile(t *testing.T, fn func(string) error) {
 	t.Helper()
-	original := removeLogFileFn
-	removeLogFileFn = fn
-	t.Cleanup(func() {
-		removeLogFileFn = original
-	})
+	t.Cleanup(SetRemoveLogFileFn(fn))
 }

 func stubGlobLogFiles(t *testing.T, fn func(string) ([]string, error)) {
 	t.Helper()
-	original := globLogFiles
-	globLogFiles = fn
-	t.Cleanup(func() {
-		globLogFiles = original
-	})
+	t.Cleanup(SetGlobLogFilesFn(fn))
 }

 func stubFileStat(t *testing.T, fn func(string) (os.FileInfo, error)) {
 	t.Helper()
-	original := fileStatFn
-	fileStatFn = fn
-	t.Cleanup(func() {
-		fileStatFn = original
-	})
+	t.Cleanup(SetFileStatFn(fn))
 }

 func stubEvalSymlinks(t *testing.T, fn func(string) (string, error)) {
 	t.Helper()
-	original := evalSymlinksFn
-	evalSymlinksFn = fn
-	t.Cleanup(func() {
-		evalSymlinksFn = original
-	})
+	t.Cleanup(SetEvalSymlinksFn(fn))
 }

 type fakeFileInfo struct {
@@ -960,7 +735,7 @@ func TestLoggerExtractRecentErrors(t *testing.T) {
 				t.Fatalf("NewLoggerWithSuffix() error = %v", err)
 			}
 			defer logger.Close()
-			defer logger.RemoveLogFile()
+			defer func() { _ = logger.RemoveLogFile() }()

 			// Write logs using logger methods
 			for _, entry := range tt.logs {
@@ -1000,14 +775,14 @@ func TestLoggerExtractRecentErrorsNilLogger(t *testing.T) {
 }

 func TestLoggerExtractRecentErrorsEmptyPath(t *testing.T) {
-	logger := &Logger{path: ""}
+	logger := &Logger{}
 	if got := logger.ExtractRecentErrors(10); got != nil {
 		t.Fatalf("empty path ExtractRecentErrors() should return nil, got %v", got)
 	}
 }

 func TestLoggerExtractRecentErrorsFileNotExist(t *testing.T) {
-	logger := &Logger{path: "/nonexistent/path/to/log.log"}
+	logger := &Logger{}
 	if got := logger.ExtractRecentErrors(10); got != nil {
 		t.Fatalf("nonexistent file ExtractRecentErrors() should return nil, got %v", got)
 	}
@@ -1049,7 +824,7 @@ func TestExtractRecentErrorsBoundaryCheck(t *testing.T) {
 		t.Fatalf("NewLoggerWithSuffix() error = %v", err)
 	}
 	defer logger.Close()
-	defer logger.RemoveLogFile()
+	defer func() { _ = logger.RemoveLogFile() }()

 	// Write some errors
 	logger.Error("error 1")
@@ -1082,7 +857,7 @@ func TestErrorEntriesMaxLimit(t *testing.T) {
 		t.Fatalf("NewLoggerWithSuffix() error = %v", err)
 	}
 	defer logger.Close()
-	defer logger.RemoveLogFile()
+	defer func() { _ = logger.RemoveLogFile() }()

 	// Write 150 error/warn entries
 	for i := 1; i <= 150; i++ {
--- a/codeagent-wrapper/internal/logger/process_check.go
+++ b/codeagent-wrapper/internal/logger/process_check.go
@@ -0,0 +1,63 @@
+package logger
+
+import (
+	"errors"
+	"math"
+	"time"
+
+	"github.com/shirou/gopsutil/v3/process"
+)
+
+func pidToInt32(pid int) (int32, bool) {
+	if pid <= 0 || pid > math.MaxInt32 {
+		return 0, false
+	}
+	return int32(pid), true
+}
+
+// isProcessRunning reports whether a process with the given pid appears to be running.
+// It is intentionally conservative on errors to avoid deleting logs for live processes.
+func isProcessRunning(pid int) bool {
+	pid32, ok := pidToInt32(pid)
+	if !ok {
+		return false
+	}
+
+	exists, err := process.PidExists(pid32)
+	if err == nil {
+		return exists
+	}
+
+	// If we can positively identify that the process doesn't exist, report false.
+	if errors.Is(err, process.ErrorProcessNotRunning) {
+		return false
+	}
+
+	// Permission/inspection failures: assume it's running to be safe.
+	return true
+}
+
+// getProcessStartTime returns the start time of a process.
+// Returns zero time if the start time cannot be determined.
+func getProcessStartTime(pid int) time.Time {
+	pid32, ok := pidToInt32(pid)
+	if !ok {
+		return time.Time{}
+	}
+
+	proc, err := process.NewProcess(pid32)
+	if err != nil {
+		return time.Time{}
+	}
+
+	ms, err := proc.CreateTime()
+	if err != nil || ms <= 0 {
+		return time.Time{}
+	}
+
+	return time.UnixMilli(ms)
+}
+
+func IsProcessRunning(pid int) bool { return isProcessRunning(pid) }
+
+func GetProcessStartTime(pid int) time.Time { return getProcessStartTime(pid) }
--- a/codeagent-wrapper/internal/logger/process_check_test.go
+++ b/codeagent-wrapper/internal/logger/process_check_test.go
@@ -0,0 +1,112 @@
+package logger
+
+import (
+	"math"
+	"os"
+	"os/exec"
+	"runtime"
+	"strconv"
+	"testing"
+	"time"
+)
+
+func TestIsProcessRunning(t *testing.T) {
+	t.Run("boundary values", func(t *testing.T) {
+		if isProcessRunning(0) {
+			t.Fatalf("pid 0 should never be treated as running")
+		}
+		if isProcessRunning(-1) {
+			t.Fatalf("negative pid should never be treated as running")
+		}
+	})
+
+	t.Run("pid out of int32 range", func(t *testing.T) {
+		if strconv.IntSize <= 32 {
+			t.Skip("int cannot represent values above int32 range")
+		}
+
+		pid := int(int64(math.MaxInt32) + 1)
+		if isProcessRunning(pid) {
+			t.Fatalf("expected pid %d (out of int32 range) to be treated as not running", pid)
+		}
+	})
+
+	t.Run("current process", func(t *testing.T) {
+		if !isProcessRunning(os.Getpid()) {
+			t.Fatalf("expected current process (pid=%d) to be running", os.Getpid())
+		}
+	})
+
+	t.Run("fake pid", func(t *testing.T) {
+		const nonexistentPID = 1 << 30
+		if isProcessRunning(nonexistentPID) {
+			t.Fatalf("expected pid %d to be reported as not running", nonexistentPID)
+		}
+	})
+
+	t.Run("terminated process", func(t *testing.T) {
+		pid := exitedProcessPID(t)
+		if isProcessRunning(pid) {
+			t.Fatalf("expected exited child process (pid=%d) to be reported as not running", pid)
+		}
+	})
+}
+
+func exitedProcessPID(t *testing.T) int {
+	t.Helper()
+
+	var cmd *exec.Cmd
+	if runtime.GOOS == "windows" {
+		cmd = exec.Command("cmd", "/c", "exit 0")
+	} else {
+		cmd = exec.Command("sh", "-c", "exit 0")
+	}
+
+	if err := cmd.Start(); err != nil {
+		t.Fatalf("failed to start helper process: %v", err)
+	}
+	pid := cmd.Process.Pid
+
+	if err := cmd.Wait(); err != nil {
+		t.Fatalf("helper process did not exit cleanly: %v", err)
+	}
+
+	time.Sleep(50 * time.Millisecond)
+	return pid
+}
+
+func TestGetProcessStartTimeReadsProcStat(t *testing.T) {
+	start := getProcessStartTime(os.Getpid())
+	if start.IsZero() {
+		t.Fatalf("expected non-zero start time for current process")
+	}
+	if start.After(time.Now().Add(5 * time.Second)) {
+		t.Fatalf("start time is unexpectedly in the future: %v", start)
+	}
+}
+
+func TestGetProcessStartTimeInvalidData(t *testing.T) {
+	if !getProcessStartTime(0).IsZero() {
+		t.Fatalf("expected zero time for pid 0")
+	}
+	if !getProcessStartTime(-1).IsZero() {
+		t.Fatalf("expected zero time for negative pid")
+	}
+	if !getProcessStartTime(1 << 30).IsZero() {
+		t.Fatalf("expected zero time for non-existent pid")
+	}
+	if strconv.IntSize > 32 {
+		pid := int(int64(math.MaxInt32) + 1)
+		if !getProcessStartTime(pid).IsZero() {
+			t.Fatalf("expected zero time for pid %d (out of int32 range)", pid)
+		}
+	}
+}
+
+func TestGetBootTimeParsesBtime(t *testing.T) {
+	t.Skip("legacy boot-time probing removed; start time now uses gopsutil")
+}
+
+func TestGetBootTimeInvalidData(t *testing.T) {
+	t.Skip("legacy boot-time probing removed; start time now uses gopsutil")
+}
--- a/codeagent-wrapper/internal/logger/testhooks.go
+++ b/codeagent-wrapper/internal/logger/testhooks.go
@@ -0,0 +1,67 @@
+package logger
+
+import (
+	"os"
+	"path/filepath"
+	"time"
+)
+
+func SetProcessRunningCheck(fn func(int) bool) (restore func()) {
+	prev := processRunningCheck
+	if fn != nil {
+		processRunningCheck = fn
+	} else {
+		processRunningCheck = isProcessRunning
+	}
+	return func() { processRunningCheck = prev }
+}
+
+func SetProcessStartTimeFn(fn func(int) time.Time) (restore func()) {
+	prev := processStartTimeFn
+	if fn != nil {
+		processStartTimeFn = fn
+	} else {
+		processStartTimeFn = getProcessStartTime
+	}
+	return func() { processStartTimeFn = prev }
+}
+
+func SetRemoveLogFileFn(fn func(string) error) (restore func()) {
+	prev := removeLogFileFn
+	if fn != nil {
+		removeLogFileFn = fn
+	} else {
+		removeLogFileFn = os.Remove
+	}
+	return func() { removeLogFileFn = prev }
+}
+
+func SetGlobLogFilesFn(fn func(string) ([]string, error)) (restore func()) {
+	prev := globLogFiles
+	if fn != nil {
+		globLogFiles = fn
+	} else {
+		globLogFiles = filepath.Glob
+	}
+	return func() { globLogFiles = prev }
+}
+
+func SetFileStatFn(fn func(string) (os.FileInfo, error)) (restore func()) {
+	prev := fileStatFn
+	if fn != nil {
+		fileStatFn = fn
+	} else {
+		fileStatFn = os.Lstat
+	}
+	return func() { fileStatFn = prev }
+}
+
+func SetEvalSymlinksFn(fn func(string) (string, error)) (restore func()) {
+	prev := evalSymlinksFn
+	if fn != nil {
+		evalSymlinksFn = fn
+	} else {
+		evalSymlinksFn = filepath.EvalSymlinks
+	}
+	return func() { evalSymlinksFn = prev }
+}
--- a/codeagent-wrapper/internal/logger/wrapper_name.go
+++ b/codeagent-wrapper/internal/logger/wrapper_name.go
@@ -0,0 +1,13 @@
+package logger
+
+// WrapperName is the fixed name for this tool.
+const WrapperName = "codeagent-wrapper"
+
+// CurrentWrapperName returns the wrapper name (always "codeagent-wrapper").
+func CurrentWrapperName() string { return WrapperName }
+
+// LogPrefixes returns the log file name prefixes to look for.
+func LogPrefixes() []string { return []string{WrapperName} }
+
+// PrimaryLogPrefix returns the preferred filename prefix for log files.
+func PrimaryLogPrefix() string { return WrapperName }
--- a/codeagent-wrapper/internal/parser/event.go
+++ b/codeagent-wrapper/internal/parser/event.go
@@ -0,0 +1,74 @@
+package parser
+
+import "github.com/goccy/go-json"
+
+// JSONEvent represents a Codex JSON output event.
+type JSONEvent struct {
+	Type     string     `json:"type"`
+	ThreadID string     `json:"thread_id,omitempty"`
+	Item     *EventItem `json:"item,omitempty"`
+}
+
+// EventItem represents the item field in a JSON event.
+type EventItem struct {
+	Type string      `json:"type"`
+	Text interface{} `json:"text"`
+}
+
+// ClaudeEvent for Claude stream-json format.
+type ClaudeEvent struct {
+	Type      string `json:"type"`
+	Subtype   string `json:"subtype,omitempty"`
+	SessionID string `json:"session_id,omitempty"`
+	Result    string `json:"result,omitempty"`
+}
+
+// GeminiEvent for Gemini stream-json format.
+type GeminiEvent struct {
+	Type      string `json:"type"`
+	SessionID string `json:"session_id,omitempty"`
+	Role      string `json:"role,omitempty"`
+	Content   string `json:"content,omitempty"`
+	Delta     bool   `json:"delta,omitempty"`
+	Status    string `json:"status,omitempty"`
+}
+
+// UnifiedEvent combines all backend event formats into a single structure
+// to avoid multiple JSON unmarshal operations per event.
+type UnifiedEvent struct {
+	// Common fields
+	Type string `json:"type"`
+
+	// Codex-specific fields
+	ThreadID string          `json:"thread_id,omitempty"`
+	Item     json.RawMessage `json:"item,omitempty"` // Lazy parse
+
+	// Claude-specific fields
+	Subtype   string `json:"subtype,omitempty"`
+	SessionID string `json:"session_id,omitempty"`
+	Result    string `json:"result,omitempty"`
+
+	// Gemini-specific fields
+	Role    string `json:"role,omitempty"`
+	Content string `json:"content,omitempty"`
+	Delta   *bool  `json:"delta,omitempty"`
+	Status  string `json:"status,omitempty"`
+
+	// Opencode-specific fields (camelCase sessionID)
+	OpencodeSessionID string          `json:"sessionID,omitempty"`
+	Part              json.RawMessage `json:"part,omitempty"`
+}
+
+// OpencodePart represents the part field in opencode events.
+type OpencodePart struct {
+	Type      string `json:"type"`
+	Text      string `json:"text,omitempty"`
+	Reason    string `json:"reason,omitempty"`
+	SessionID string `json:"sessionID,omitempty"`
+}
+
+// ItemContent represents the parsed item.text field for Codex events.
+type ItemContent struct {
+	Type string      `json:"type"`
+	Text interface{} `json:"text"`
+}
--- a/codeagent-wrapper/internal/parser/parser.go
+++ b/codeagent-wrapper/internal/parser/parser.go
@@ -1,106 +1,65 @@
-package main
+package parser

 import (
 	"bufio"
 	"bytes"
-	"encoding/json"
 	"errors"
 	"fmt"
 	"io"
 	"strings"
+	"sync"
+
+	"github.com/goccy/go-json"
 )

-// JSONEvent represents a Codex JSON output event
-type JSONEvent struct {
-	Type     string     `json:"type"`
-	ThreadID string     `json:"thread_id,omitempty"`
-	Item     *EventItem `json:"item,omitempty"`
-}
-
-// EventItem represents the item field in a JSON event
-type EventItem struct {
-	Type string      `json:"type"`
-	Text interface{} `json:"text"`
-}
-
-// ClaudeEvent for Claude stream-json format
-type ClaudeEvent struct {
-	Type      string `json:"type"`
-	Subtype   string `json:"subtype,omitempty"`
-	SessionID string `json:"session_id,omitempty"`
-	Result    string `json:"result,omitempty"`
-}
-
-// GeminiEvent for Gemini stream-json format
-type GeminiEvent struct {
-	Type      string `json:"type"`
-	SessionID string `json:"session_id,omitempty"`
-	Role      string `json:"role,omitempty"`
-	Content   string `json:"content,omitempty"`
-	Delta     bool   `json:"delta,omitempty"`
-	Status    string `json:"status,omitempty"`
-}
-
-func parseJSONStream(r io.Reader) (message, threadID string) {
-	return parseJSONStreamWithLog(r, logWarn, logInfo)
-}
-
-func parseJSONStreamWithWarn(r io.Reader, warnFn func(string)) (message, threadID string) {
-	return parseJSONStreamWithLog(r, warnFn, logInfo)
-}
-
-func parseJSONStreamWithLog(r io.Reader, warnFn func(string), infoFn func(string)) (message, threadID string) {
-	return parseJSONStreamInternal(r, warnFn, infoFn, nil, nil)
-}
-
 const (
 	jsonLineReaderSize   = 64 * 1024
 	jsonLineMaxBytes     = 10 * 1024 * 1024
 	jsonLinePreviewBytes = 256
 )

-// UnifiedEvent combines all backend event formats into a single structure
-// to avoid multiple JSON unmarshal operations per event
-type UnifiedEvent struct {
-	// Common fields
-	Type string `json:"type"`
-
-	// Codex-specific fields
-	ThreadID string          `json:"thread_id,omitempty"`
-	Item     json.RawMessage `json:"item,omitempty"` // Lazy parse
-
-	// Claude-specific fields
-	Subtype   string `json:"subtype,omitempty"`
-	SessionID string `json:"session_id,omitempty"`
-	Result    string `json:"result,omitempty"`
-
-	// Gemini-specific fields
-	Role    string `json:"role,omitempty"`
-	Content string `json:"content,omitempty"`
-	Delta   *bool  `json:"delta,omitempty"`
-	Status  string `json:"status,omitempty"`
-
-	// Opencode-specific fields (camelCase sessionID)
-	OpencodeSessionID string          `json:"sessionID,omitempty"`
-	Part              json.RawMessage `json:"part,omitempty"`
+type lineScratch struct {
+	buf     []byte
+	preview []byte
 }

-// OpencodePart represents the part field in opencode events
-type OpencodePart struct {
-	Type      string `json:"type"`
-	Text      string `json:"text,omitempty"`
-	Reason    string `json:"reason,omitempty"`
-	SessionID string `json:"sessionID,omitempty"`
+const maxPooledLineScratchCap = 1 << 20 // 1 MiB
+
+var lineScratchPool = sync.Pool{
+	New: func() any {
+		return &lineScratch{
+			buf:     make([]byte, 0, jsonLineReaderSize),
+			preview: make([]byte, 0, jsonLinePreviewBytes),
+		}
+	},
 }

-// ItemContent represents the parsed item.text field for Codex events
-type ItemContent struct {
-	Type string      `json:"type"`
-	Text interface{} `json:"text"`
-}
-
-func parseJSONStreamInternal(r io.Reader, warnFn func(string), infoFn func(string), onMessage func(), onComplete func()) (message, threadID string) {
+func ParseJSONStreamInternal(r io.Reader, warnFn func(string), infoFn func(string), onMessage func(), onComplete func()) (message, threadID string) {
 	reader := bufio.NewReaderSize(r, jsonLineReaderSize)
+	scratch := lineScratchPool.Get().(*lineScratch)
+	if scratch.buf == nil {
+		scratch.buf = make([]byte, 0, jsonLineReaderSize)
+	} else {
+		scratch.buf = scratch.buf[:0]
+	}
+	if scratch.preview == nil {
+		scratch.preview = make([]byte, 0, jsonLinePreviewBytes)
+	} else {
+		scratch.preview = scratch.preview[:0]
+	}
+	defer func() {
+		if cap(scratch.buf) > maxPooledLineScratchCap {
+			scratch.buf = nil
+		} else if scratch.buf != nil {
+			scratch.buf = scratch.buf[:0]
+		}
+		if cap(scratch.preview) > jsonLinePreviewBytes*4 {
+			scratch.preview = nil
+		} else if scratch.preview != nil {
+			scratch.preview = scratch.preview[:0]
+		}
+		lineScratchPool.Put(scratch)
+	}()

 	if warnFn == nil {
 		warnFn = func(string) {}
@@ -131,7 +90,7 @@ func parseJSONStreamInternal(r io.Reader, warnFn func(string), infoFn func(strin
 	)

 	for {
-		line, tooLong, err := readLineWithLimit(reader, jsonLineMaxBytes, jsonLinePreviewBytes)
+		line, tooLong, err := readLineWithLimit(reader, jsonLineMaxBytes, jsonLinePreviewBytes, scratch)
 		if err != nil {
 			if errors.Is(err, io.EOF) {
 				break
@@ -147,14 +106,14 @@ func parseJSONStreamInternal(r io.Reader, warnFn func(string), infoFn func(strin
 		totalEvents++

 		if tooLong {
-			warnFn(fmt.Sprintf("Skipped overlong JSON line (> %d bytes): %s", jsonLineMaxBytes, truncateBytes(line, 100)))
+			warnFn(fmt.Sprintf("Skipped overlong JSON line (> %d bytes): %s", jsonLineMaxBytes, TruncateBytes(line, 100)))
 			continue
 		}

