mirror of
https://github.com/catlog22/Claude-Code-Workflow.git
synced 2026-02-13 02:41:50 +08:00
- Move -s parameter to end of command: codex --full-auto exec "..." -s danger-full-access - Change default sandbox mode from workspace-write to danger-full-access - Add missing --full-auto parameter to commands requiring autonomous development - Add missing exec parameter to all codex commands - Add missing -s danger-full-access parameter to multi-line commands - Update all documentation and examples to use consistent syntax - Fix CLAUDE.md detection functionality in detect_changed_modules.sh script Files updated: - Core configuration: CLAUDE.md, RELEASE_NOTES_v1.3.0.md - Documentation: tools-implementation-guide.md, intelligent-tools-strategy.md - Command files: All .claude/commands/codex/*.md and workflow/*.md files - Agent templates: action-planning-agent.md - Scripts: detect_changed_modules.sh 🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>
5.7 KiB
5.7 KiB
name, description, type
| name | description | type |
|---|---|---|
| intelligent-tools-strategy | Strategic guide for intelligent tool selection - quick start and decision framework | strategic-guideline |
Intelligent Tools Selection Strategy
⚡ Quick Start
Essential Commands
Gemini (Analysis & Pattern Recognition):
~/.claude/scripts/gemini-wrapper -p "analyze authentication patterns"
Codex (Development & Implementation):
codex --full-auto exec "implement user authentication system" -s danger-full-access
⚠️ CRITICAL Command Differences
| Tool | Command | Has Wrapper | Key Feature |
|---|---|---|---|
| Gemini | ~/.claude/scripts/gemini-wrapper |
✅ YES | Large context window, pattern recognition |
| Codex | codex --full-auto exec ... -s danger-full-access |
❌ NO | Autonomous development, math reasoning |
❌ NEVER use: ~/.claude/scripts/codex - this wrapper does not exist!
🔒 Codex Sandbox Modes
Codex requires the -s (sandbox) parameter for write operations:
-s read-only: Safe analysis mode, no file modifications-s workspace-write: Standard development mode, can modify project files-s danger-full-access: Full system access (RECOMMENDED for development tasks)
Default Choice: Use -s danger-full-access for all development and implementation tasks.
🎯 Tool Selection Matrix
When to Use Gemini
- Command:
~/.claude/scripts/gemini-wrapper -p "prompt" - Strengths: Large context window, pattern recognition across modules
- Best For:
- Project architecture analysis (>50 files)
- Cross-module pattern detection
- Coding convention analysis
- Refactoring with broad dependencies
- Large codebase understanding
When to Use Codex
- Command:
codex --full-auto exec "prompt" -s danger-full-access - Strengths: Mathematical reasoning, autonomous development
- Best For:
- Complex algorithm analysis
- Security vulnerability assessment
- Performance optimization
- Database schema design
- API protocol specifications
- Autonomous feature development
📊 Decision Framework
| Analysis Need | Recommended Tool | Rationale |
|---|---|---|
| Project Architecture | Gemini | Needs broad context across many files |
| Algorithm Optimization | Codex | Requires deep mathematical reasoning |
| Security Analysis | Codex | Leverages deeper security knowledge |
| Code Patterns | Gemini | Pattern recognition across modules |
| Refactoring | Gemini | Needs understanding of all dependencies |
| API Design | Codex | Technical specification expertise |
| Test Coverage | Gemini | Cross-module test understanding |
| Performance Tuning | Codex | Mathematical optimization capabilities |
| Feature Implementation | Codex | Autonomous development capabilities |
| Architectural Review | Gemini | Large context analysis |
🔄 Parallel Analysis Strategy
For complex projects requiring both broad context and deep analysis:
# Use Task agents to run both tools in parallel
Task(subagent_type="general-purpose",
prompt="Use Gemini (see @~/.claude/workflows/tools-implementation-guide.md) for architectural analysis")
+
Task(subagent_type="general-purpose",
prompt="Use Codex (see @~/.claude/workflows/tools-implementation-guide.md) for algorithmic analysis")
📈 Complexity-Based Selection
Simple Projects (≤50 files)
- Content-driven choice: Mathematical → Codex, Structural → Gemini
Medium Projects (50-200 files)
- Gemini first for overview and patterns
- Codex second for specific implementations
Large Projects (>200 files)
- Parallel analysis with both tools
- Gemini for architectural understanding
- Codex for focused development tasks
🎯 Quick Reference
Gemini Quick Commands
# Pattern analysis
~/.claude/scripts/gemini-wrapper -p "analyze existing patterns in auth module"
# Architecture review
cd src && ~/.claude/scripts/gemini-wrapper -p "review overall architecture"
# Code conventions
~/.claude/scripts/gemini-wrapper -p "identify coding standards and conventions"
Codex Quick Commands
# Feature development
codex --full-auto exec "implement JWT authentication with refresh tokens" -s danger-full-access
# Performance optimization
codex --full-auto exec "optimize database queries in user service" -s danger-full-access
# Security enhancement
codex --full-auto exec "add input validation and sanitization" -s danger-full-access
📋 Implementation Guidelines
- Default Selection: Let project characteristics drive tool choice
- Start Simple: Begin with single tool, escalate to parallel if needed
- Context First: Understand scope before selecting approach
- Trust the Tools: Let autonomous capabilities handle complexity
🔗 Detailed Implementation
For comprehensive syntax, patterns, and advanced usage:
- Implementation Guide: @~/.claude/workflows/tools-implementation-guide.md
📊 Tools Comparison Summary
| Feature | Gemini | Codex |
|---|---|---|
| Command Syntax | Has wrapper script | Direct command only |
| File Loading | --all-files available |
@ patterns required |
| Default Mode | Interactive analysis | --full-auto exec ... -s danger-full-access automation |
| Primary Use | Analysis & planning | Development & implementation |
| Context Window | Very large | Standard with smart discovery |
| Automation Level | Manual implementation | Autonomous execution |
| Best For | Understanding codebases | Building features |
Remember: Choose based on task nature, not personal preference. Gemini excels at understanding, Codex excels at building.