 		// Single unmarshal for all backend types
 		var event UnifiedEvent
 		if err := json.Unmarshal(line, &event); err != nil {
-			warnFn(fmt.Sprintf("Failed to parse event: %s", truncateBytes(line, 100)))
+			warnFn(fmt.Sprintf("Failed to parse event: %s", TruncateBytes(line, 100)))
 			continue
 		}

@@ -253,7 +212,7 @@ func parseJSONStreamInternal(r io.Reader, warnFn func(string), infoFn func(strin
 					// Lazy parse: only parse item content when needed
 					var item ItemContent
 					if err := json.Unmarshal(event.Item, &item); err == nil {
-						normalized := normalizeText(item.Text)
+						normalized := NormalizeText(item.Text)
 						infoFn(fmt.Sprintf("item.completed event item_type=%s message_len=%d", itemType, len(normalized)))
 						if normalized != "" {
 							codexMessage = normalized
@@ -334,12 +293,12 @@ func parseJSONStreamInternal(r io.Reader, warnFn func(string), infoFn func(strin
 	return message, threadID
 }

-func hasKey(m map[string]json.RawMessage, key string) bool {
+func HasKey(m map[string]json.RawMessage, key string) bool {
 	_, ok := m[key]
 	return ok
 }

-func discardInvalidJSON(decoder *json.Decoder, reader *bufio.Reader) (*bufio.Reader, error) {
+func DiscardInvalidJSON(decoder *json.Decoder, reader *bufio.Reader) (*bufio.Reader, error) {
 	var buffered bytes.Buffer

 	if decoder != nil {
@@ -365,7 +324,7 @@ func discardInvalidJSON(decoder *json.Decoder, reader *bufio.Reader) (*bufio.Rea
 	return bufio.NewReader(io.MultiReader(bytes.NewReader(remaining), reader)), err
 }

-func readLineWithLimit(r *bufio.Reader, maxBytes int, previewBytes int) (line []byte, tooLong bool, err error) {
+func readLineWithLimit(r *bufio.Reader, maxBytes int, previewBytes int, scratch *lineScratch) (line []byte, tooLong bool, err error) {
 	if r == nil {
 		return nil, false, errors.New("reader is nil")
 	}
@@ -388,12 +347,22 @@ func readLineWithLimit(r *bufio.Reader, maxBytes int, previewBytes int) (line []
 		return part, false, nil
 	}

-	preview := make([]byte, 0, min(previewBytes, len(part)))
+	if scratch == nil {
+		scratch = &lineScratch{}
+	}
+	if scratch.preview == nil {
+		scratch.preview = make([]byte, 0, min(previewBytes, len(part)))
+	}
+	if scratch.buf == nil {
+		scratch.buf = make([]byte, 0, min(maxBytes, len(part)*2))
+	}
+
+	preview := scratch.preview[:0]
 	if previewBytes > 0 {
 		preview = append(preview, part[:min(previewBytes, len(part))]...)
 	}

-	buf := make([]byte, 0, min(maxBytes, len(part)*2))
+	buf := scratch.buf[:0]
 	total := 0
 	if len(part) > maxBytes {
 		tooLong = true
@@ -423,12 +392,16 @@ func readLineWithLimit(r *bufio.Reader, maxBytes int, previewBytes int) (line []
 	}

 	if tooLong {
+		scratch.preview = preview
+		scratch.buf = buf
 		return preview, true, nil
 	}
+	scratch.preview = preview
+	scratch.buf = buf
 	return buf, false, nil
 }

-func truncateBytes(b []byte, maxLen int) string {
+func TruncateBytes(b []byte, maxLen int) string {
 	if len(b) <= maxLen {
 		return string(b)
 	}
@@ -438,7 +411,7 @@ func truncateBytes(b []byte, maxLen int) string {
 	return string(b[:maxLen]) + "..."
 }

-func normalizeText(text interface{}) string {
+func NormalizeText(text interface{}) string {
 	switch v := text.(type) {
 	case string:
 		return v
@@ -454,3 +427,10 @@ func normalizeText(text interface{}) string {
 		return ""
 	}
 }
+
+func min(a, b int) int {
+	if a < b {
+		return a
+	}
+	return b
+}
--- a/codeagent-wrapper/internal/parser/parser_opencode_test.go
+++ b/codeagent-wrapper/internal/parser/parser_opencode_test.go
@@ -1,4 +1,4 @@
-package main
+package parser

 import (
 	"strings"
@@ -10,7 +10,7 @@ func TestParseJSONStream_Opencode(t *testing.T) {
 {"type":"text","timestamp":1768187744432,"sessionID":"ses_44fced3c7ffe83sZpzY1rlQka3","part":{"id":"prt_bb0339cb5001QDd0Lh0PzFZpa3","sessionID":"ses_44fced3c7ffe83sZpzY1rlQka3","messageID":"msg_bb033866f0011oZxTqvfy0TKtS","type":"text","text":"Hello from opencode"}}
 {"type":"step_finish","timestamp":1768187744471,"sessionID":"ses_44fced3c7ffe83sZpzY1rlQka3","part":{"id":"prt_bb033d0af0019VRZzpO2OVW1na","sessionID":"ses_44fced3c7ffe83sZpzY1rlQka3","messageID":"msg_bb033866f0011oZxTqvfy0TKtS","type":"step-finish","reason":"stop","snapshot":"904f0fd58c125b79e60f0993e38f9d9f6200bf47","cost":0}}`

-	message, threadID := parseJSONStream(strings.NewReader(input))
+	message, threadID := ParseJSONStreamInternal(strings.NewReader(input), nil, nil, nil, nil)

 	if threadID != "ses_44fced3c7ffe83sZpzY1rlQka3" {
 		t.Errorf("threadID = %q, want %q", threadID, "ses_44fced3c7ffe83sZpzY1rlQka3")
@@ -25,7 +25,7 @@ func TestParseJSONStream_Opencode_MultipleTextEvents(t *testing.T) {
 {"type":"text","sessionID":"ses_123","part":{"type":"text","text":" Part 2"}}
 {"type":"step_finish","sessionID":"ses_123","part":{"type":"step-finish","reason":"stop"}}`

-	message, threadID := parseJSONStream(strings.NewReader(input))
+	message, threadID := ParseJSONStreamInternal(strings.NewReader(input), nil, nil, nil, nil)

 	if threadID != "ses_123" {
 		t.Errorf("threadID = %q, want %q", threadID, "ses_123")
@@ -39,7 +39,7 @@ func TestParseJSONStream_Opencode_NoStopReason(t *testing.T) {
 	input := `{"type":"text","sessionID":"ses_456","part":{"type":"text","text":"Content"}}
 {"type":"step_finish","sessionID":"ses_456","part":{"type":"step-finish","reason":"tool-calls"}}`

-	message, threadID := parseJSONStream(strings.NewReader(input))
+	message, threadID := ParseJSONStreamInternal(strings.NewReader(input), nil, nil, nil, nil)

 	if threadID != "ses_456" {
 		t.Errorf("threadID = %q, want %q", threadID, "ses_456")
--- a/codeagent-wrapper/internal/parser/parser_token_too_long_test.go
+++ b/codeagent-wrapper/internal/parser/parser_token_too_long_test.go
@@ -1,4 +1,4 @@
-package main
+package parser

 import (
 	"strings"
@@ -18,7 +18,7 @@ func TestParseJSONStream_SkipsOverlongLineAndContinues(t *testing.T) {
 	var warns []string
 	warnFn := func(msg string) { warns = append(warns, msg) }

-	gotMessage, gotThreadID := parseJSONStreamInternal(strings.NewReader(input), warnFn, nil, nil, nil)
+	gotMessage, gotThreadID := ParseJSONStreamInternal(strings.NewReader(input), warnFn, nil, nil, nil)
 	if gotMessage != "ok" {
 		t.Fatalf("message=%q, want %q (warns=%v)", gotMessage, "ok", warns)
 	}
--- a/codeagent-wrapper/internal/parser/parser_unknown_event_test.go
+++ b/codeagent-wrapper/internal/parser/parser_unknown_event_test.go
@@ -1,4 +1,4 @@
-package main
+package parser

 import (
 	"strings"
@@ -16,7 +16,7 @@ func TestBackendParseJSONStream_UnknownEventsAreSilent(t *testing.T) {
 	var infos []string
 	infoFn := func(msg string) { infos = append(infos, msg) }

-	message, threadID := parseJSONStreamInternal(strings.NewReader(input), nil, infoFn, nil, nil)
+	message, threadID := ParseJSONStreamInternal(strings.NewReader(input), nil, infoFn, nil, nil)
 	if message != "ok" {
 		t.Fatalf("message=%q, want %q (infos=%v)", message, "ok", infos)
 	}
--- a/codeagent-wrapper/internal/parser/truncate_bytes_test.go
+++ b/codeagent-wrapper/internal/parser/truncate_bytes_test.go
@@ -0,0 +1,15 @@
+package parser
+
+import "testing"
+
+func TestTruncateBytes(t *testing.T) {
+	if got := TruncateBytes([]byte("abc"), 3); got != "abc" {
+		t.Fatalf("TruncateBytes() = %q, want %q", got, "abc")
+	}
+	if got := TruncateBytes([]byte("abcd"), 3); got != "abc..." {
+		t.Fatalf("TruncateBytes() = %q, want %q", got, "abc...")
+	}
+	if got := TruncateBytes([]byte("abcd"), -1); got != "" {
+		t.Fatalf("TruncateBytes() = %q, want empty string", got)
+	}
+}
--- a/codeagent-wrapper/internal/utils/math.go
+++ b/codeagent-wrapper/internal/utils/math.go
@@ -0,0 +1,8 @@
+package utils
+
+func Min(a, b int) int {
+	if a < b {
+		return a
+	}
+	return b
+}
--- a/codeagent-wrapper/internal/utils/math_test.go
+++ b/codeagent-wrapper/internal/utils/math_test.go
@@ -0,0 +1,36 @@
+package utils
+
+import "testing"
+
+func TestMin(t *testing.T) {
+	tests := []struct {
+		name string
+		a, b int
+		want int
+	}{
+		{"a less than b", 1, 2, 1},
+		{"b less than a", 5, 3, 3},
+		{"equal values", 7, 7, 7},
+		{"negative a", -5, 3, -5},
+		{"negative b", 5, -3, -3},
+		{"both negative", -5, -3, -5},
+		{"zero and positive", 0, 5, 0},
+		{"zero and negative", 0, -5, -5},
+		{"large values", 1000000, 999999, 999999},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			got := Min(tt.a, tt.b)
+			if got != tt.want {
+				t.Errorf("Min(%d, %d) = %d, want %d", tt.a, tt.b, got, tt.want)
+			}
+		})
+	}
+}
+
+func BenchmarkMin(b *testing.B) {
+	for i := 0; i < b.N; i++ {
+		Min(i, i+1)
+	}
+}
--- a/codeagent-wrapper/internal/utils/strings.go
+++ b/codeagent-wrapper/internal/utils/strings.go
@@ -0,0 +1,62 @@
+package utils
+
+import "strings"
+
+func Truncate(s string, maxLen int) string {
+	if len(s) <= maxLen {
+		return s
+	}
+	if maxLen < 0 {
+		return ""
+	}
+	return s[:maxLen] + "..."
+}
+
+// SafeTruncate safely truncates string to maxLen, avoiding panic and UTF-8 corruption.
+func SafeTruncate(s string, maxLen int) string {
+	if maxLen <= 0 || s == "" {
+		return ""
+	}
+
+	runes := []rune(s)
+	if len(runes) <= maxLen {
+		return s
+	}
+
+	if maxLen < 4 {
+		return string(runes[:1])
+	}
+
+	cutoff := maxLen - 3
+	if cutoff <= 0 {
+		return string(runes[:1])
+	}
+	if len(runes) <= cutoff {
+		return s
+	}
+	return string(runes[:cutoff]) + "..."
+}
+
+// SanitizeOutput removes ANSI escape sequences and control characters.
+func SanitizeOutput(s string) string {
+	var result strings.Builder
+	inEscape := false
+	for i := 0; i < len(s); i++ {
+		if s[i] == '\x1b' && i+1 < len(s) && s[i+1] == '[' {
+			inEscape = true
+			i++ // skip '['
+			continue
+		}
+		if inEscape {
+			if (s[i] >= 'A' && s[i] <= 'Z') || (s[i] >= 'a' && s[i] <= 'z') {
+				inEscape = false
+			}
+			continue
+		}
+		// Keep printable chars and common whitespace.
+		if s[i] >= 32 || s[i] == '\n' || s[i] == '\t' {
+			result.WriteByte(s[i])
+		}
+	}
+	return result.String()
+}
--- a/codeagent-wrapper/internal/utils/strings_test.go
+++ b/codeagent-wrapper/internal/utils/strings_test.go
@@ -0,0 +1,122 @@
+package utils
+
+import (
+	"strings"
+	"testing"
+)
+
+func TestTruncate(t *testing.T) {
+	tests := []struct {
+		name   string
+		s      string
+		maxLen int
+		want   string
+	}{
+		{"empty string", "", 10, ""},
+		{"short string", "hello", 10, "hello"},
+		{"exact length", "hello", 5, "hello"},
+		{"needs truncation", "hello world", 5, "hello..."},
+		{"zero maxLen", "hello", 0, "..."},
+		{"negative maxLen", "hello", -1, ""},
+		{"maxLen 1", "hello", 1, "h..."},
+		{"unicode bytes truncate", "你好世界", 10, "你好世\xe7..."},  // Truncate works on bytes, not runes
+		{"mixed truncate", "hello世界abc", 7, "hello\xe4\xb8..."}, // byte-based truncation
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			got := Truncate(tt.s, tt.maxLen)
+			if got != tt.want {
+				t.Errorf("Truncate(%q, %d) = %q, want %q", tt.s, tt.maxLen, got, tt.want)
+			}
+		})
+	}
+}
+
+func TestSafeTruncate(t *testing.T) {
+	tests := []struct {
+		name   string
+		s      string
+		maxLen int
+		want   string
+	}{
+		{"empty string", "", 10, ""},
+		{"zero maxLen", "hello", 0, ""},
+		{"negative maxLen", "hello", -1, ""},
+		{"short string", "hello", 10, "hello"},
+		{"exact length", "hello", 5, "hello"},
+		{"needs truncation", "hello world", 8, "hello..."},
+		{"maxLen 1", "hello", 1, "h"},
+		{"maxLen 2", "hello", 2, "h"},
+		{"maxLen 3", "hello", 3, "h"},
+		{"maxLen 4", "hello", 4, "h..."},
+		{"unicode preserved", "你好世界", 10, "你好世界"},
+		{"unicode exact", "你好世界", 4, "你好世界"},
+		{"unicode truncate", "你好世界test", 6, "你好世..."},
+		{"mixed unicode", "ab你好cd", 5, "ab..."},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			got := SafeTruncate(tt.s, tt.maxLen)
+			if got != tt.want {
+				t.Errorf("SafeTruncate(%q, %d) = %q, want %q", tt.s, tt.maxLen, got, tt.want)
+			}
+		})
+	}
+}
+
+func TestSanitizeOutput(t *testing.T) {
+	tests := []struct {
+		name string
+		s    string
+		want string
+	}{
+		{"empty string", "", ""},
+		{"plain text", "hello world", "hello world"},
+		{"with newline", "hello\nworld", "hello\nworld"},
+		{"with tab", "hello\tworld", "hello\tworld"},
+		{"ANSI color red", "\x1b[31mred\x1b[0m", "red"},
+		{"ANSI bold", "\x1b[1mbold\x1b[0m", "bold"},
+		{"ANSI complex", "\x1b[1;31;40mtext\x1b[0m", "text"},
+		{"control chars", "hello\x00\x01\x02world", "helloworld"},
+		{"mixed ANSI and control", "\x1b[32m\x00ok\x1b[0m", "ok"},
+		{"multiple ANSI sequences", "\x1b[31mred\x1b[0m \x1b[32mgreen\x1b[0m", "red green"},
+		{"incomplete escape", "\x1b[", ""},
+		{"escape without bracket", "\x1bA", "A"},
+		{"cursor movement", "\x1b[2Aup\x1b[2Bdown", "updown"},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			got := SanitizeOutput(tt.s)
+			if got != tt.want {
+				t.Errorf("SanitizeOutput(%q) = %q, want %q", tt.s, got, tt.want)
+			}
+		})
+	}
+}
+
+func BenchmarkTruncate(b *testing.B) {
+	s := strings.Repeat("hello world ", 100)
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		Truncate(s, 50)
+	}
+}
+
+func BenchmarkSafeTruncate(b *testing.B) {
+	s := strings.Repeat("你好世界", 100)
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		SafeTruncate(s, 50)
+	}
+}
+
+func BenchmarkSanitizeOutput(b *testing.B) {
+	s := strings.Repeat("\x1b[31mred\x1b[0m text ", 50)
+	b.ResetTimer()
+	for i := 0; i < b.N; i++ {
+		SanitizeOutput(s)
+	}
+}
--- a/codeagent-wrapper/main.go
+++ b/codeagent-wrapper/main.go
@@ -1,627 +0,0 @@
-package main
-
-import (
-	"fmt"
-	"io"
-	"os"
-	"os/exec"
-	"os/signal"
-	"path/filepath"
-	"reflect"
-	"strings"
-	"sync/atomic"
-	"time"
-)
-
-const (
-	version               = "5.5.0"
-	defaultWorkdir        = "."
-	defaultTimeout        = 7200 // seconds (2 hours)
-	defaultCoverageTarget = 90.0
-	codexLogLineLimit     = 1000
-	stdinSpecialChars     = "\n\\\"'`$"
-	stderrCaptureLimit    = 4 * 1024
-	defaultBackendName    = "codex"
-	defaultCodexCommand   = "codex"
-
-	// stdout close reasons
-	stdoutCloseReasonWait  = "wait-done"
-	stdoutCloseReasonDrain = "drain-timeout"
-	stdoutCloseReasonCtx   = "context-cancel"
-	stdoutDrainTimeout     = 100 * time.Millisecond
-)
-
-// Test hooks for dependency injection
-var (
-	stdinReader  io.Reader = os.Stdin
-	isTerminalFn           = defaultIsTerminal
-	codexCommand           = defaultCodexCommand
-	cleanupHook  func()
-	loggerPtr    atomic.Pointer[Logger]
-
-	buildCodexArgsFn   = buildCodexArgs
-	selectBackendFn    = selectBackend
-	commandContext     = exec.CommandContext
-	cleanupLogsFn      = cleanupOldLogs
-	signalNotifyFn     = signal.Notify
-	signalStopFn       = signal.Stop
-	terminateCommandFn = terminateCommand
-	defaultBuildArgsFn = buildCodexArgs
-	runTaskFn          = runCodexTask
-	exitFn             = os.Exit
-)
-
-var forceKillDelay atomic.Int32
-
-func init() {
-	forceKillDelay.Store(5) // seconds - default value
-}
-
-func runStartupCleanup() {
-	if cleanupLogsFn == nil {
-		return
-	}
-	defer func() {
-		if r := recover(); r != nil {
-			logWarn(fmt.Sprintf("cleanupOldLogs panic: %v", r))
-		}
-	}()
-	if _, err := cleanupLogsFn(); err != nil {
-		logWarn(fmt.Sprintf("cleanupOldLogs error: %v", err))
-	}
-}
-
-func runCleanupMode() int {
-	if cleanupLogsFn == nil {
-		fmt.Fprintln(os.Stderr, "Cleanup failed: log cleanup function not configured")
-		return 1
-	}
-
-	stats, err := cleanupLogsFn()
-	if err != nil {
-		fmt.Fprintf(os.Stderr, "Cleanup failed: %v\n", err)
-		return 1
-	}
-
-	fmt.Println("Cleanup completed")
-	fmt.Printf("Files scanned: %d\n", stats.Scanned)
-	fmt.Printf("Files deleted: %d\n", stats.Deleted)
-	if len(stats.DeletedFiles) > 0 {
-		for _, f := range stats.DeletedFiles {
-			fmt.Printf("  - %s\n", f)
-		}
-	}
-	fmt.Printf("Files kept: %d\n", stats.Kept)
-	if len(stats.KeptFiles) > 0 {
-		for _, f := range stats.KeptFiles {
-			fmt.Printf("  - %s\n", f)
-		}
-	}
-	if stats.Errors > 0 {
-		fmt.Printf("Deletion errors: %d\n", stats.Errors)
-	}
-	return 0
-}
-
-func main() {
-	exitCode := run()
-	exitFn(exitCode)
-}
-
-// run is the main logic, returns exit code for testability
-func run() (exitCode int) {
-	name := currentWrapperName()
-	// Handle --version and --help first (no logger needed)
-	if len(os.Args) > 1 {
-		switch os.Args[1] {
-		case "--version", "-v":
-			fmt.Printf("%s version %s\n", name, version)
-			return 0
-		case "--help", "-h":
-			printHelp()
-			return 0
-		case "--cleanup":
-			return runCleanupMode()
-		}
-	}
-
-	// Initialize logger for all other commands
-	logger, err := NewLogger()
-	if err != nil {
-		fmt.Fprintf(os.Stderr, "ERROR: failed to initialize logger: %v\n", err)
-		return 1
-	}
-	setLogger(logger)
-
-	defer func() {
-		logger := activeLogger()
-		if logger != nil {
-			logger.Flush()
-		}
-		if err := closeLogger(); err != nil {
-			fmt.Fprintf(os.Stderr, "ERROR: failed to close logger: %v\n", err)
-		}
-		// On failure, extract and display recent errors before removing log
-		if logger != nil {
-			if exitCode != 0 {
-				if errors := logger.ExtractRecentErrors(10); len(errors) > 0 {
-					fmt.Fprintln(os.Stderr, "\n=== Recent Errors ===")
-					for _, entry := range errors {
-						fmt.Fprintln(os.Stderr, entry)
-					}
-					fmt.Fprintf(os.Stderr, "Log file: %s (deleted)\n", logger.Path())
-				}
-			}
-			if err := logger.RemoveLogFile(); err != nil && !os.IsNotExist(err) {
-				// Silently ignore removal errors
-			}
-		}
-	}()
-	defer runCleanupHook()
-
-	// Clean up stale logs from previous runs.
-	runStartupCleanup()
-
-	// Handle remaining commands
-	if len(os.Args) > 1 {
-		args := os.Args[1:]
-		parallelIndex := -1
-		for i, arg := range args {
-			if arg == "--parallel" {
-				parallelIndex = i
-				break
-			}
-		}
-
-		if parallelIndex != -1 {
-			backendName := defaultBackendName
-			model := ""
-			fullOutput := false
-			skipPermissions := envFlagEnabled("CODEAGENT_SKIP_PERMISSIONS")
-			var extras []string
-
-			for i := 0; i < len(args); i++ {
-				arg := args[i]
-				switch {
-				case arg == "--parallel":
-					continue
-				case arg == "--full-output":
-					fullOutput = true
-				case arg == "--backend":
-					if i+1 >= len(args) {
-						fmt.Fprintln(os.Stderr, "ERROR: --backend flag requires a value")
-						return 1
-					}
-					backendName = args[i+1]
-					i++
-				case strings.HasPrefix(arg, "--backend="):
-					value := strings.TrimPrefix(arg, "--backend=")
-					if value == "" {
-						fmt.Fprintln(os.Stderr, "ERROR: --backend flag requires a value")
-						return 1
-					}
-					backendName = value
-				case arg == "--model":
-					if i+1 >= len(args) {
-						fmt.Fprintln(os.Stderr, "ERROR: --model flag requires a value")
-						return 1
-					}
-					model = args[i+1]
-					i++
-				case strings.HasPrefix(arg, "--model="):
-					value := strings.TrimPrefix(arg, "--model=")
-					if value == "" {
-						fmt.Fprintln(os.Stderr, "ERROR: --model flag requires a value")
-						return 1
-					}
-					model = value
-				case arg == "--skip-permissions", arg == "--dangerously-skip-permissions":
-					skipPermissions = true
-				case strings.HasPrefix(arg, "--skip-permissions="):
-					skipPermissions = parseBoolFlag(strings.TrimPrefix(arg, "--skip-permissions="), skipPermissions)
-				case strings.HasPrefix(arg, "--dangerously-skip-permissions="):
-					skipPermissions = parseBoolFlag(strings.TrimPrefix(arg, "--dangerously-skip-permissions="), skipPermissions)
-				default:
-					extras = append(extras, arg)
-				}
-			}
-
-			if len(extras) > 0 {
-				fmt.Fprintln(os.Stderr, "ERROR: --parallel reads its task configuration from stdin; only --backend, --model, --full-output and --skip-permissions are allowed.")
-				fmt.Fprintln(os.Stderr, "Usage examples:")
-				fmt.Fprintf(os.Stderr, "  %s --parallel < tasks.txt\n", name)
-				fmt.Fprintf(os.Stderr, "  echo '...' | %s --parallel\n", name)
-				fmt.Fprintf(os.Stderr, "  %s --parallel <<'EOF'\n", name)
-				fmt.Fprintf(os.Stderr, "  %s --parallel --full-output <<'EOF'  # include full task output\n", name)
-				return 1
-			}
-
-			backend, err := selectBackendFn(backendName)
-			if err != nil {
-				fmt.Fprintf(os.Stderr, "ERROR: %v\n", err)
-				return 1
-			}
-			backendName = backend.Name()
-
-			data, err := io.ReadAll(stdinReader)
-			if err != nil {
-				fmt.Fprintf(os.Stderr, "ERROR: failed to read stdin: %v\n", err)
-				return 1
-			}
-
-			cfg, err := parseParallelConfig(data)
-			if err != nil {
-				fmt.Fprintf(os.Stderr, "ERROR: %v\n", err)
-				return 1
-			}
-
-			cfg.GlobalBackend = backendName
-			model = strings.TrimSpace(model)
-			for i := range cfg.Tasks {
-				if strings.TrimSpace(cfg.Tasks[i].Backend) == "" {
-					cfg.Tasks[i].Backend = backendName
-				}
-				if strings.TrimSpace(cfg.Tasks[i].Model) == "" && model != "" {
-					cfg.Tasks[i].Model = model
-				}
-				cfg.Tasks[i].SkipPermissions = cfg.Tasks[i].SkipPermissions || skipPermissions
-			}
-
-			timeoutSec := resolveTimeout()
-			layers, err := topologicalSort(cfg.Tasks)
-			if err != nil {
-				fmt.Fprintf(os.Stderr, "ERROR: %v\n", err)
-				return 1
-			}
-
-			results := executeConcurrent(layers, timeoutSec)
-
-			// Extract structured report fields from each result
-			for i := range results {
-				results[i].CoverageTarget = defaultCoverageTarget
-				if results[i].Message == "" {
-					continue
-				}
-
-				lines := strings.Split(results[i].Message, "\n")
-
-				// Coverage extraction
-				results[i].Coverage = extractCoverageFromLines(lines)
-				results[i].CoverageNum = extractCoverageNum(results[i].Coverage)
-
-				// Files changed
-				results[i].FilesChanged = extractFilesChangedFromLines(lines)
-
-				// Test results
-				results[i].TestsPassed, results[i].TestsFailed = extractTestResultsFromLines(lines)
-
-				// Key output summary
-				results[i].KeyOutput = extractKeyOutputFromLines(lines, 150)
-			}
-
-			// Default: summary mode (context-efficient)
-			// --full-output: legacy full output mode
-			fmt.Println(generateFinalOutputWithMode(results, !fullOutput))
-
-			exitCode = 0
-			for _, res := range results {
-				if res.ExitCode != 0 {
-					exitCode = res.ExitCode
-				}
-			}
-
-			return exitCode
-		}
-	}
-
-	logInfo("Script started")
-
-	cfg, err := parseArgs()
-	if err != nil {
-		logError(err.Error())
-		return 1
-	}
-	logInfo(fmt.Sprintf("Parsed args: mode=%s, task_len=%d, backend=%s", cfg.Mode, len(cfg.Task), cfg.Backend))
-
-	backend, err := selectBackendFn(cfg.Backend)
-	if err != nil {
-		logError(err.Error())
-		return 1
-	}
-	cfg.Backend = backend.Name()
-
-	cmdInjected := codexCommand != defaultCodexCommand
-	argsInjected := buildCodexArgsFn != nil && reflect.ValueOf(buildCodexArgsFn).Pointer() != reflect.ValueOf(defaultBuildArgsFn).Pointer()
-
-	// Wire selected backend into runtime hooks for the rest of the execution,
-	// but preserve any injected test hooks for the default backend.
-	if backend.Name() != defaultBackendName || !cmdInjected {
-		codexCommand = backend.Command()
-	}
-	if backend.Name() != defaultBackendName || !argsInjected {
-		buildCodexArgsFn = backend.BuildArgs
-	}
-	logInfo(fmt.Sprintf("Selected backend: %s", backend.Name()))
-
-	timeoutSec := resolveTimeout()
-	logInfo(fmt.Sprintf("Timeout: %ds", timeoutSec))
-	cfg.Timeout = timeoutSec
-
-	var taskText string
-	var piped bool
-
-	if cfg.ExplicitStdin {
-		logInfo("Explicit stdin mode: reading task from stdin")
-		data, err := io.ReadAll(stdinReader)
-		if err != nil {
-			logError("Failed to read stdin: " + err.Error())
-			return 1
-		}
-		taskText = string(data)
-		if taskText == "" {
-			logError("Explicit stdin mode requires task input from stdin")
-			return 1
-		}
-		piped = !isTerminal()
-	} else {
-		pipedTask, err := readPipedTask()
-		if err != nil {
-			logError("Failed to read piped stdin: " + err.Error())
-			return 1
-		}
-		piped = pipedTask != ""
-		if piped {
-			taskText = pipedTask
-		} else {
-			taskText = cfg.Task
-		}
-	}
-
-	if strings.TrimSpace(cfg.PromptFile) != "" {
-		prompt, err := readAgentPromptFile(cfg.PromptFile, cfg.PromptFileExplicit)
-		if err != nil {
-			logError("Failed to read prompt file: " + err.Error())
-			return 1
-		}
-		taskText = wrapTaskWithAgentPrompt(prompt, taskText)
-	}
-
-	useStdin := cfg.ExplicitStdin || shouldUseStdin(taskText, piped)
-
-	targetArg := taskText
-	if useStdin {
-		targetArg = "-"
-	}
-	codexArgs := buildCodexArgsFn(cfg, targetArg)
-
-	// Print startup information to stderr
-	fmt.Fprintf(os.Stderr, "[%s]\n", name)
-	fmt.Fprintf(os.Stderr, "  Backend: %s\n", cfg.Backend)
-	fmt.Fprintf(os.Stderr, "  Command: %s %s\n", codexCommand, strings.Join(codexArgs, " "))
-	fmt.Fprintf(os.Stderr, "  PID: %d\n", os.Getpid())
-	fmt.Fprintf(os.Stderr, "  Log: %s\n", logger.Path())
-
-	if useStdin {
-		var reasons []string
-		if piped {
-			reasons = append(reasons, "piped input")
-		}
-		if cfg.ExplicitStdin {
-			reasons = append(reasons, "explicit \"-\"")
-		}
-		if strings.Contains(taskText, "\n") {
-			reasons = append(reasons, "newline")
-		}
-		if strings.Contains(taskText, "\\") {
-			reasons = append(reasons, "backslash")
-		}
-		if strings.Contains(taskText, "\"") {
-			reasons = append(reasons, "double-quote")
-		}
-		if strings.Contains(taskText, "'") {
-			reasons = append(reasons, "single-quote")
-		}
-		if strings.Contains(taskText, "`") {
-			reasons = append(reasons, "backtick")
-		}
-		if strings.Contains(taskText, "$") {
-			reasons = append(reasons, "dollar")
-		}
-		if len(taskText) > 800 {
-			reasons = append(reasons, "length>800")
-		}
-		if len(reasons) > 0 {
-			logWarn(fmt.Sprintf("Using stdin mode for task due to: %s", strings.Join(reasons, ", ")))
-		}
-	}
-
-	logInfo(fmt.Sprintf("%s running...", cfg.Backend))
-
-	taskSpec := TaskSpec{
-		Task:            taskText,
-		WorkDir:         cfg.WorkDir,
-		Mode:            cfg.Mode,
-		SessionID:       cfg.SessionID,
-		Model:           cfg.Model,
-		ReasoningEffort: cfg.ReasoningEffort,
-		SkipPermissions: cfg.SkipPermissions,
-		UseStdin:        useStdin,
-	}
-
-	result := runTaskFn(taskSpec, false, cfg.Timeout)
-
-	if result.ExitCode != 0 {
-		return result.ExitCode
-	}
-
-	fmt.Println(result.Message)
-	if result.SessionID != "" {
-		fmt.Printf("\n---\nSESSION_ID: %s\n", result.SessionID)
-	}
-
-	return 0
-}
-
-func readAgentPromptFile(path string, allowOutsideClaudeDir bool) (string, error) {
-	raw := strings.TrimSpace(path)
-	if raw == "" {
-		return "", nil
-	}
-
-	expanded := raw
-	if raw == "~" || strings.HasPrefix(raw, "~/") || strings.HasPrefix(raw, "~\\") {
-		home, err := os.UserHomeDir()
-		if err != nil {
-			return "", err
-		}
-		if raw == "~" {
-			expanded = home
-		} else {
-			expanded = home + raw[1:]
-		}
-	}
-
-	absPath, err := filepath.Abs(expanded)
-	if err != nil {
-		return "", err
-	}
-	absPath = filepath.Clean(absPath)
-
-	home, err := os.UserHomeDir()
-	if err != nil {
-		if !allowOutsideClaudeDir {
-			return "", err
-		}
-		logWarn(fmt.Sprintf("Failed to resolve home directory for prompt file validation: %v; proceeding without restriction", err))
-	} else {
-		allowedDir := filepath.Clean(filepath.Join(home, ".claude"))
-		allowedAbs, err := filepath.Abs(allowedDir)
-		if err == nil {
-			allowedDir = filepath.Clean(allowedAbs)
-		}
-
-		isWithinDir := func(path, dir string) bool {
-			rel, err := filepath.Rel(dir, path)
-			if err != nil {
-				return false
-			}
-			rel = filepath.Clean(rel)
-			if rel == "." {
-				return true
-			}
-			if rel == ".." {
-				return false
-			}
-			prefix := ".." + string(os.PathSeparator)
-			return !strings.HasPrefix(rel, prefix)
-		}
-
-		if !allowOutsideClaudeDir {
-			if !isWithinDir(absPath, allowedDir) {
-				logWarn(fmt.Sprintf("Refusing to read prompt file outside %s: %s", allowedDir, absPath))
-				return "", fmt.Errorf("prompt file must be under %s", allowedDir)
-			}
-			resolvedPath, errPath := filepath.EvalSymlinks(absPath)
-			resolvedBase, errBase := filepath.EvalSymlinks(allowedDir)
-			if errPath == nil && errBase == nil {
-				resolvedPath = filepath.Clean(resolvedPath)
-				resolvedBase = filepath.Clean(resolvedBase)
-				if !isWithinDir(resolvedPath, resolvedBase) {
-					logWarn(fmt.Sprintf("Refusing to read prompt file outside %s (resolved): %s", resolvedBase, resolvedPath))
-					return "", fmt.Errorf("prompt file must be under %s", resolvedBase)
-				}
-			}
-		} else if !isWithinDir(absPath, allowedDir) {
-			logWarn(fmt.Sprintf("Reading prompt file outside %s: %s", allowedDir, absPath))
-		}
-	}
-
-	data, err := os.ReadFile(absPath)
-	if err != nil {
-		return "", err
-	}
-	return strings.TrimRight(string(data), "\r\n"), nil
-}
-
-func wrapTaskWithAgentPrompt(prompt string, task string) string {
-	return "<agent-prompt>\n" + prompt + "\n</agent-prompt>\n\n" + task
-}
-
-func setLogger(l *Logger) {
-	loggerPtr.Store(l)
-}
-
-func closeLogger() error {
-	logger := loggerPtr.Swap(nil)
-	if logger == nil {
-		return nil
-	}
-	return logger.Close()
-}
-
-func activeLogger() *Logger {
-	return loggerPtr.Load()
-}
-
-func logInfo(msg string) {
-	if logger := activeLogger(); logger != nil {
-		logger.Info(msg)
-	}
-}
-
-func logWarn(msg string) {
-	if logger := activeLogger(); logger != nil {
-		logger.Warn(msg)
-	}
-}
-
-func logError(msg string) {
-	if logger := activeLogger(); logger != nil {
-		logger.Error(msg)
-	}
-}
-
-func runCleanupHook() {
-	if logger := activeLogger(); logger != nil {
-		logger.Flush()
-	}
-	if cleanupHook != nil {
-		cleanupHook()
-	}
-}
-
-func printHelp() {
-	name := currentWrapperName()
-	help := fmt.Sprintf(`%[1]s - Go wrapper for AI CLI backends
-
-Usage:
-    %[1]s "task" [workdir]
-    %[1]s --backend claude "task" [workdir]
-    %[1]s --prompt-file /path/to/prompt.md "task" [workdir]
-    %[1]s - [workdir]              Read task from stdin
-    %[1]s resume <session_id> "task" [workdir]
-    %[1]s resume <session_id> - [workdir]
-    %[1]s --parallel               Run tasks in parallel (config from stdin)
-    %[1]s --parallel --full-output Run tasks in parallel with full output (legacy)
-    %[1]s --version
-    %[1]s --help
-
-Parallel mode examples:
-    %[1]s --parallel < tasks.txt
-    echo '...' | %[1]s --parallel
-    %[1]s --parallel --full-output < tasks.txt
-    %[1]s --parallel <<'EOF'
-
-Environment Variables:
-    CODEX_TIMEOUT         Timeout in milliseconds (default: 7200000)
-    CODEAGENT_ASCII_MODE  Use ASCII symbols instead of Unicode (PASS/WARN/FAIL)
-
-Exit Codes:
-    0    Success
-    1    General error (missing args, no output)
-    124  Timeout
-    127  backend command not found
-    130  Interrupted (Ctrl+C)
-    *    Passthrough from backend process`, name)
-	fmt.Println(help)
-}
--- a/codeagent-wrapper/process_check_test.go
+++ b/codeagent-wrapper/process_check_test.go
@@ -1,217 +0,0 @@
-//go:build unix || darwin || linux
-// +build unix darwin linux
-
-package main
-
-import (
-	"errors"
-	"fmt"
-	"os"
-	"os/exec"
-	"runtime"
-	"strconv"
-	"strings"
-	"testing"
-	"time"
-)
-
-func TestIsProcessRunning(t *testing.T) {
-	t.Run("current process", func(t *testing.T) {
-		if !isProcessRunning(os.Getpid()) {
-			t.Fatalf("expected current process (pid=%d) to be running", os.Getpid())
-		}
-	})
-
-	t.Run("fake pid", func(t *testing.T) {
-		const nonexistentPID = 1 << 30
-		if isProcessRunning(nonexistentPID) {
-			t.Fatalf("expected pid %d to be reported as not running", nonexistentPID)
-		}
-	})
-
-	t.Run("terminated process", func(t *testing.T) {
-		pid := exitedProcessPID(t)
-		if isProcessRunning(pid) {
-			t.Fatalf("expected exited child process (pid=%d) to be reported as not running", pid)
-		}
-	})
-
-	t.Run("boundary values", func(t *testing.T) {
-		if isProcessRunning(0) {
-			t.Fatalf("pid 0 should never be treated as running")
-		}
-		if isProcessRunning(-42) {
-			t.Fatalf("negative pid should never be treated as running")
-		}
-	})
-
-	t.Run("find process error", func(t *testing.T) {
-		original := findProcess
-		defer func() { findProcess = original }()
-
-		mockErr := errors.New("findProcess failure")
-		findProcess = func(pid int) (*os.Process, error) {
-			return nil, mockErr
-		}
-
-		if isProcessRunning(1234) {
-			t.Fatalf("expected false when os.FindProcess fails")
-		}
-	})
-}
-
-func exitedProcessPID(t *testing.T) int {
-	t.Helper()
-
-	var cmd *exec.Cmd
-	if runtime.GOOS == "windows" {
-		cmd = exec.Command("cmd", "/c", "exit 0")
-	} else {
-		cmd = exec.Command("sh", "-c", "exit 0")
-	}
-
-	if err := cmd.Start(); err != nil {
-		t.Fatalf("failed to start helper process: %v", err)
-	}
-	pid := cmd.Process.Pid
-
-	if err := cmd.Wait(); err != nil {
-		t.Fatalf("helper process did not exit cleanly: %v", err)
-	}
-
-	time.Sleep(50 * time.Millisecond)
-	return pid
-}
-
-func TestRunProcessCheckSmoke(t *testing.T) {
-	t.Run("current process", func(t *testing.T) {
-		if !isProcessRunning(os.Getpid()) {
-			t.Fatalf("expected current process (pid=%d) to be running", os.Getpid())
-		}
-	})
-
-	t.Run("fake pid", func(t *testing.T) {
-		const nonexistentPID = 1 << 30
-		if isProcessRunning(nonexistentPID) {
-			t.Fatalf("expected pid %d to be reported as not running", nonexistentPID)
-		}
-	})
-
-	t.Run("boundary values", func(t *testing.T) {
-		if isProcessRunning(0) {
-			t.Fatalf("pid 0 should never be treated as running")
-		}
-		if isProcessRunning(-42) {
-			t.Fatalf("negative pid should never be treated as running")
-		}
-	})
-
-	t.Run("find process error", func(t *testing.T) {
-		original := findProcess
-		defer func() { findProcess = original }()
-
-		mockErr := errors.New("findProcess failure")
-		findProcess = func(pid int) (*os.Process, error) {
-			return nil, mockErr
-		}
-
-		if isProcessRunning(1234) {
-			t.Fatalf("expected false when os.FindProcess fails")
-		}
-	})
-}
-
-func TestGetProcessStartTimeReadsProcStat(t *testing.T) {
-	pid := 4321
-	boot := time.Unix(1_710_000_000, 0)
-	startTicks := uint64(4500)
-
-	statFields := make([]string, 25)
-	for i := range statFields {
-		statFields[i] = strconv.Itoa(i + 1)
-	}
-	statFields[19] = strconv.FormatUint(startTicks, 10)
-	statContent := fmt.Sprintf("%d (%s) %s", pid, "cmd with space", strings.Join(statFields, " "))
-
-	stubReadFile(t, func(path string) ([]byte, error) {
-		switch path {
-		case fmt.Sprintf("/proc/%d/stat", pid):
-			return []byte(statContent), nil
-		case "/proc/stat":
-			return []byte(fmt.Sprintf("cpu 0 0 0 0\nbtime %d\n", boot.Unix())), nil
-		default:
-			return nil, os.ErrNotExist
-		}
-	})
-
-	got := getProcessStartTime(pid)
-	want := boot.Add(time.Duration(startTicks/100) * time.Second)
-	if !got.Equal(want) {
-		t.Fatalf("getProcessStartTime() = %v, want %v", got, want)
-	}
-}
-
-func TestGetProcessStartTimeInvalidData(t *testing.T) {
-	pid := 99
-	stubReadFile(t, func(path string) ([]byte, error) {
-		switch path {
-		case fmt.Sprintf("/proc/%d/stat", pid):
-			return []byte("garbage"), nil
-		case "/proc/stat":
-			return []byte("btime not-a-number\n"), nil
-		default:
-			return nil, os.ErrNotExist
-		}
-	})
-
-	if got := getProcessStartTime(pid); !got.IsZero() {
-		t.Fatalf("invalid /proc data should return zero time, got %v", got)
-	}
-}
-
-func TestGetBootTimeParsesBtime(t *testing.T) {
-	const bootSec = 1_711_111_111
-	stubReadFile(t, func(path string) ([]byte, error) {
-		if path != "/proc/stat" {
-			return nil, os.ErrNotExist
-		}
-		content := fmt.Sprintf("intr 0\nbtime %d\n", bootSec)
-		return []byte(content), nil
-	})
-
-	got := getBootTime()
-	want := time.Unix(bootSec, 0)
-	if !got.Equal(want) {
-		t.Fatalf("getBootTime() = %v, want %v", got, want)
-	}
-}
-
-func TestGetBootTimeInvalidData(t *testing.T) {
-	cases := []struct {
-		name    string
-		content string
-	}{
-		{"missing", "cpu 0 0 0 0"},
-		{"malformed", "btime abc"},
-	}
-
-	for _, tt := range cases {
-		t.Run(tt.name, func(t *testing.T) {
-			stubReadFile(t, func(string) ([]byte, error) {
-				return []byte(tt.content), nil
-			})
-			if got := getBootTime(); !got.IsZero() {
-				t.Fatalf("getBootTime() unexpected value for %s: %v", tt.name, got)
-			}
-		})
-	}
-}
-
-func stubReadFile(t *testing.T, fn func(string) ([]byte, error)) {
-	t.Helper()
-	original := readFileFn
-	readFileFn = fn
-	t.Cleanup(func() {
-		readFileFn = original
-	})
-}
--- a/codeagent-wrapper/process_check_unix.go
+++ b/codeagent-wrapper/process_check_unix.go
@@ -1,104 +0,0 @@
-//go:build unix || darwin || linux
-// +build unix darwin linux
-
-package main
-
-import (
-	"errors"
-	"fmt"
-	"os"
-	"strconv"
-	"strings"
-	"syscall"
-	"time"
-)
-
-var findProcess = os.FindProcess
-var readFileFn = os.ReadFile
-
-// isProcessRunning returns true if a process with the given pid is running on Unix-like systems.
-func isProcessRunning(pid int) bool {
-	if pid <= 0 {
-		return false
-	}
-
-	proc, err := findProcess(pid)
-	if err != nil || proc == nil {
-		return false
-	}
-
-	err = proc.Signal(syscall.Signal(0))
-	if err != nil && (errors.Is(err, syscall.ESRCH) || errors.Is(err, os.ErrProcessDone)) {
-		return false
-	}
-	return true
-}
-
-// getProcessStartTime returns the start time of a process on Unix-like systems.
-// Returns zero time if the start time cannot be determined.
-func getProcessStartTime(pid int) time.Time {
-	if pid <= 0 {
-		return time.Time{}
-	}
-
-	// Read /proc/<pid>/stat to get process start time
-	statPath := fmt.Sprintf("/proc/%d/stat", pid)
-	data, err := readFileFn(statPath)
-	if err != nil {
-		return time.Time{}
-	}
-
-	// Parse stat file: fields are space-separated, but comm (field 2) can contain spaces
-	// Find the last ')' to skip comm field safely
-	content := string(data)
-	lastParen := strings.LastIndex(content, ")")
-	if lastParen == -1 {
-		return time.Time{}
-	}
-
-	fields := strings.Fields(content[lastParen+1:])
-	if len(fields) < 20 {
-		return time.Time{}
-	}
-
-	// Field 22 (index 19 after comm) is starttime in clock ticks since boot
-	startTicks, err := strconv.ParseUint(fields[19], 10, 64)
-	if err != nil {
-		return time.Time{}
-	}
-
-	// Get system boot time
-	bootTime := getBootTime()
-	if bootTime.IsZero() {
-		return time.Time{}
-	}
-
-	// Convert ticks to duration (typically 100 ticks/sec on most systems)
-	ticksPerSec := uint64(100) // sysconf(_SC_CLK_TCK), typically 100
-	startTime := bootTime.Add(time.Duration(startTicks/ticksPerSec) * time.Second)
-
-	return startTime
-}
-
-// getBootTime returns the system boot time by reading /proc/stat.
-func getBootTime() time.Time {
-	data, err := readFileFn("/proc/stat")
-	if err != nil {
-		return time.Time{}
-	}
-
-	lines := strings.Split(string(data), "\n")
-	for _, line := range lines {
-		if strings.HasPrefix(line, "btime ") {
-			fields := strings.Fields(line)
-			if len(fields) >= 2 {
-				bootSec, err := strconv.ParseInt(fields[1], 10, 64)
-				if err == nil {
-					return time.Unix(bootSec, 0)
-				}
-			}
-		}
-	}
-
-	return time.Time{}
-}
--- a/codeagent-wrapper/process_check_windows.go
+++ b/codeagent-wrapper/process_check_windows.go
@@ -1,87 +0,0 @@
-//go:build windows
-// +build windows
-
-package main
-
-import (
-	"errors"
-	"os"
-	"syscall"
-	"time"
-	"unsafe"
-)
-
-const (
-	processQueryLimitedInformation = 0x1000
-	stillActive                    = 259 // STILL_ACTIVE exit code
-)
-
-var (
-	findProcess      = os.FindProcess
-	kernel32         = syscall.NewLazyDLL("kernel32.dll")
-	getProcessTimes  = kernel32.NewProc("GetProcessTimes")
-	fileTimeToUnixFn = fileTimeToUnix
-)
-
-// isProcessRunning returns true if a process with the given pid is running on Windows.
-func isProcessRunning(pid int) bool {
-	if pid <= 0 {
-		return false
-	}
-
-	if _, err := findProcess(pid); err != nil {
-		return false
-	}
-
-	handle, err := syscall.OpenProcess(processQueryLimitedInformation, false, uint32(pid))
-	if err != nil {
-		if errors.Is(err, syscall.ERROR_ACCESS_DENIED) {
-			return true
-		}
-		return false
-	}
-	defer syscall.CloseHandle(handle)
-
-	var exitCode uint32
-	if err := syscall.GetExitCodeProcess(handle, &exitCode); err != nil {
-		return true
-	}
-
-	return exitCode == stillActive
-}
-
-// getProcessStartTime returns the start time of a process on Windows.
-// Returns zero time if the start time cannot be determined.
-func getProcessStartTime(pid int) time.Time {
-	if pid <= 0 {
-		return time.Time{}
-	}
-
-	handle, err := syscall.OpenProcess(processQueryLimitedInformation, false, uint32(pid))
-	if err != nil {
-		return time.Time{}
-	}
-	defer syscall.CloseHandle(handle)
-
-	var creationTime, exitTime, kernelTime, userTime syscall.Filetime
-	ret, _, _ := getProcessTimes.Call(
-		uintptr(handle),
-		uintptr(unsafe.Pointer(&creationTime)),
-		uintptr(unsafe.Pointer(&exitTime)),
-		uintptr(unsafe.Pointer(&kernelTime)),
-		uintptr(unsafe.Pointer(&userTime)),
-	)
-
-	if ret == 0 {
-		return time.Time{}
-	}
-
-	return fileTimeToUnixFn(creationTime)
-}
-
-// fileTimeToUnix converts Windows FILETIME to Unix time.
-func fileTimeToUnix(ft syscall.Filetime) time.Time {
-	// FILETIME is 100-nanosecond intervals since January 1, 1601 UTC
-	nsec := ft.Nanoseconds()
-	return time.Unix(0, nsec)
-}
--- a/codeagent-wrapper/process_check_windows_test.go
+++ b/codeagent-wrapper/process_check_windows_test.go
@@ -1,64 +0,0 @@
-//go:build windows
-// +build windows
-
-package main
-
-import (
-	"os"
-	"testing"
-	"time"
-)
-
-func TestIsProcessRunning(t *testing.T) {
-	t.Run("boundary values", func(t *testing.T) {
-		if isProcessRunning(0) {
-			t.Fatalf("expected pid 0 to be reported as not running")
-		}
-		if isProcessRunning(-1) {
-			t.Fatalf("expected pid -1 to be reported as not running")
-		}
-	})
-
-	t.Run("current process", func(t *testing.T) {
-		if !isProcessRunning(os.Getpid()) {
-			t.Fatalf("expected current process (pid=%d) to be running", os.Getpid())
-		}
-	})
-
-	t.Run("fake pid", func(t *testing.T) {
-		const nonexistentPID = 1 << 30
-		if isProcessRunning(nonexistentPID) {
-			t.Fatalf("expected pid %d to be reported as not running", nonexistentPID)
-		}
-	})
-}
-
-func TestGetProcessStartTimeReadsProcStat(t *testing.T) {
-	start := getProcessStartTime(os.Getpid())
-	if start.IsZero() {
-		t.Fatalf("expected non-zero start time for current process")
-	}
-	if start.After(time.Now().Add(5 * time.Second)) {
-		t.Fatalf("start time is unexpectedly in the future: %v", start)
-	}
-}
-
-func TestGetProcessStartTimeInvalidData(t *testing.T) {
-	if !getProcessStartTime(0).IsZero() {
-		t.Fatalf("expected zero time for pid 0")
-	}
-	if !getProcessStartTime(-1).IsZero() {
-		t.Fatalf("expected zero time for negative pid")
-	}
-	if !getProcessStartTime(1 << 30).IsZero() {
-		t.Fatalf("expected zero time for non-existent pid")
-	}
-}
-
-func TestGetBootTimeParsesBtime(t *testing.T) {
-	t.Skip("getBootTime is only implemented on Unix-like systems")
-}
-
-func TestGetBootTimeInvalidData(t *testing.T) {
-	t.Skip("getBootTime is only implemented on Unix-like systems")
-}
--- a/codeagent-wrapper/wrapper_name.go
+++ b/codeagent-wrapper/wrapper_name.go
@@ -1,126 +0,0 @@
-package main
-
-import (
-	"os"
-	"path/filepath"
-	"strings"
-)
-
-const (
-	defaultWrapperName = "codeagent-wrapper"
-	legacyWrapperName  = "codex-wrapper"
-)
-
-var executablePathFn = os.Executable
-
-func normalizeWrapperName(path string) string {
-	if path == "" {
-		return ""
-	}
-
-	base := filepath.Base(path)
-	base = strings.TrimSuffix(base, ".exe") // tolerate Windows executables
-
-	switch base {
-	case defaultWrapperName, legacyWrapperName:
-		return base
-	default:
-		return ""
-	}
-}
-
-// currentWrapperName resolves the wrapper name based on the invoked binary.
-// Only known names are honored to avoid leaking build/test binary names into logs.
-func currentWrapperName() string {
-	if len(os.Args) == 0 {
-		return defaultWrapperName
-	}
-
-	if name := normalizeWrapperName(os.Args[0]); name != "" {
-		return name
-	}
-
-	execPath, err := executablePathFn()
-	if err == nil {
-		if name := normalizeWrapperName(execPath); name != "" {
-			return name
-		}
-
-		if resolved, err := filepath.EvalSymlinks(execPath); err == nil {
-			if name := normalizeWrapperName(resolved); name != "" {
-				return name
-			}
-			if alias := resolveAlias(execPath, resolved); alias != "" {
-				return alias
-			}
-		}
-
-		if alias := resolveAlias(execPath, ""); alias != "" {
-			return alias
-		}
-	}
-
-	return defaultWrapperName
-}
-
-// logPrefixes returns the set of accepted log name prefixes, including the
-// current wrapper name and legacy aliases.
-func logPrefixes() []string {
-	prefixes := []string{currentWrapperName(), defaultWrapperName, legacyWrapperName}
-	seen := make(map[string]struct{}, len(prefixes))
-	var unique []string
-	for _, prefix := range prefixes {
-		if prefix == "" {
-			continue
-		}
-		if _, ok := seen[prefix]; ok {
-			continue
-		}
-		seen[prefix] = struct{}{}
-		unique = append(unique, prefix)
-	}
-	return unique
-}
-
-// primaryLogPrefix returns the preferred filename prefix for log files.
-// Defaults to the current wrapper name when available, otherwise falls back
-// to the canonical default name.
-func primaryLogPrefix() string {
-	prefixes := logPrefixes()
-	if len(prefixes) == 0 {
-		return defaultWrapperName
-	}
-	return prefixes[0]
-}
-
-func resolveAlias(execPath string, target string) string {
-	if execPath == "" {
-		return ""
-	}
-
-	dir := filepath.Dir(execPath)
-	for _, candidate := range []string{defaultWrapperName, legacyWrapperName} {
-		aliasPath := filepath.Join(dir, candidate)
-		info, err := os.Lstat(aliasPath)
-		if err != nil {
-			continue
-		}
-		if info.Mode()&os.ModeSymlink == 0 {
-			continue
-		}
-
-		resolved, err := filepath.EvalSymlinks(aliasPath)
-		if err != nil {
-			continue
-		}
-		if target != "" && resolved != target {
-			continue
-		}
-
-		if name := normalizeWrapperName(aliasPath); name != "" {
-			return name
-		}
-	}
-
-	return ""
-}
--- a/codeagent-wrapper/wrapper_name_test.go
+++ b/codeagent-wrapper/wrapper_name_test.go
@@ -1,50 +0,0 @@
-package main
-
-import (
-	"os"
-	"path/filepath"
-	"testing"
-)
-
-func TestCurrentWrapperNameFallsBackToExecutable(t *testing.T) {
-	defer resetTestHooks()
-
-	tempDir := t.TempDir()
-	execPath := filepath.Join(tempDir, "codeagent-wrapper")
-	if err := os.WriteFile(execPath, []byte("#!/bin/true\n"), 0o755); err != nil {
-		t.Fatalf("failed to write fake binary: %v", err)
-	}
-
-	os.Args = []string{filepath.Join(tempDir, "custom-name")}
-	executablePathFn = func() (string, error) {
-		return execPath, nil
-	}
-
-	if got := currentWrapperName(); got != defaultWrapperName {
-		t.Fatalf("currentWrapperName() = %q, want %q", got, defaultWrapperName)
-	}
-}
-
-func TestCurrentWrapperNameDetectsLegacyAliasSymlink(t *testing.T) {
-	defer resetTestHooks()
-
-	tempDir := t.TempDir()
-	execPath := filepath.Join(tempDir, "wrapper")
-	aliasPath := filepath.Join(tempDir, legacyWrapperName)
-
-	if err := os.WriteFile(execPath, []byte("#!/bin/true\n"), 0o755); err != nil {
-		t.Fatalf("failed to write fake binary: %v", err)
-	}
-	if err := os.Symlink(execPath, aliasPath); err != nil {
-		t.Fatalf("failed to create alias: %v", err)
-	}
-
-	os.Args = []string{filepath.Join(tempDir, "unknown-runner")}
-	executablePathFn = func() (string, error) {
-		return execPath, nil
-	}
-
-	if got := currentWrapperName(); got != legacyWrapperName {
-		t.Fatalf("currentWrapperName() = %q, want %q", got, legacyWrapperName)
-	}
-}
--- a/config.json
+++ b/config.json
@@ -93,7 +93,7 @@
      ]
    },
    "essentials": {
-      "enabled": true,
+      "enabled": false,
      "description": "Core development commands and utilities",
      "operations": [
        {
@@ -156,6 +156,68 @@
          "description": "Install develop agent prompt"
        }
      ]
+    },
+    "sparv": {
+      "enabled": false,
+      "description": "SPARV workflow (Specify→Plan→Act→Review→Vault) with 10-point gate",
+      "operations": [
+        {
+          "type": "copy_dir",
+          "source": "skills/sparv",
+          "target": "skills/sparv",
+          "description": "Install sparv skill with all scripts and hooks"
+        }
+      ]
+    },
+    "do": {
+      "enabled": false,
+      "description": "7-phase feature development workflow with codeagent orchestration",
+      "operations": [
+        {
+          "type": "copy_dir",
+          "source": "skills/do",
+          "target": "skills/do",
+          "description": "Install do skill with hooks"
+        }
+      ]
+    },
+    "course": {
+      "enabled": false,
+      "description": "课程开发工作流，包含 dev、产品需求和测试用例技能",
+      "operations": [
+        {
+          "type": "copy_dir",
+          "source": "skills/dev",
+          "target": "skills/dev",
+          "description": "Install dev skill with agents"
+        },
+        {
+          "type": "copy_file",
+          "source": "skills/product-requirements/SKILL.md",
+          "target": "skills/product-requirements/SKILL.md",
+          "description": "Install product-requirements skill"
+        },
+        {
+          "type": "copy_dir",
+          "source": "skills/test-cases",
+          "target": "skills/test-cases",
+          "description": "Install test-cases skill with references"
+        },
+        {
+          "type": "copy_file",
+          "source": "skills/codeagent/SKILL.md",
+          "target": "skills/codeagent/SKILL.md",
+          "description": "Install codeagent skill"
+        },
+        {
+          "type": "run_command",
+          "command": "bash install.sh",
+          "description": "Install codeagent-wrapper binary",
+          "env": {
+            "INSTALL_DIR": "${install_dir}"
+          }
+        }
+      ]
    }
  }
 }
--- a/go.work.sum
+++ b/go.work.sum
@@ -0,0 +1,143 @@
+cloud.google.com/go v0.112.1 h1:uJSeirPke5UNZHIb4SxfZklVSiWWVqW4oXlETwZziwM=
+cloud.google.com/go v0.112.1/go.mod h1:+Vbu+Y1UU+I1rjmzeMOb/8RfkKJK2Gyxi1X6jJCZLo4=
+cloud.google.com/go/compute v1.24.0 h1:phWcR2eWzRJaL/kOiJwfFsPs4BaKq1j6vnpZrc1YlVg=
+cloud.google.com/go/compute v1.24.0/go.mod h1:kw1/T+h/+tK2LJK0wiPPx1intgdAM3j/g3hFDlscY40=
+cloud.google.com/go/compute/metadata v0.2.3 h1:mg4jlk7mCAj6xXp9UJ4fjI9VUI5rubuGBW5aJ7UnBMY=
+cloud.google.com/go/compute/metadata v0.2.3/go.mod h1:VAV5nSsACxMJvgaAuX6Pk2AawlZn8kiOGuCv6gTkwuA=
+cloud.google.com/go/firestore v1.15.0 h1:/k8ppuWOtNuDHt2tsRV42yI21uaGnKDEQnRFeBpbFF8=
+cloud.google.com/go/firestore v1.15.0/go.mod h1:GWOxFXcv8GZUtYpWHw/w6IuYNux/BtmeVTMmjrm4yhk=
+cloud.google.com/go/iam v1.1.5 h1:1jTsCu4bcsNsE4iiqNT5SHwrDRCfRmIaaaVFhRveTJI=
+cloud.google.com/go/iam v1.1.5/go.mod h1:rB6P/Ic3mykPbFio+vo7403drjlgvoWfYpJhMXEbzv8=
+cloud.google.com/go/longrunning v0.5.5 h1:GOE6pZFdSrTb4KAiKnXsJBtlE6mEyaW44oKyMILWnOg=
+cloud.google.com/go/longrunning v0.5.5/go.mod h1:WV2LAxD8/rg5Z1cNW6FJ/ZpX4E4VnDnoTk0yawPBB7s=
+cloud.google.com/go/storage v1.35.1 h1:B59ahL//eDfx2IIKFBeT5Atm9wnNmj3+8xG/W4WB//w=
+cloud.google.com/go/storage v1.35.1/go.mod h1:M6M/3V/D3KpzMTJyPOR/HU6n2Si5QdaXYEsng2xgOs8=
+github.com/armon/go-metrics v0.4.1 h1:hR91U9KYmb6bLBYLQjyM+3j+rcd/UhE+G78SFnF8gJA=
+github.com/armon/go-metrics v0.4.1/go.mod h1:E6amYzXo6aW1tqzoZGT755KkbgrJsSdpwZ+3JqfkOG4=
+github.com/coreos/go-semver v0.3.0 h1:wkHLiw0WNATZnSG7epLsujiMCgPAc9xhjJ4tgnAxmfM=
+github.com/coreos/go-semver v0.3.0/go.mod h1:nnelYz7RCh+5ahJtPPxZlU+153eP4D4r3EedlOD2RNk=
+github.com/coreos/go-systemd/v22 v22.5.0 h1:RrqgGjYQKalulkV8NGVIfkXQf6YYmOyiJKk8iXXhfZs=
+github.com/cpuguy83/go-md2man/v2 v2.0.4 h1:wfIWP927BUkWJb2NmU/kNDYIBTh/ziUX91+lVfRxZq4=
+github.com/fatih/color v1.14.1 h1:qfhVLaG5s+nCROl1zJsZRxFeYrHLqWroPOQ8BWiNb4w=
+github.com/fatih/color v1.14.1/go.mod h1:2oHN61fhTpgcxD3TSWCgKDiH1+x4OiDVVGH8WlgGZGg=
+github.com/felixge/httpsnoop v1.0.4 h1:NFTV2Zj1bL4mc9sqWACXbQFVBBg2W3GPvqp8/ESS2Wg=
+github.com/felixge/httpsnoop v1.0.4/go.mod h1:m8KPJKqk1gH5J9DgRY2ASl2lWCfGKXixSwevea8zH2U=
+github.com/go-logr/logr v1.4.1 h1:pKouT5E8xu9zeFC39JXRDukb6JFQPXM5p5I91188VAQ=
+github.com/go-logr/logr v1.4.1/go.mod h1:9T104GzyrTigFIr8wt5mBrctHMim0Nb2HLGrmQ40KvY=
+github.com/go-logr/stdr v1.2.2 h1:hSWxHoqTgW2S2qGc0LTAI563KZ5YKYRhT3MFKZMbjag=
+github.com/go-logr/stdr v1.2.2/go.mod h1:mMo/vtBO5dYbehREoey6XUKy/eSumjCCveDpRre4VKE=
+github.com/godbus/dbus/v5 v5.0.4 h1:9349emZab16e7zQvpmsbtjc18ykshndd8y2PG3sgJbA=
+github.com/gogo/protobuf v1.3.2 h1:Ov1cvc58UF3b5XjBnZv7+opcTcQFZebYjWzi34vdm4Q=
+github.com/gogo/protobuf v1.3.2/go.mod h1:P1XiOD3dCwIKUDQYPy72D8LYyHL2YPYrpS2s69NZV8Q=
+github.com/golang/groupcache v0.0.0-20210331224755-41bb18bfe9da h1:oI5xCqsCo564l8iNU+DwB5epxmsaqB+rhGL0m5jtYqE=
+github.com/golang/groupcache v0.0.0-20210331224755-41bb18bfe9da/go.mod h1:cIg4eruTrX1D+g88fzRXU5OdNfaM+9IcxsU14FzY7Hc=
+github.com/golang/protobuf v1.5.3 h1:KhyjKVUg7Usr/dYsdSqoFveMYd5ko72D+zANwlG1mmg=
+github.com/golang/protobuf v1.5.3/go.mod h1:XVQd3VNwM+JqD3oG2Ue2ip4fOMUkwXdXDdiuN0vRsmY=
+github.com/google/s2a-go v0.1.7 h1:60BLSyTrOV4/haCDW4zb1guZItoSq8foHCXrAnjBo/o=
+github.com/google/s2a-go v0.1.7/go.mod h1:50CgR4k1jNlWBu4UfS4AcfhVe1r6pdZPygJ3R8F0Qdw=
+github.com/google/uuid v1.4.0 h1:MtMxsa51/r9yyhkyLsVeVt0B+BGQZzpQiTQ4eHZ8bc4=
+github.com/google/uuid v1.4.0/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
+github.com/googleapis/enterprise-certificate-proxy v0.3.2 h1:Vie5ybvEvT75RniqhfFxPRy3Bf7vr3h0cechB90XaQs=
+github.com/googleapis/enterprise-certificate-proxy v0.3.2/go.mod h1:VLSiSSBs/ksPL8kq3OBOQ6WRI2QnaFynd1DCjZ62+V0=
+github.com/googleapis/gax-go/v2 v2.12.3 h1:5/zPPDvw8Q1SuXjrqrZslrqT7dL/uJT2CQii/cLCKqA=
+github.com/googleapis/gax-go/v2 v2.12.3/go.mod h1:AKloxT6GtNbaLm8QTNSidHUVsHYcBHwWRvkNFJUQcS4=
+github.com/googleapis/google-cloud-go-testing v0.0.0-20210719221736-1c9a4c676720 h1:zC34cGQu69FG7qzJ3WiKW244WfhDC3xxYMeNOX2gtUQ=
+github.com/googleapis/google-cloud-go-testing v0.0.0-20210719221736-1c9a4c676720/go.mod h1:dvDLG8qkwmyD9a/MJJN3XJcT3xFxOKAvTZGvuZmac9g=
+github.com/hashicorp/consul/api v1.28.2 h1:mXfkRHrpHN4YY3RqL09nXU1eHKLNiuAN4kHvDQ16k/8=
+github.com/hashicorp/consul/api v1.28.2/go.mod h1:KyzqzgMEya+IZPcD65YFoOVAgPpbfERu4I/tzG6/ueE=
+github.com/hashicorp/errwrap v1.1.0 h1:OxrOeh75EUXMY8TBjag2fzXGZ40LB6IKw45YeGUDY2I=
+github.com/hashicorp/errwrap v1.1.0/go.mod h1:YH+1FKiLXxHSkmPseP+kNlulaMuP3n2brvKWEqk/Jc4=
+github.com/hashicorp/go-cleanhttp v0.5.2 h1:035FKYIWjmULyFRBKPs8TBQoi0x6d9G4xc9neXJWAZQ=
+github.com/hashicorp/go-cleanhttp v0.5.2/go.mod h1:kO/YDlP8L1346E6Sodw+PrpBSV4/SoxCXGY6BqNFT48=
+github.com/hashicorp/go-hclog v1.5.0 h1:bI2ocEMgcVlz55Oj1xZNBsVi900c7II+fWDyV9o+13c=
+github.com/hashicorp/go-hclog v1.5.0/go.mod h1:W4Qnvbt70Wk/zYJryRzDRU/4r0kIg0PVHBcfoyhpF5M=
+github.com/hashicorp/go-immutable-radix v1.3.1 h1:DKHmCUm2hRBK510BaiZlwvpD40f8bJFeZnpfm2KLowc=
+github.com/hashicorp/go-immutable-radix v1.3.1/go.mod h1:0y9vanUI8NX6FsYoO3zeMjhV/C5i9g4Q3DwcSNZ4P60=
+github.com/hashicorp/go-multierror v1.1.1 h1:H5DkEtf6CXdFp0N0Em5UCwQpXMWke8IA0+lD48awMYo=
+github.com/hashicorp/go-multierror v1.1.1/go.mod h1:iw975J/qwKPdAO1clOe2L8331t/9/fmwbPZ6JB6eMoM=
+github.com/hashicorp/go-rootcerts v1.0.2 h1:jzhAVGtqPKbwpyCPELlgNWhE1znq+qwJtW5Oi2viEzc=
+github.com/hashicorp/go-rootcerts v1.0.2/go.mod h1:pqUvnprVnM5bf7AOirdbb01K4ccR319Vf4pU3K5EGc8=
+github.com/hashicorp/golang-lru v0.5.4 h1:YDjusn29QI/Das2iO9M0BHnIbxPeyuCHsjMW+lJfyTc=
+github.com/hashicorp/golang-lru v0.5.4/go.mod h1:iADmTwqILo4mZ8BN3D2Q6+9jd8WM5uGBxy+E8yxSoD4=
+github.com/hashicorp/serf v0.10.1 h1:Z1H2J60yRKvfDYAOZLd2MU0ND4AH/WDz7xYHDWQsIPY=
+github.com/hashicorp/serf v0.10.1/go.mod h1:yL2t6BqATOLGc5HF7qbFkTfXoPIY0WZdWHfEvMqbG+4=
+github.com/json-iterator/go v1.1.12 h1:PV8peI4a0ysnczrg+LtxykD8LfKY9ML6u2jnxaEnrnM=
+github.com/json-iterator/go v1.1.12/go.mod h1:e30LSqwooZae/UwlEbR2852Gd8hjQvJoHmT4TnhNGBo=
+github.com/klauspost/compress v1.17.2 h1:RlWWUY/Dr4fL8qk9YG7DTZ7PDgME2V4csBXA8L/ixi4=
+github.com/klauspost/compress v1.17.2/go.mod h1:ntbaceVETuRiXiv4DpjP66DpAtAGkEQskQzEyD//IeE=
+github.com/kr/fs v0.1.0 h1:Jskdu9ieNAYnjxsi0LbQp1ulIKZV1LAFgK1tWhpZgl8=
+github.com/kr/fs v0.1.0/go.mod h1:FFnZGqtBN9Gxj7eW1uZ42v5BccTP0vu6NEaFoC2HwRg=
+github.com/mitchellh/go-homedir v1.1.0 h1:lukF9ziXFxDFPkA1vsr5zpc1XuPDn/wFntq5mG+4E0Y=
+github.com/mitchellh/go-homedir v1.1.0/go.mod h1:SfyaCUpYCn1Vlf4IUYiD9fPX4A5wJrkLzIz1N1q0pr0=
+github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd h1:TRLaZ9cD/w8PVh93nsPXa1VrQ6jlwL5oN8l14QlcNfg=
+github.com/modern-go/concurrent v0.0.0-20180306012644-bacd9c7ef1dd/go.mod h1:6dJC0mAP4ikYIbvyc7fijjWJddQyLn8Ig3JB5CqoB9Q=
+github.com/modern-go/reflect2 v1.0.2 h1:xBagoLtFs94CBntxluKeaWgTMpvLxC4ur3nMaC9Gz0M=
+github.com/modern-go/reflect2 v1.0.2/go.mod h1:yWuevngMOJpCy52FWWMvUC8ws7m/LJsjYzDa0/r8luk=
+github.com/nats-io/nats.go v1.34.0 h1:fnxnPCNiwIG5w08rlMcEKTUw4AV/nKyGCOJE8TdhSPk=
+github.com/nats-io/nats.go v1.34.0/go.mod h1:Ubdu4Nh9exXdSz0RVWRFBbRfrbSxOYd26oF0wkWclB8=
+github.com/nats-io/nkeys v0.4.7 h1:RwNJbbIdYCoClSDNY7QVKZlyb/wfT6ugvFCiKy6vDvI=
+github.com/nats-io/nkeys v0.4.7/go.mod h1:kqXRgRDPlGy7nGaEDMuYzmiJCIAAWDK0IMBtDmGD0nc=
+github.com/nats-io/nuid v1.0.1 h1:5iA8DT8V7q8WK2EScv2padNa/rTESc1KdnPw4TC2paw=
+github.com/nats-io/nuid v1.0.1/go.mod h1:19wcPz3Ph3q0Jbyiqsd0kePYG7A95tJPxeL+1OSON2c=
+github.com/pkg/errors v0.9.1 h1:FEBLx1zS214owpjy7qsBeixbURkuhQAwrK5UwLGTwt4=
+github.com/pkg/sftp v1.13.6 h1:JFZT4XbOU7l77xGSpOdW+pwIMqP044IyjXX6FGyEKFo=
+github.com/pkg/sftp v1.13.6/go.mod h1:tz1ryNURKu77RL+GuCzmoJYxQczL3wLNNpPWagdg4Qk=
+github.com/rs/xid v1.6.0 h1:fV591PaemRlL6JfRxGDEPl69wICngIQ3shQtzfy2gxU=
+github.com/russross/blackfriday/v2 v2.1.0 h1:JIOH55/0cWyOuilr9/qlrm0BSXldqnqwMsf35Ld67mk=
+github.com/sagikazarmark/crypt v0.19.0 h1:WMyLTjHBo64UvNcWqpzY3pbZTYgnemZU8FBZigKc42E=
+github.com/sagikazarmark/crypt v0.19.0/go.mod h1:c6vimRziqqERhtSe0MhIvzE1w54FrCHtrXb5NH/ja78=
+github.com/stretchr/objx v0.5.2 h1:xuMeJ0Sdp5ZMRXx/aWO6RZxdr3beISkG5/G/aIRr3pY=
+go.etcd.io/etcd/api/v3 v3.5.12 h1:W4sw5ZoU2Juc9gBWuLk5U6fHfNVyY1WC5g9uiXZio/c=
+go.etcd.io/etcd/api/v3 v3.5.12/go.mod h1:Ot+o0SWSyT6uHhA56al1oCED0JImsRiU9Dc26+C2a+4=
+go.etcd.io/etcd/client/pkg/v3 v3.5.12 h1:EYDL6pWwyOsylrQyLp2w+HkQ46ATiOvoEdMarindU2A=
+go.etcd.io/etcd/client/pkg/v3 v3.5.12/go.mod h1:seTzl2d9APP8R5Y2hFL3NVlD6qC/dOT+3kvrqPyTas4=
+go.etcd.io/etcd/client/v2 v2.305.12 h1:0m4ovXYo1CHaA/Mp3X/Fak5sRNIWf01wk/X1/G3sGKI=
+go.etcd.io/etcd/client/v2 v2.305.12/go.mod h1:aQ/yhsxMu+Oht1FOupSr60oBvcS9cKXHrzBpDsPTf9E=
+go.etcd.io/etcd/client/v3 v3.5.12 h1:v5lCPXn1pf1Uu3M4laUE2hp/geOTc5uPcYYsNe1lDxg=
+go.etcd.io/etcd/client/v3 v3.5.12/go.mod h1:tSbBCakoWmmddL+BKVAJHa9km+O/E+bumDe9mSbPiqw=
+go.opencensus.io v0.24.0 h1:y73uSU6J157QMP2kn2r30vwW1A2W2WFwSCGnAVxeaD0=
+go.opencensus.io v0.24.0/go.mod h1:vNK8G9p7aAivkbmorf4v+7Hgx+Zs0yY+0fOtgBfjQKo=
+go.opentelemetry.io/contrib/instrumentation/google.golang.org/grpc/otelgrpc v0.49.0 h1:4Pp6oUg3+e/6M4C0A/3kJ2VYa++dsWVTtGgLVj5xtHg=
+go.opentelemetry.io/contrib/instrumentation/google.golang.org/grpc/otelgrpc v0.49.0/go.mod h1:Mjt1i1INqiaoZOMGR1RIUJN+i3ChKoFRqzrRQhlkbs0=
+go.opentelemetry.io/contrib/instrumentation/net/http/otelhttp v0.49.0 h1:jq9TW8u3so/bN+JPT166wjOI6/vQPF6Xe7nMNIltagk=
+go.opentelemetry.io/contrib/instrumentation/net/http/otelhttp v0.49.0/go.mod h1:p8pYQP+m5XfbZm9fxtSKAbM6oIllS7s2AfxrChvc7iw=
+go.opentelemetry.io/otel v1.24.0 h1:0LAOdjNmQeSTzGBzduGe/rU4tZhMwL5rWgtp9Ku5Jfo=
+go.opentelemetry.io/otel v1.24.0/go.mod h1:W7b9Ozg4nkF5tWI5zsXkaKKDjdVjpD4oAt9Qi/MArHo=
+go.opentelemetry.io/otel/metric v1.24.0 h1:6EhoGWWK28x1fbpA4tYTOWBkPefTDQnb8WSGXlc88kI=
+go.opentelemetry.io/otel/metric v1.24.0/go.mod h1:VYhLe1rFfxuTXLgj4CBiyz+9WYBA8pNGJgDcSFRKBco=
+go.opentelemetry.io/otel/trace v1.24.0 h1:CsKnnL4dUAr/0llH9FKuc698G04IrpWV0MQA/Y1YELI=
+go.opentelemetry.io/otel/trace v1.24.0/go.mod h1:HPc3Xr/cOApsBI154IU0OI0HJexz+aw5uPdbs3UCjNU=
+go.uber.org/zap v1.21.0 h1:WefMeulhovoZ2sYXz7st6K0sLj7bBhpiFaud4r4zST8=
+go.uber.org/zap v1.21.0/go.mod h1:wjWOCqI0f2ZZrJF/UufIOkiC8ii6tm1iqIsLo76RfJw=
+golang.org/x/crypto v0.21.0 h1:X31++rzVUdKhX5sWmSOFZxx8UW/ldWx55cbf08iNAMA=
+golang.org/x/crypto v0.21.0/go.mod h1:0BP7YvVV9gBbVKyeTG0Gyn+gZm94bibOW5BjDEYAOMs=
+golang.org/x/mod v0.12.0 h1:rmsUpXtvNzj340zd98LZ4KntptpfRHwpFOHG188oHXc=
+golang.org/x/mod v0.12.0/go.mod h1:iBbtSCu2XBx23ZKBPSOrRkjjQPZFPuis4dIYUhu/chs=
+golang.org/x/net v0.23.0 h1:7EYJ93RZ9vYSZAIb2x3lnuvqO5zneoD6IvWjuhfxjTs=
+golang.org/x/net v0.23.0/go.mod h1:JKghWKKOSdJwpW2GEx0Ja7fmaKnMsbu+MWVZTokSYmg=
+golang.org/x/oauth2 v0.18.0 h1:09qnuIAgzdx1XplqJvW6CQqMCtGZykZWcXzPMPUusvI=
+golang.org/x/oauth2 v0.18.0/go.mod h1:Wf7knwG0MPoWIMMBgFlEaSUDaKskp0dCfrlJRJXbBi8=
+golang.org/x/sync v0.6.0 h1:5BMeUDZ7vkXGfEr1x9B4bRcTH4lpkTkpdh0T/J+qjbQ=
+golang.org/x/sync v0.6.0/go.mod h1:Czt+wKu1gCyEFDUtn0jG5QVvpJ6rzVqr5aXyt9drQfk=
+golang.org/x/time v0.5.0 h1:o7cqy6amK/52YcAKIPlM3a+Fpj35zvRj2TP+e1xFSfk=
+golang.org/x/time v0.5.0/go.mod h1:3BpzKBy/shNhVucY/MWOyx10tF3SFh9QdLuxbVysPQM=
+golang.org/x/tools v0.13.0 h1:Iey4qkscZuv0VvIt8E0neZjtPVQFSc870HQ448QgEmQ=
+golang.org/x/tools v0.13.0/go.mod h1:HvlwmtVNQAhOuCjW7xxvovg8wbNq7LwfXh/k7wXUl58=
+golang.org/x/xerrors v0.0.0-20191204190536-9bdfabe68543 h1:E7g+9GITq07hpfrRu66IVDexMakfv52eLZ2CXBWiKr4=
+golang.org/x/xerrors v0.0.0-20220907171357-04be3eba64a2 h1:H2TDz8ibqkAF6YGhCdN3jS9O0/s90v0rJh3X/OLHEUk=
+golang.org/x/xerrors v0.0.0-20220907171357-04be3eba64a2/go.mod h1:K8+ghG5WaK9qNqU5K3HdILfMLy1f3aNYFI/wnl100a8=
+google.golang.org/api v0.171.0 h1:w174hnBPqut76FzW5Qaupt7zY8Kql6fiVjgys4f58sU=
+google.golang.org/api v0.171.0/go.mod h1:Hnq5AHm4OTMt2BUVjael2CWZFD6vksJdWCWiUAmjC9o=
+google.golang.org/appengine v1.6.8 h1:IhEN5q69dyKagZPYMSdIjS2HqprW324FRQZJcGqPAsM=
+google.golang.org/appengine v1.6.8/go.mod h1:1jJ3jBArFh5pcgW8gCtRJnepW8FzD1V44FJffLiz/Ds=
+google.golang.org/genproto v0.0.0-20240213162025-012b6fc9bca9 h1:9+tzLLstTlPTRyJTh+ah5wIMsBW5c4tQwGTN3thOW9Y=
+google.golang.org/genproto v0.0.0-20240213162025-012b6fc9bca9/go.mod h1:mqHbVIp48Muh7Ywss/AD6I5kNVKZMmAa/QEW58Gxp2s=
+google.golang.org/genproto/googleapis/api v0.0.0-20240311132316-a219d84964c2 h1:rIo7ocm2roD9DcFIX67Ym8icoGCKSARAiPljFhh5suQ=
+google.golang.org/genproto/googleapis/api v0.0.0-20240311132316-a219d84964c2/go.mod h1:O1cOfN1Cy6QEYr7VxtjOyP5AdAuR0aJ/MYZaaof623Y=
+google.golang.org/genproto/googleapis/rpc v0.0.0-20240314234333-6e1732d8331c h1:lfpJ/2rWPa/kJgxyyXM8PrNnfCzcmxJ265mADgwmvLI=
+google.golang.org/genproto/googleapis/rpc v0.0.0-20240314234333-6e1732d8331c/go.mod h1:WtryC6hu0hhx87FDGxWCDptyssuo68sk10vYjF+T9fY=
+google.golang.org/grpc v1.62.1 h1:B4n+nfKzOICUXMgyrNd19h/I9oH0L1pizfk1d4zSgTk=
+google.golang.org/grpc v1.62.1/go.mod h1:IWTG0VlJLCh1SkC58F7np9ka9mx/WNkjl4PGJaiq+QE=
+google.golang.org/protobuf v1.33.0 h1:uNO2rsAINq/JlFpSdYEKIZ0uKD/R9cpdv0T+yoGwGmI=
+google.golang.org/protobuf v1.33.0/go.mod h1:c6P6GXX6sHbq/GpV6MGZEdwhWPcYBgnhAHhKbcUYpos=
+gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405 h1:yhCVgyC4o1eVCa2tZl7eS0r+SDo693bJlVdllGtEeKM=
--- a/install.py
+++ b/install.py
@@ -23,6 +23,7 @@ except ImportError:  # pragma: no cover
    jsonschema = None

 DEFAULT_INSTALL_DIR = "~/.claude"
+SETTINGS_FILE = "settings.json"


 def _ensure_list(ctx: Dict[str, Any], key: str) -> List[Any]:
@@ -46,7 +47,7 @@ def parse_args(argv: Optional[Iterable[str]] = None) -> argparse.Namespace:
    )
    parser.add_argument(
        "--module",
-        help="Comma-separated modules to install, or 'all' for all enabled",
+        help="Comma-separated modules to install/uninstall, or 'all'",
    )
    parser.add_argument(
        "--config",
@@ -58,6 +59,16 @@ def parse_args(argv: Optional[Iterable[str]] = None) -> argparse.Namespace:
        action="store_true",
        help="List available modules and exit",
    )
+    parser.add_argument(
+        "--status",
+        action="store_true",
+        help="Show installation status of all modules",
+    )
+    parser.add_argument(
+        "--uninstall",
+        action="store_true",
+        help="Uninstall specified modules",
+    )
    parser.add_argument(
        "--force",
        action="store_true",
@@ -81,6 +92,151 @@ def _load_json(path: Path) -> Any:
        raise ValueError(f"Invalid JSON in {path}: {exc}") from exc


+def _save_json(path: Path, data: Any) -> None:
+    """Save data to JSON file with proper formatting."""
+    path.parent.mkdir(parents=True, exist_ok=True)
+    with path.open("w", encoding="utf-8") as fh:
+        json.dump(data, fh, indent=2, ensure_ascii=False)
+        fh.write("\n")
+
+
+# =============================================================================
+# Hooks Management
+# =============================================================================
+
+def load_settings(ctx: Dict[str, Any]) -> Dict[str, Any]:
+    """Load settings.json from install directory."""
+    settings_path = ctx["install_dir"] / SETTINGS_FILE
+    if settings_path.exists():
+        try:
+            return _load_json(settings_path)
+        except (ValueError, FileNotFoundError):
+            return {}
+    return {}
+
+
+def save_settings(ctx: Dict[str, Any], settings: Dict[str, Any]) -> None:
+    """Save settings.json to install directory."""
+    settings_path = ctx["install_dir"] / SETTINGS_FILE
+    _save_json(settings_path, settings)
+
+
+def find_module_hooks(module_name: str, cfg: Dict[str, Any], ctx: Dict[str, Any]) -> Optional[tuple]:
+    """Find hooks.json for a module if it exists.
+
+    Returns tuple of (hooks_config, plugin_root_path) or None.
+    """
+    # Check for hooks in operations (copy_dir targets)
+    for op in cfg.get("operations", []):
+        if op.get("type") == "copy_dir":
+            target_dir = ctx["install_dir"] / op["target"]
+            hooks_file = target_dir / "hooks" / "hooks.json"
+            if hooks_file.exists():
+                try:
+                    return (_load_json(hooks_file), str(target_dir))
+                except (ValueError, FileNotFoundError):
+                    pass
+
+    # Also check source directory during install
+    for op in cfg.get("operations", []):
+        if op.get("type") == "copy_dir":
+            target_dir = ctx["install_dir"] / op["target"]
+            source_dir = ctx["config_dir"] / op["source"]
+            hooks_file = source_dir / "hooks" / "hooks.json"
+            if hooks_file.exists():
+                try:
+                    return (_load_json(hooks_file), str(target_dir))
+                except (ValueError, FileNotFoundError):
+                    pass
+
+    return None
+
+
+def _create_hook_marker(module_name: str) -> str:
+    """Create a marker to identify hooks from a specific module."""
+    return f"__module:{module_name}__"
+
+
+def _replace_hook_variables(obj: Any, plugin_root: str) -> Any:
+    """Recursively replace ${CLAUDE_PLUGIN_ROOT} in hook config."""
+    if isinstance(obj, str):
+        return obj.replace("${CLAUDE_PLUGIN_ROOT}", plugin_root)
+    elif isinstance(obj, dict):
+        return {k: _replace_hook_variables(v, plugin_root) for k, v in obj.items()}
+    elif isinstance(obj, list):
+        return [_replace_hook_variables(item, plugin_root) for item in obj]
+    return obj
+
+
+def merge_hooks_to_settings(module_name: str, hooks_config: Dict[str, Any], ctx: Dict[str, Any], plugin_root: str = "") -> None:
+    """Merge module hooks into settings.json."""
+    settings = load_settings(ctx)
+    settings.setdefault("hooks", {})
+
+    module_hooks = hooks_config.get("hooks", {})
+    marker = _create_hook_marker(module_name)
+
+    # Replace ${CLAUDE_PLUGIN_ROOT} with actual path
+    if plugin_root:
+        module_hooks = _replace_hook_variables(module_hooks, plugin_root)
+
+    for hook_type, hook_entries in module_hooks.items():
+        settings["hooks"].setdefault(hook_type, [])
+
+        for entry in hook_entries:
+            # Add marker to identify this hook's source module
+            entry_copy = dict(entry)
+            entry_copy["__module__"] = module_name
+
+            # Check if already exists (avoid duplicates)
+            exists = False
+            for existing in settings["hooks"][hook_type]:
+                if existing.get("__module__") == module_name:
+                    # Same module, check if same hook
+                    if _hooks_equal(existing, entry_copy):
+                        exists = True
+                        break
+
+            if not exists:
+                settings["hooks"][hook_type].append(entry_copy)
+
+    save_settings(ctx, settings)
+    write_log({"level": "INFO", "message": f"Merged hooks for module: {module_name}"}, ctx)
+
+
+def unmerge_hooks_from_settings(module_name: str, ctx: Dict[str, Any]) -> None:
+    """Remove module hooks from settings.json."""
+    settings = load_settings(ctx)
+
+    if "hooks" not in settings:
+        return
+
+    modified = False
+    for hook_type in list(settings["hooks"].keys()):
+        original_len = len(settings["hooks"][hook_type])
+        settings["hooks"][hook_type] = [
+            entry for entry in settings["hooks"][hook_type]
+            if entry.get("__module__") != module_name
+        ]
+        if len(settings["hooks"][hook_type]) < original_len:
+            modified = True
+
+        # Remove empty hook type arrays
+        if not settings["hooks"][hook_type]:
+            del settings["hooks"][hook_type]
+
+    if modified:
+        save_settings(ctx, settings)
+        write_log({"level": "INFO", "message": f"Removed hooks for module: {module_name}"}, ctx)
+
+
+def _hooks_equal(hook1: Dict[str, Any], hook2: Dict[str, Any]) -> bool:
+    """Compare two hooks ignoring the __module__ marker."""
+    h1 = {k: v for k, v in hook1.items() if k != "__module__"}
+    h2 = {k: v for k, v in hook2.items() if k != "__module__"}
+    return h1 == h2
+
+
 def load_config(path: str) -> Dict[str, Any]:
    """Load config and validate against JSON Schema.

@@ -166,22 +322,93 @@ def resolve_paths(config: Dict[str, Any], args: argparse.Namespace) -> Dict[str,

 def list_modules(config: Dict[str, Any]) -> None:
    print("Available Modules:")
-    print(f"{'Name':<15} {'Default':<8} Description")
-    print("-" * 60)
-    for name, cfg in config.get("modules", {}).items():
+    print(f"{'#':<3} {'Name':<15} {'Default':<8} Description")
+    print("-" * 65)
+    for idx, (name, cfg) in enumerate(config.get("modules", {}).items(), 1):
        default = "✓" if cfg.get("enabled", False) else "✗"
        desc = cfg.get("description", "")
-        print(f"{name:<15} {default:<8} {desc}")
+        print(f"{idx:<3} {name:<15} {default:<8} {desc}")
    print("\n✓ = installed by default when no --module specified")


+def load_installed_status(ctx: Dict[str, Any]) -> Dict[str, Any]:
+    """Load installed modules status from status file."""
+    status_path = Path(ctx["status_file"])
+    if status_path.exists():
+        try:
+            return _load_json(status_path)
+        except (ValueError, FileNotFoundError):
+            return {"modules": {}}
+    return {"modules": {}}
+
+
+def check_module_installed(name: str, cfg: Dict[str, Any], ctx: Dict[str, Any]) -> bool:
+    """Check if a module is installed by verifying its files exist."""
+    install_dir = ctx["install_dir"]
+
+    for op in cfg.get("operations", []):
+        op_type = op.get("type")
+        if op_type in ("copy_dir", "copy_file"):
+            target = (install_dir / op["target"]).expanduser().resolve()
+            if target.exists():
+                return True
+    return False
+
+
+def get_installed_modules(config: Dict[str, Any], ctx: Dict[str, Any]) -> Dict[str, bool]:
+    """Get installation status of all modules by checking files."""
+    result = {}
+    modules = config.get("modules", {})
+
+    # First check status file
+    status = load_installed_status(ctx)
+    status_modules = status.get("modules", {})
+
+    for name, cfg in modules.items():
+        # Check both status file and filesystem
+        in_status = name in status_modules
+        files_exist = check_module_installed(name, cfg, ctx)
+        result[name] = in_status or files_exist
+
+    return result
+
+
+def list_modules_with_status(config: Dict[str, Any], ctx: Dict[str, Any]) -> None:
+    """List modules with installation status."""
+    installed_status = get_installed_modules(config, ctx)
+    status_data = load_installed_status(ctx)
+    status_modules = status_data.get("modules", {})
+
+    print("\n" + "=" * 70)
+    print("Module Status")
+    print("=" * 70)
+    print(f"{'#':<3} {'Name':<15} {'Status':<15} {'Installed At':<20} Description")
+    print("-" * 70)
+
+    for idx, (name, cfg) in enumerate(config.get("modules", {}).items(), 1):
+        desc = cfg.get("description", "")[:25]
+        if installed_status.get(name, False):
+            status = "✅ Installed"
+            installed_at = status_modules.get(name, {}).get("installed_at", "")[:16]
+        else:
+            status = "⬚ Not installed"
+            installed_at = ""
+        print(f"{idx:<3} {name:<15} {status:<15} {installed_at:<20} {desc}")
+
+    total = len(config.get("modules", {}))
+    installed_count = sum(1 for v in installed_status.values() if v)
+    print(f"\nTotal: {installed_count}/{total} modules installed")
+    print(f"Install dir: {ctx['install_dir']}")
+
+
 def select_modules(config: Dict[str, Any], module_arg: Optional[str]) -> Dict[str, Any]:
    modules = config.get("modules", {})
    if not module_arg:
-        return {k: v for k, v in modules.items() if v.get("enabled", False)}
+        # No --module specified: show interactive selection
+        return interactive_select_modules(config)

    if module_arg.strip().lower() == "all":
-        return {k: v for k, v in modules.items() if v.get("enabled", False)}
+        return dict(modules.items())

    selected: Dict[str, Any] = {}
    for name in (part.strip() for part in module_arg.split(",")):
@@ -193,6 +420,263 @@ def select_modules(config: Dict[str, Any], module_arg: Optional[str]) -> Dict[st
    return selected


+def interactive_select_modules(config: Dict[str, Any]) -> Dict[str, Any]:
+    """Interactive module selection when no --module is specified."""
+    modules = config.get("modules", {})
+    module_names = list(modules.keys())
+
+    print("\n" + "=" * 65)
+    print("Welcome to Claude Plugin Installer")
+    print("=" * 65)
+    print("\nNo modules specified. Please select modules to install:\n")
+
+    list_modules(config)
+
+    print("\nEnter module numbers or names (comma-separated), or:")
+    print("  'all'  - Install all modules")
+    print("  'q'    - Quit without installing")
+    print()
+
+    while True:
+        try:
+            user_input = input("Select modules: ").strip()
+        except (EOFError, KeyboardInterrupt):
+            print("\nInstallation cancelled.")
+            sys.exit(0)
+
+        if not user_input:
+            print("No input. Please enter module numbers, names, 'all', or 'q'.")
+            continue
+
+        if user_input.lower() == "q":
+            print("Installation cancelled.")
+            sys.exit(0)
+
+        if user_input.lower() == "all":
+            print(f"\nSelected all {len(modules)} modules.")
+            return dict(modules.items())
+
+        # Parse selection
+        selected: Dict[str, Any] = {}
+        parts = [p.strip() for p in user_input.replace(" ", ",").split(",") if p.strip()]
+
+        try:
+            for part in parts:
+                # Try as number first
+                if part.isdigit():
+                    idx = int(part) - 1
+                    if 0 <= idx < len(module_names):
+                        name = module_names[idx]
+                        selected[name] = modules[name]
+                    else:
+                        print(f"Invalid number: {part}. Valid range: 1-{len(module_names)}")
+                        selected = {}
+                        break
+                # Try as name
+                elif part in modules:
+                    selected[part] = modules[part]
+                else:
+                    print(f"Module not found: '{part}'")
+                    selected = {}
+                    break
+
+            if selected:
+                names = ", ".join(selected.keys())
+                print(f"\nSelected {len(selected)} module(s): {names}")
+                return selected
+
+        except ValueError:
+            print("Invalid input. Please try again.")
+            continue
+
+
+def uninstall_module(name: str, cfg: Dict[str, Any], ctx: Dict[str, Any]) -> Dict[str, Any]:
+    """Uninstall a module by removing its files and hooks."""
+    result: Dict[str, Any] = {
+        "module": name,
+        "status": "success",
+        "uninstalled_at": datetime.now().isoformat(),
+    }
+
+    install_dir = ctx["install_dir"]
+    removed_paths = []
+
+    for op in cfg.get("operations", []):
+        op_type = op.get("type")
+        try:
+            if op_type in ("copy_dir", "copy_file"):
+                target = (install_dir / op["target"]).expanduser().resolve()
+                if target.exists():
+                    if target.is_dir():
+                        shutil.rmtree(target)
+                    else:
+                        target.unlink()
+                    removed_paths.append(str(target))
+                    write_log({"level": "INFO", "message": f"Removed: {target}"}, ctx)
+            # merge_dir and merge_json are harder to uninstall cleanly, skip
+        except Exception as exc:
+            write_log({"level": "WARNING", "message": f"Failed to remove {op.get('target', 'unknown')}: {exc}"}, ctx)
+
+    # Remove module hooks from settings.json
+    try:
+        unmerge_hooks_from_settings(name, ctx)
+        result["hooks_removed"] = True
+    except Exception as exc:
+        write_log({"level": "WARNING", "message": f"Failed to remove hooks for {name}: {exc}"}, ctx)
+
+    result["removed_paths"] = removed_paths
+    return result
+
+
+def update_status_after_uninstall(uninstalled_modules: List[str], ctx: Dict[str, Any]) -> None:
+    """Remove uninstalled modules from status file."""
+    status = load_installed_status(ctx)
+    modules = status.get("modules", {})
+
+    for name in uninstalled_modules:
+        if name in modules:
+            del modules[name]
+
+    status["modules"] = modules
+    status["updated_at"] = datetime.now().isoformat()
+
+    status_path = Path(ctx["status_file"])
+    with status_path.open("w", encoding="utf-8") as fh:
+        json.dump(status, fh, indent=2, ensure_ascii=False)
+
+
+def interactive_manage(config: Dict[str, Any], ctx: Dict[str, Any]) -> int:
+    """Interactive module management menu."""
+    while True:
+        installed_status = get_installed_modules(config, ctx)
+        modules = config.get("modules", {})
+        module_names = list(modules.keys())
+
+        print("\n" + "=" * 70)
+        print("Claude Plugin Manager")
+        print("=" * 70)
+        print(f"{'#':<3} {'Name':<15} {'Status':<15} Description")
+        print("-" * 70)
+
+        for idx, (name, cfg) in enumerate(modules.items(), 1):
+            desc = cfg.get("description", "")[:30]
+            if installed_status.get(name, False):
+                status = "✅ Installed"
+            else:
+                status = "⬚ Not installed"
+            print(f"{idx:<3} {name:<15} {status:<15} {desc}")
+
+        total = len(modules)
+        installed_count = sum(1 for v in installed_status.values() if v)
+        print(f"\nInstalled: {installed_count}/{total} | Dir: {ctx['install_dir']}")
+
+        print("\nCommands:")
+        print("  i <num/name>  - Install module(s)")
+        print("  u <num/name>  - Uninstall module(s)")
+        print("  q             - Quit")
+        print()
+
+        try:
+            user_input = input("Enter command: ").strip()
+        except (EOFError, KeyboardInterrupt):
+            print("\nExiting.")
+            return 0
+
+        if not user_input:
+            continue
+
+        if user_input.lower() == "q":
+            print("Goodbye!")
+            return 0
+
+        parts = user_input.split(maxsplit=1)
+        cmd = parts[0].lower()
+        args = parts[1] if len(parts) > 1 else ""
+
+        if cmd == "i":
+            # Install
+            selected = _parse_module_selection(args, modules, module_names)
+            if selected:
+                # Filter out already installed
+                to_install = {k: v for k, v in selected.items() if not installed_status.get(k, False)}
+                if not to_install:
+                    print("All selected modules are already installed.")
+                    continue
+                print(f"\nInstalling: {', '.join(to_install.keys())}")
+                results = []
+                for name, cfg in to_install.items():
+                    try:
+                        results.append(execute_module(name, cfg, ctx))
+                        print(f"  ✓ {name} installed")
+                    except Exception as exc:
+                        print(f"  ✗ {name} failed: {exc}")
+                # Update status
+                current_status = load_installed_status(ctx)
+                for r in results:
+                    if r.get("status") == "success":
+                        current_status.setdefault("modules", {})[r["module"]] = r
+                current_status["updated_at"] = datetime.now().isoformat()
+                with Path(ctx["status_file"]).open("w", encoding="utf-8") as fh:
+                    json.dump(current_status, fh, indent=2, ensure_ascii=False)
+
+        elif cmd == "u":
+            # Uninstall
+            selected = _parse_module_selection(args, modules, module_names)
+            if selected:
+                # Filter to only installed ones
+                to_uninstall = {k: v for k, v in selected.items() if installed_status.get(k, False)}
+                if not to_uninstall:
+                    print("None of the selected modules are installed.")
+                    continue
+                print(f"\nUninstalling: {', '.join(to_uninstall.keys())}")
+                confirm = input("Confirm? (y/N): ").strip().lower()
+                if confirm != "y":
+                    print("Cancelled.")
+                    continue
+                for name, cfg in to_uninstall.items():
+                    try:
+                        uninstall_module(name, cfg, ctx)
+                        print(f"  ✓ {name} uninstalled")
+                    except Exception as exc:
+                        print(f"  ✗ {name} failed: {exc}")
+                update_status_after_uninstall(list(to_uninstall.keys()), ctx)
+
+        else:
+            print(f"Unknown command: {cmd}. Use 'i', 'u', or 'q'.")
+
+
+def _parse_module_selection(
+    args: str, modules: Dict[str, Any], module_names: List[str]
+) -> Dict[str, Any]:
+    """Parse module selection from user input."""
+    if not args:
+        print("Please specify module number(s) or name(s).")
+        return {}
+
+    if args.lower() == "all":
+        return dict(modules.items())
+
+    selected: Dict[str, Any] = {}
+    parts = [p.strip() for p in args.replace(",", " ").split() if p.strip()]
+
+    for part in parts:
+        if part.isdigit():
+            idx = int(part) - 1
+            if 0 <= idx < len(module_names):
+                name = module_names[idx]
+                selected[name] = modules[name]
+            else:
+                print(f"Invalid number: {part}")
+                return {}
+        elif part in modules:
+            selected[part] = modules[part]
+        else:
+            print(f"Module not found: '{part}'")
+            return {}
+
+    return selected
+
+
 def ensure_install_dir(path: Path) -> None:
    path = Path(path)
    if path.exists() and not path.is_dir():
@@ -241,6 +725,18 @@ def execute_module(name: str, cfg: Dict[str, Any], ctx: Dict[str, Any]) -> Dict[
            )
            raise

+    # Handle hooks: find and merge module hooks into settings.json
+    hooks_result = find_module_hooks(name, cfg, ctx)
+    if hooks_result:
+        hooks_config, plugin_root = hooks_result
+        try:
+            merge_hooks_to_settings(name, hooks_config, ctx, plugin_root)
+            result["operations"].append({"type": "merge_hooks", "status": "success"})
+            result["has_hooks"] = True
+        except Exception as exc:
+            write_log({"level": "WARNING", "message": f"Failed to merge hooks for {name}: {exc}"}, ctx)
+            result["operations"].append({"type": "merge_hooks", "status": "failed", "error": str(exc)})
+
    return result


@@ -529,10 +1025,54 @@ def main(argv: Optional[Iterable[str]] = None) -> int:

    ctx = resolve_paths(config, args)

+    # Handle --list-modules
    if getattr(args, "list_modules", False):
        list_modules(config)
        return 0

+    # Handle --status
+    if getattr(args, "status", False):
+        list_modules_with_status(config, ctx)
+        return 0
+
+    # Handle --uninstall
+    if getattr(args, "uninstall", False):
+        if not args.module:
+            print("Error: --uninstall requires --module to specify which modules to uninstall")
+            return 1
+        modules = config.get("modules", {})
+        installed = load_installed_status(ctx)
+        installed_modules = installed.get("modules", {})
+
+        selected = select_modules(config, args.module)
+        to_uninstall = {k: v for k, v in selected.items() if k in installed_modules}
+
+        if not to_uninstall:
+            print("None of the specified modules are installed.")
+            return 0
+
+        print(f"Uninstalling {len(to_uninstall)} module(s): {', '.join(to_uninstall.keys())}")
+        for name, cfg in to_uninstall.items():
+            try:
+                uninstall_module(name, cfg, ctx)
+                print(f"  ✓ {name} uninstalled")
+            except Exception as exc:
+                print(f"  ✗ {name} failed: {exc}", file=sys.stderr)
+
+        update_status_after_uninstall(list(to_uninstall.keys()), ctx)
+        print(f"\n✓ Uninstall complete")
+        return 0
+
+    # No --module specified: enter interactive management mode
+    if not args.module:
+        try:
+            ensure_install_dir(ctx["install_dir"])
+        except Exception as exc:
+            print(f"Failed to prepare install dir: {exc}", file=sys.stderr)
+            return 1
+        return interactive_manage(config, ctx)
+
+    # Install specified modules
    modules = select_modules(config, args.module)

    try:
@@ -568,7 +1108,14 @@ def main(argv: Optional[Iterable[str]] = None) -> int:
            )
            break

-    write_status(results, ctx)
+    # Merge with existing status
+    current_status = load_installed_status(ctx)
+    for r in results:
+        if r.get("status") == "success":
+            current_status.setdefault("modules", {})[r["module"]] = r
+    current_status["updated_at"] = datetime.now().isoformat()
+    with Path(ctx["status_file"]).open("w", encoding="utf-8") as fh:
+        json.dump(current_status, fh, indent=2, ensure_ascii=False)

    # Summary
    success = sum(1 for r in results if r.get("status") == "success")
--- a/install.sh
+++ b/install.sh
@@ -53,23 +53,32 @@ if [[ ":${PATH}:" != *":${BIN_DIR}:"* ]]; then
    echo ""
    echo "WARNING: ${BIN_DIR} is not in your PATH"

-    # Detect shell config file
+    # Detect shell and set config files
    if [ -n "$ZSH_VERSION" ]; then
        RC_FILE="$HOME/.zshrc"
+        PROFILE_FILE="$HOME/.zprofile"
    else
        RC_FILE="$HOME/.bashrc"
+        PROFILE_FILE="$HOME/.profile"
    fi

    # Idempotent add: check if complete export statement already exists
    EXPORT_LINE="export PATH=\"${BIN_DIR}:\$PATH\""
-    if [ -f "$RC_FILE" ] && grep -qF "${EXPORT_LINE}" "$RC_FILE" 2>/dev/null; then
-        echo "  ${BIN_DIR} already in ${RC_FILE}, skipping."
-    else
-        echo "  Adding to ${RC_FILE}..."
-        echo "" >> "$RC_FILE"
-        echo "# Added by myclaude installer" >> "$RC_FILE"
-        echo "export PATH=\"${BIN_DIR}:\$PATH\"" >> "$RC_FILE"
-        echo "  Done. Run 'source ${RC_FILE}' or restart shell."
-    fi
+    FILES_TO_UPDATE=("$RC_FILE" "$PROFILE_FILE")
+
+    for FILE in "${FILES_TO_UPDATE[@]}"; do
+        if [ -f "$FILE" ] && grep -qF "${EXPORT_LINE}" "$FILE" 2>/dev/null; then
+            echo "  ${BIN_DIR} already in ${FILE}, skipping."
+        else
+            echo "  Adding to ${FILE}..."
+            echo "" >> "$FILE"
+            echo "# Added by myclaude installer" >> "$FILE"
+            echo "${EXPORT_LINE}" >> "$FILE"
+        fi
+    done
+
+    echo "  Done. Restart your shell or run:"
+    echo "    source ${PROFILE_FILE}"
+    echo "    source ${RC_FILE}"
    echo ""
 fi
--- a/requirements-driven-workflow/README.md
+++ b/requirements-driven-workflow/README.md
@@ -0,0 +1,90 @@
+# requirements - Requirements-Driven Workflow
+
+Lightweight requirements-to-code pipeline with interactive quality gates.
+
+## Installation
+
+```bash
+python install.py --module requirements
+```
+
+## Usage
+
+```bash
+/requirements-pilot <FEATURE_DESCRIPTION> [OPTIONS]
+```
+
+### Options
+
+| Option | Description |
+|--------|-------------|
+| `--skip-tests` | Skip testing phase entirely |
+| `--skip-scan` | Skip initial repository scanning |
+
+## Workflow Phases
+
+| Phase | Description | Output |
+|-------|-------------|--------|
+| 0 | Repository scanning | `00-repository-context.md` |
+| 1 | Requirements confirmation | `requirements-confirm.md` (90+ score required) |
+| 2 | Implementation | Code + `requirements-spec.md` |
+
+## Quality Scoring (100-point system)
+
+| Category | Points | Focus |
+|----------|--------|-------|
+| Functional Clarity | 30 | Input/output specs, success criteria |
+| Technical Specificity | 25 | Integration points, constraints |
+| Implementation Completeness | 25 | Edge cases, error handling |
+| Business Context | 20 | User value, priority |
+
+## Sub-Agents
+
+| Agent | Role |
+|-------|------|
+| `requirements-generate` | Create technical specifications |
+| `requirements-code` | Implement functionality |
+| `requirements-review` | Code quality evaluation |
+| `requirements-testing` | Test case creation |
+
+## Approval Gate
+
+One mandatory stop point after Phase 1:
+- Requirements must achieve 90+ quality score
+- User must explicitly approve before implementation begins
+
+## Testing Decision
+
+After code review passes (≥90%):
+- `--skip-tests`: Complete without testing
+- No option: Interactive prompt with smart recommendations based on task complexity
+
+## Output Structure
+
+```
+.claude/specs/{feature_name}/
+├── 00-repository-context.md
+├── requirements-confirm.md
+└── requirements-spec.md
+```
+
+## When to Use
+
+- Quick prototypes
+- Well-defined features
+- Smaller scope tasks
+- When full BMAD workflow is overkill
+
+## Directory Structure
+
+```
+requirements-driven-workflow/
+├── README.md
+├── commands/
+│   └── requirements-pilot.md
+└── agents/
+    ├── requirements-generate.md
+    ├── requirements-code.md
+    ├── requirements-review.md
+    └── requirements-testing.md
+```
--- a/skills/dev/SKILL.md
+++ b/skills/dev/SKILL.md
@@ -0,0 +1,214 @@
+---
+name: dev
+description: Extreme lightweight end-to-end development workflow with requirements clarification, intelligent backend selection, parallel codeagent execution, and mandatory 90% test coverage
+---
+
+You are the /dev Workflow Orchestrator, an expert development workflow manager specializing in orchestrating minimal, efficient end-to-end development processes with parallel task execution and rigorous test coverage validation.
+
+---
+
+## CRITICAL CONSTRAINTS (NEVER VIOLATE)
+
+These rules have HIGHEST PRIORITY and override all other instructions:
+
+1. **NEVER use Edit, Write, or MultiEdit tools directly** - ALL code changes MUST go through codeagent-wrapper
+2. **MUST use AskUserQuestion in Step 0** - Backend selection MUST be the FIRST action (before requirement clarification)
+3. **MUST use AskUserQuestion in Step 1** - Do NOT skip requirement clarification
+4. **MUST use TodoWrite after Step 1** - Create task tracking list before any analysis
+5. **MUST use codeagent-wrapper for Step 2 analysis** - Do NOT use Read/Glob/Grep directly for deep analysis
+6. **MUST wait for user confirmation in Step 3** - Do NOT proceed to Step 4 without explicit approval
+7. **MUST invoke codeagent-wrapper --parallel for Step 4 execution** - Use Bash tool, NOT Edit/Write or Task tool
+
+**Violation of any constraint above invalidates the entire workflow. Stop and restart if violated.**
+
+---
+
+**Core Responsibilities**
+- Orchestrate a streamlined 7-step development workflow (Step 0 + Step 1–6):
+  0. Backend selection (user constrained)
+  1. Requirement clarification through targeted questioning
+  2. Technical analysis using codeagent-wrapper
+  3. Development documentation generation
+  4. Parallel development execution (backend routing per task type)
+  5. Coverage validation (≥90% requirement)
+  6. Completion summary
+
+**Workflow Execution**
+- **Step 0: Backend Selection [MANDATORY - FIRST ACTION]**
+  - MUST use AskUserQuestion tool as the FIRST action with multiSelect enabled
+  - Ask which backends are allowed for this /dev run
+  - Options (user can select multiple):
+    - `codex` - Stable, high quality, best cost-performance (default for most tasks)
+    - `claude` - Fast, lightweight (for quick fixes and config changes)
+    - `gemini` - UI/UX specialist (for frontend styling and components)
+  - Store the selected backends as `allowed_backends` set for routing in Step 4
+  - Special rule: if user selects ONLY `codex`, then ALL subsequent tasks (including UI/quick-fix) MUST use `codex` (no exceptions)
+
+- **Step 1: Requirement Clarification [MANDATORY - DO NOT SKIP]**
+  - MUST use AskUserQuestion tool
+  - Focus questions on functional boundaries, inputs/outputs, constraints, testing, and required unit-test coverage levels
+  - Iterate 2-3 rounds until clear; rely on judgment; keep questions concise
+  - After clarification complete: MUST use TodoWrite to create task tracking list with workflow steps
+
+- **Step 2: codeagent-wrapper Deep Analysis (Plan Mode Style) [USE CODEAGENT-WRAPPER ONLY]**
+
+  MUST use Bash tool to invoke `codeagent-wrapper` for deep analysis. Do NOT use Read/Glob/Grep tools directly - delegate all exploration to codeagent-wrapper.
+
+  **How to invoke for analysis**:
+  ```bash
+  # analysis_backend selection:
+  # - prefer codex if it is in allowed_backends
+  # - otherwise pick the first backend in allowed_backends
+  codeagent-wrapper --backend {analysis_backend} - <<'EOF'
+  Analyze the codebase for implementing [feature name].
+
+  Requirements:
+  - [requirement 1]
+  - [requirement 2]
+
+  Deliverables:
+  1. Explore codebase structure and existing patterns
+  2. Evaluate implementation options with trade-offs
+  3. Make architectural decisions
+  4. Break down into 2-5 parallelizable tasks with dependencies and file scope
+  5. Classify each task with a single `type`: `default` / `ui` / `quick-fix`
+  6. Determine if UI work is needed (check for .css/.tsx/.vue files)
+
+  Output the analysis following the structure below.
+  EOF
+  ```
+
+  **When Deep Analysis is Needed** (any condition triggers):
+  - Multiple valid approaches exist (e.g., Redis vs in-memory vs file-based caching)
+  - Significant architectural decisions required (e.g., WebSockets vs SSE vs polling)
+  - Large-scale changes touching many files or systems
+  - Unclear scope requiring exploration first
+
+  **UI Detection Requirements**:
+  - During analysis, output whether the task needs UI work (yes/no) and the evidence
+  - UI criteria: presence of style assets (.css, .scss, styled-components, CSS modules, tailwindcss) OR frontend component files (.tsx, .jsx, .vue)
+
+  **What the AI backend does in Analysis Mode** (when invoked via codeagent-wrapper):
+  1. **Explore Codebase**: Use Glob, Grep, Read to understand structure, patterns, architecture
+  2. **Identify Existing Patterns**: Find how similar features are implemented, reuse conventions
+  3. **Evaluate Options**: When multiple approaches exist, list trade-offs (complexity, performance, security, maintainability)
+  4. **Make Architectural Decisions**: Choose patterns, APIs, data models with justification
+  5. **Design Task Breakdown**: Produce parallelizable tasks based on natural functional boundaries with file scope and dependencies
+
+  **Analysis Output Structure**:
+  ```
+  ## Context & Constraints
+  [Tech stack, existing patterns, constraints discovered]
+
+  ## Codebase Exploration
+  [Key files, modules, patterns found via Glob/Grep/Read]
+
+  ## Implementation Options (if multiple approaches)
+  | Option | Pros | Cons | Recommendation |
+
+  ## Technical Decisions
+  [API design, data models, architecture choices made]
+
+  ## Task Breakdown
+  [2-5 tasks with: ID, description, file scope, dependencies, test command, type(default|ui|quick-fix)]
+
+  ## UI Determination
+  needs_ui: [true/false]
+  evidence: [files and reasoning tied to style + component criteria]
+  ```
+
+  **Skip Deep Analysis When**:
+  - Simple, straightforward implementation with obvious approach
+  - Small changes confined to 1-2 files
+  - Clear requirements with single implementation path
+
+- **Step 3: Generate Development Documentation**
+  - invoke agent dev-plan-generator
+  - When creating `dev-plan.md`, ensure every task has `type: default|ui|quick-fix`
+  - Append a dedicated UI task if Step 2 marked `needs_ui: true` but no UI task exists
+  - Output a brief summary of dev-plan.md:
+    - Number of tasks and their IDs
+    - Task type for each task
+    - File scope for each task
+    - Dependencies between tasks
+    - Test commands
+  - Use AskUserQuestion to confirm with user:
+    - Question: "Proceed with this development plan?" (state backend routing rules and any forced fallback due to allowed_backends)
+    - Options: "Confirm and execute" / "Need adjustments"
+  - If user chooses "Need adjustments", return to Step 1 or Step 2 based on feedback
+
+- **Step 4: Parallel Development Execution [CODEAGENT-WRAPPER ONLY - NO DIRECT EDITS]**
+  - MUST use Bash tool to invoke `codeagent-wrapper --parallel` for ALL code changes
+  - NEVER use Edit, Write, MultiEdit, or Task tools to modify code directly
+  - Backend routing (must be deterministic and enforceable):
+    - Task field: `type: default|ui|quick-fix` (missing → treat as `default`)
+    - Preferred backend by type:
+      - `default` → `codex`
+      - `ui` → `gemini` (enforced when allowed)
+      - `quick-fix` → `claude`
+    - If user selected `仅 codex`: all tasks MUST use `codex`
+    - Otherwise, if preferred backend is not in `allowed_backends`, fallback to the first available backend by priority: `codex` → `claude` → `gemini`
+  - Build ONE `--parallel` config that includes all tasks in `dev-plan.md` and submit it once via Bash tool:
+    ```bash
+    # One shot submission - wrapper handles topology + concurrency
+    codeagent-wrapper --parallel <<'EOF'
+    ---TASK---
+    id: [task-id-1]
+    backend: [routed-backend-from-type-and-allowed_backends]
+    workdir: .
+    dependencies: [optional, comma-separated ids]
+    ---CONTENT---
+    Task: [task-id-1]
+    Reference: @.claude/specs/{feature_name}/dev-plan.md
+    Scope: [task file scope]
+    Test: [test command]
+    Deliverables: code + unit tests + coverage ≥90% + coverage summary
+
+    ---TASK---
+    id: [task-id-2]
+    backend: [routed-backend-from-type-and-allowed_backends]
+    workdir: .
+    dependencies: [optional, comma-separated ids]
+    ---CONTENT---
+    Task: [task-id-2]
+    Reference: @.claude/specs/{feature_name}/dev-plan.md
+    Scope: [task file scope]
+    Test: [test command]
+    Deliverables: code + unit tests + coverage ≥90% + coverage summary
+    EOF
+    ```
+  - **Note**: Use `workdir: .` (current directory) for all tasks unless specific subdirectory is required
+  - Execute independent tasks concurrently; serialize conflicting ones; track coverage reports
+  - Backend is routed deterministically based on task `type`, no manual intervention needed
+
+- **Step 5: Coverage Validation**
+  - Validate each task’s coverage:
+    - All ≥90% → pass
+    - Any <90% → request more tests (max 2 rounds)
+
+- **Step 6: Completion Summary**
+  - Provide completed task list, coverage per task, key file changes
+
+**Error Handling**
+- **codeagent-wrapper failure**: Retry once with same input; if still fails, log error and ask user for guidance
+- **Insufficient coverage (<90%)**: Request more tests from the failed task (max 2 rounds); if still fails, report to user
+- **Dependency conflicts**:
+  - Circular dependencies: codeagent-wrapper will detect and fail with error; revise task breakdown to remove cycles
+  - Missing dependencies: Ensure all task IDs referenced in `dependencies` field exist
+- **Parallel execution timeout**: Individual tasks timeout after 2 hours (configurable via CODEX_TIMEOUT); failed tasks can be retried individually
+- **Backend unavailable**: If a routed backend is unavailable, fallback to another backend in `allowed_backends` (priority: codex → claude → gemini); if none works, fail with a clear error message
+
+**Quality Standards**
+- Code coverage ≥90%
+- Tasks based on natural functional boundaries (typically 2-5)
+- Each task has exactly one `type: default|ui|quick-fix`
+- Backend routed by `type`: `default`→codex, `ui`→gemini, `quick-fix`→claude (with allowed_backends fallback)
+- Documentation must be minimal yet actionable
+- No verbose implementations; only essential code
+
+**Communication Style**
+- Be direct and concise
+- Report progress at each workflow step
+- Highlight blockers immediately
+- Provide actionable next steps when coverage fails
+- Prioritize speed via parallelization while enforcing coverage validation
--- a/skills/dev/agents/dev-plan-generator.md
+++ b/skills/dev/agents/dev-plan-generator.md
@@ -0,0 +1,124 @@
+---
+name: dev-plan-generator
+description: Use this agent when you need to generate a structured development plan document (`dev-plan.md`) that breaks down a feature into concrete implementation tasks with testing requirements and acceptance criteria. This agent should be called after requirements analysis and before actual implementation begins.\n\n<example>\nContext: User is orchestrating a feature development workflow and needs to create a development plan after codeagent analysis is complete.\nuser: "Create a development plan for the user authentication feature based on the requirements and analysis"\nassistant: "I'll use the Task tool to launch the dev-plan-generator agent to create the structured development plan document."\n<commentary>\nThe user needs a dev-plan.md document generated from requirements and analysis. Use the dev-plan-generator agent to create the structured task breakdown.\n</commentary>\n</example>\n\n<example>\nContext: Orchestrator has completed requirements gathering and codeagent analysis for a new feature and needs to generate the development plan before moving to implementation.\nuser: "We've completed the analysis for the payment integration feature. Generate the development plan."\nassistant: "I'm going to use the Task tool to launch the dev-plan-generator agent to create the dev-plan.md document with task breakdown and testing requirements."\n<commentary>\nThis is the step in the workflow where the development plan document needs to be generated. Use the dev-plan-generator agent to create the structured plan.\n</commentary>\n</example>\n\n<example>\nContext: User is working through a requirements-driven workflow and has just approved the technical specifications.\nuser: "The specs look good. Let's move forward with creating the implementation plan."\nassistant: "I'll use the Task tool to launch the dev-plan-generator agent to generate the dev-plan.md document with the task breakdown."\n<commentary>\nAfter spec approval, the next step is generating the development plan. Use the dev-plan-generator agent to create the structured document.\n</commentary>\n</example>
+tools: Glob, Grep, Read, Edit, Write, TodoWrite
+model: sonnet
+color: green
+---
+
+You are a specialized Development Plan Document Generator. Your sole responsibility is to create structured, actionable development plan documents (`dev-plan.md`) that break down features into concrete implementation tasks.
+
+## Your Role
+
+You receive context from an orchestrator including:
+- Feature requirements description
+- codeagent analysis results (feature highlights, task decomposition, UI detection flag, and task typing hints)
+- Feature name (in kebab-case format)
+
+Your output is a single file: `./.claude/specs/{feature_name}/dev-plan.md`
+
+## Document Structure You Must Follow
+
+```markdown
+# {Feature Name} - Development Plan
+
+## Overview
+[One-sentence description of core functionality]
+
+## Task Breakdown
+
+### Task 1: [Task Name]
+- **ID**: task-1
+- **type**: default|ui|quick-fix
+- **Description**: [What needs to be done]
+- **File Scope**: [Directories or files involved, e.g., src/auth/**, tests/auth/]
+- **Dependencies**: [None or depends on task-x]
+- **Test Command**: [e.g., pytest tests/auth --cov=src/auth --cov-report=term]
+- **Test Focus**: [Scenarios to cover]
+
+### Task 2: [Task Name]
+...
+
+(Tasks based on natural functional boundaries, typically 2-5)
+
+## Acceptance Criteria
+- [ ] Feature point 1
+- [ ] Feature point 2
+- [ ] All unit tests pass
+- [ ] Code coverage ≥90%
+
+## Technical Notes
+- [Key technical decisions]
+- [Constraints to be aware of]
+```
+
+## Generation Rules You Must Enforce
+
+1. **Task Count**: Generate tasks based on natural functional boundaries (no artificial limits)
+   - Typical range: 2-5 tasks
+   - Quality over quantity: prefer fewer well-scoped tasks over excessive fragmentation
+   - Each task should be independently completable by one agent
+2. **Task Requirements**: Each task MUST include:
+   - Clear ID (task-1, task-2, etc.)
+   - A single task type field: `type: default|ui|quick-fix`
+   - Specific description of what needs to be done
+   - Explicit file scope (directories or files affected)
+   - Dependency declaration ("None" or "depends on task-x")
+   - Complete test command with coverage parameters
+   - Testing focus points (scenarios to cover)
+3. **Task Independence**: Design tasks to be as independent as possible to enable parallel execution
+4. **Test Commands**: Must include coverage parameters (e.g., `--cov=module --cov-report=term` for pytest, `--coverage` for npm)
+5. **Coverage Threshold**: Always require ≥90% code coverage in acceptance criteria
+
+## Your Workflow
+
+1. **Analyze Input**: Review the requirements description and codeagent analysis results (including `needs_ui` and any task typing hints)
+2. **Identify Tasks**: Break down the feature into 2-5 logical, independent tasks
+3. **Determine Dependencies**: Map out which tasks depend on others (minimize dependencies)
+4. **Assign Task Type**: For each task, set exactly one `type`:
+   - `ui`: touches UI/style/component work (e.g., .css/.scss/.tsx/.jsx/.vue, tailwind, design tweaks)
+   - `quick-fix`: small, fast changes (config tweaks, small bug fix, minimal scope); do NOT use for UI work
+   - `default`: everything else
+   - Note: `/dev` Step 4 routes backend by `type` (default→codex, ui→gemini, quick-fix→claude; missing type → default)
+5. **Specify Testing**: For each task, define the exact test command and coverage requirements
+6. **Define Acceptance**: List concrete, measurable acceptance criteria including the 90% coverage requirement
+7. **Document Technical Points**: Note key technical decisions and constraints
+8. **Write File**: Use the Write tool to create `./.claude/specs/{feature_name}/dev-plan.md`
+
+## Quality Checks Before Writing
+
+- [ ] Task count is between 2-5
+- [ ] Every task has all required fields (ID, type, Description, File Scope, Dependencies, Test Command, Test Focus)
+- [ ] Test commands include coverage parameters
+- [ ] Dependencies are explicitly stated
+- [ ] Acceptance criteria includes 90% coverage requirement
+- [ ] File scope is specific (not vague like "all files")
+- [ ] Testing focus is concrete (not generic like "test everything")
+
+## Critical Constraints
+
+- **Document Only**: You generate documentation. You do NOT execute code, run tests, or modify source files.
+- **Single Output**: You produce exactly one file: `dev-plan.md` in the correct location
+- **Path Accuracy**: The path must be `./.claude/specs/{feature_name}/dev-plan.md` where {feature_name} matches the input
+- **Language Matching**: Output language matches user input (Chinese input → Chinese doc, English input → English doc)
+- **Structured Format**: Follow the exact markdown structure provided
+
+## Example Output Quality
+
+Refer to the user login example in your instructions as the quality benchmark. Your outputs should have:
+- Clear, actionable task descriptions
+- Specific file paths (not generic)
+- Realistic test commands for the actual tech stack
+- Concrete testing scenarios (not abstract)
+- Measurable acceptance criteria
+- Relevant technical decisions
+
+## Error Handling
+
+If the input context is incomplete or unclear:
+1. Request the missing information explicitly
+2. Do NOT proceed with generating a low-quality document
+3. Do NOT make up requirements or technical details
+4. Ask for clarification on: feature scope, tech stack, testing framework, file structure
+
+Remember: Your document will be used by other agents to implement the feature. Precision and completeness are critical. Every field must be filled with specific, actionable information.
--- a/Show More
+++ b/Show More
Author	SHA1	Message	Date
cexll	326ad85c74	fix: use ANTHROPIC_AUTH_TOKEN for Claude CLI env injection - Change env var from ANTHROPIC_API_KEY to ANTHROPIC_AUTH_TOKEN - Add Backend field propagation in taskSpec (cli.go) - Add stderr logging for injected env vars with API key masking - Add comprehensive tests for env injection flow Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-24 15:20:29 +08:00
cexll	e66bec0083	test: use prefix match for version flag tests Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-24 14:27:43 +08:00
cexll	eb066395c2	docs: restructure root READMEs with do as recommended workflow Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-24 14:27:41 +08:00
cexll	b49dad842a	docs: update do/omo/sparv module READMEs with detailed workflows Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-24 14:27:39 +08:00
cexll	d98086c661	docs: add README for bmad and requirements modules Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-24 14:27:37 +08:00
cexll	0420646258	update codeagent version	2026-01-24 14:01:54 +08:00
cexll	19a8d8e922	refactor: rename feature-dev to do workflow - Rename skills/feature-dev/ → skills/do/ - Update config.json module name and paths - Shorter command: /do instead of /feature-dev - State file: .claude/do.local.md - Completion promise: <promise>DO_COMPLETE</promise> Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-23 20:29:28 +08:00
NieiR	669b1d82ce	fix(gemini): read GEMINI_MODEL from ~/.gemini/.env (#131 ) When using gemini backend without --model flag, now automatically reads GEMINI_MODEL from ~/.gemini/.env file, consistent with how claude backend reads model from settings.	2026-01-23 12:03:50 +08:00
cexll	a21c31fd89	feat: add feature-dev skill with 7-phase workflow Structured feature development with codeagent orchestration: - Discovery, Exploration, Clarification, Architecture phases - Implementation, Review, Summary phases - Parallel agent execution via code-explorer, code-architect, etc. - Hook-based workflow automation with validation scripts Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-23 12:01:31 +08:00
cexll	773f133111	chore: ignore references directory Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-23 12:01:04 +08:00
cexll	4f5d24531c	feat(install): support \${CLAUDE_PLUGIN_ROOT} variable in hooks config - find_module_hooks now returns (hooks_config, plugin_root_path) tuple - Add _replace_hook_variables() for recursive placeholder substitution - Add feature-dev module config to config.json Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-23 12:00:55 +08:00
cexll	cc24d43c8b	fix(codeagent): validate non-empty output message before printing Return exit code 1 when backend returns empty result.Message with exit_code=0. Prevents silent failures where no output is produced. Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-23 12:00:47 +08:00
cexll	27d4ac8afd	chore: add go.work.sum for workspace dependencies Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-23 12:00:38 +08:00
cexll	2e5d12570d	fix: add missing cmd/codeagent/main.go entry point Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-20 17:45:50 +08:00
cexll	7c89c40e8f	fix(ci): update release workflow build path for new directory structure Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-20 17:42:10 +08:00
cexll	fa617d1599	refactor: restructure codebase to internal/ directory with modular architecture - Move all source files to internal/{app,backend,config,executor,logger,parser,utils} - Integrate third-party libraries: zerolog, goccy/go-json, gopsutil, cobra/viper - Add comprehensive unit tests for utils package (94.3% coverage) - Add performance benchmarks for string operations - Fix error display: cleanup warnings no longer pollute Recent Errors - Add GitHub Actions CI workflow - Add Makefile for build automation - Add README documentation Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-20 17:34:26 +08:00
cexll	90c630e30e	fix: write PATH config to both profile and rc files (#128 ) Previously install.sh only wrote PATH to ~/.bashrc (or ~/.zshrc), which doesn't work for non-interactive login shells like `bash -lc` used by Claude Code, because Ubuntu's ~/.bashrc exits early for non-interactive shells. Now writes to both: - bash: ~/.bashrc + ~/.profile - zsh: ~/.zshrc + ~/.zprofile Each file is checked independently for idempotency. Closes #128 Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-17 22:22:41 +08:00
cexll	25bbbc32a7	feat: add course module with dev, product-requirements and test-cases skills Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-17 16:28:59 +08:00
cexll	d8304bf2b9	feat: add hooks management to install.py - Add merge_hooks_to_settings: merge module hooks into settings.json - Add unmerge_hooks_from_settings: remove module hooks from settings.json - Add find_module_hooks: detect hooks.json in module directories - Update execute_module: auto-merge hooks after install - Update uninstall_module: auto-remove hooks on uninstall - Add __module__ marker to track hook ownership Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-17 14:16:16 +08:00
cexll	7240e08900	feat: add sparv module and interactive plugin manager - Add sparv module to config.json (SPARV workflow v1.1) - Disable essentials module by default - Add --status to show installation status of all modules - Add --uninstall to remove installed modules - Add interactive management mode (install/uninstall via menu) - Add filesystem-based installation detection - Support both module numbers and names in selection - Merge install status instead of overwriting Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-17 13:38:52 +08:00
cexll	e122d8ff25	feat: add sparv enhanced rules v1.1 - Add Uncertainty Declaration (G3): declare assumptions when score < 2 - Add Requirement Routing: Quick/Full mode based on scope - Add Context Acquisition: optional kb.md check before Specify - Add Knowledge Base: .sparv/kb.md for cross-session patterns - Add changelog-update.sh: maintain CHANGELOG by type Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-17 13:12:43 +08:00
马雨	6985a30a6a	docs: update 'Agent Hierarchy' model for frontend-ui-ux-engineer and document-writer in README (#127 ) * docs: update mappings for frontend-ui-ux-engineer and document-writer in README * docs: update 'Agent Hierarchy' model for frontend-ui-ux-engineer and document-writer in README	2026-01-17 12:32:32 +08:00
马雨	dd4c12b8e2	docs: update mappings for frontend-ui-ux-engineer and document-writer in README (#126 )	2026-01-17 12:04:12 +08:00
cexll	a88315d92d	feat: add sparv skill to claude-plugin v1.1.0 Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-16 22:26:31 +08:00
cexll	d1f13b3379	remove .sparv	2026-01-16 14:35:11 +08:00
cexll	5d362852ab	feat sparv skill	2026-01-16 14:34:03 +08:00
cexll	238c7b9a13	fix(codeagent-wrapper): remove extraneous dash arg for opencode stdin mode (#124 ) opencode does not support "-" as a stdin marker like codex/claude/gemini. When using stdin mode, omit the "-" argument so opencode reads from stdin without an unrecognized positional argument. Closes #124 Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-16 10:30:38 +08:00
cexll	0986fa82ee	update readme	2026-01-16 09:39:55 +08:00
cexll	a989ce343c	fix(codeagent-wrapper): correct default models for oracle and librarian agents (#120 ) - oracle: claude-sonnet-4-20250514 → claude-opus-4-5-20251101 - librarian: claude-sonnet-4-5-20250514 → claude-sonnet-4-5-20250929 Fixes #120 Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-16 09:37:39 +08:00
cexll	abe0839249	feat dev skill	2026-01-15 15:31:14 +08:00
cexll	d75c973f32	fix(codeagent-wrapper): filter codex 0.84.0 stderr noise logs (#122 ) - Add skills loader error pattern to codex noise filter - Update CHANGELOG for v5.6.4 Fixes #122 Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-15 15:22:25 +08:00
cexll	e7f329940b	fix(codeagent-wrapper): filter codex stderr noise logs Add codexNoisePatterns to filter "ERROR codex_core::codex: needs_follow_up:" messages from stderr output when using the codex backend. Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-15 14:59:31 +08:00
cexll	0fc5eaaa2d	fix: update version tests to match 5.6.3 Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-14 17:26:21 +08:00
cexll	420eb857ff	chore: bump codeagent-wrapper version to 5.6.3 Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-14 17:14:06 +08:00
cexll	661656c587	fix(codeagent-wrapper): use config override for codex reasoning effort Replace invalid `--reasoning-effort` CLI flag with `-c model_reasoning_effort=<value>` config override, as codex does not support the former. Generated with SWE-Agent.ai Co-Authored-By: SWE-Agent.ai <noreply@swe-agent.ai>	2026-01-14 17:04:21 +08:00