🚀 Claude Code Workflow (CCW)
📋 Overview
Claude Code Workflow (CCW) is a next-generation multi-agent automation framework for software development that orchestrates complex development tasks through intelligent workflow management and autonomous execution.
🎯 Latest Release v1.3: Enhanced task decomposition standards, advanced search strategies with bash command combinations, free exploration phases for agents, and comprehensive workflow system improvements. See CHANGELOG.md for details.
🌟 Key Innovations
- 🧠 Intelligent Task Decomposition: New core standards prevent over-fragmentation with functional completeness principles
- 🔍 Advanced Search Strategies: Powerful command combinations using ripgrep, grep, find, awk, sed for comprehensive analysis
- ⚡ Free Exploration Phase: Agents can gather supplementary context after structured analysis
- 🎯 JSON-First Architecture: Single source of truth with atomic session management
- 🤖 Dual CLI Integration: Gemini for analysis, Codex for autonomous development
🏗️ System Architecture
🔧 Core Architectural Principles
🏛️ Three-Pillar Foundation
| 🏗️ JSON-First Data Model |
⚡ Atomic Session Management |
🧩 Adaptive Complexity |
| Single source of truth |
Marker-based session state |
Auto-adjusts to project size |
| Sub-millisecond queries |
Zero-overhead switching |
Simple → Medium → Complex |
| Generated Markdown views |
Conflict-free concurrency |
Task limit enforcement |
| Data consistency guaranteed |
Instant context switching |
Intelligent decomposition |
✨ Major Enhancements v1.3
🎯 Core Task Decomposition Standards
Revolutionary task decomposition system with four core principles:
- 🎯 Functional Completeness Principle - Complete, runnable functional units
- 📏 Minimum Size Threshold - 3+ files or 200+ lines minimum
- 🔗 Dependency Cohesion Principle - Tightly coupled components together
- 📊 Hierarchy Control Rule - Flat ≤5, hierarchical 6-10, re-scope >10
🔍 Advanced Search Strategies
Powerful command combinations for comprehensive codebase analysis:
🚀 Free Exploration Phase
Agents can enter supplementary context gathering using bash commands (grep, find, rg, awk, sed) after completing structured pre-analysis steps.
🧠 Intelligent Gemini Wrapper
Smart automation with token management and approval modes:
- Analysis Detection: Keywords trigger
--approval-mode default
- Development Detection: Action words trigger
--approval-mode yolo
- Auto Token Management: Handles
--all-files based on project size
- Error Logging: Comprehensive error tracking and recovery
📊 Complexity Management System
CCW automatically adapts workflow structure based on project complexity:
| Complexity |
Task Count |
Structure |
Features |
| 🟢 Simple |
<5 tasks |
Single-level |
Minimal overhead, direct execution |
| 🟡 Medium |
5-10 tasks |
Two-level hierarchy |
Progress tracking, automated docs |
| 🔴 Complex |
>10 tasks |
Force re-scoping |
Multi-iteration planning required |
🛠️ Complete Command Reference
🎮 Core System Commands
| Command |
Function |
Example |
🎯 /enhance-prompt |
Technical context enhancement |
/enhance-prompt "add auth system" |
📊 /context |
Unified context management |
/context --analyze --format=tree |
📝 /update-memory-full |
Complete documentation update |
/update-memory-full |
🔄 /update-memory-related |
Smart context-aware updates |
/update-memory-related |
🔍 Gemini CLI Commands (Analysis & Investigation)
| Command |
Purpose |
Usage |
🔍 /gemini:analyze |
Deep codebase analysis |
/gemini:analyze "authentication patterns" |
💬 /gemini:chat |
Direct Gemini interaction |
/gemini:chat "explain this architecture" |
⚡ /gemini:execute |
Intelligent execution |
/gemini:execute task-001 |
🎯 /gemini:mode:auto |
Auto template selection |
/gemini:mode:auto "analyze security" |
🐛 /gemini:mode:bug-index |
Bug analysis workflow |
/gemini:mode:bug-index "payment fails" |
🤖 Codex CLI Commands (Development & Implementation)
| Command |
Purpose |
Usage |
🔍 /codex:analyze |
Development analysis |
/codex:analyze "optimization opportunities" |
💬 /codex:chat |
Direct Codex interaction |
/codex:chat "implement JWT auth" |
⚡ /codex:execute |
Controlled development |
/codex:execute "refactor user service" |
🚀 /codex:mode:auto |
PRIMARY: Full autonomous |
/codex:mode:auto "build payment system" |
🐛 /codex:mode:bug-index |
Autonomous bug fixing |
/codex:mode:bug-index "fix race condition" |
🎯 Workflow Management
📋 Session Management
| Command |
Function |
Usage |
🚀 /workflow:session:start |
Create new session |
/workflow:session:start "OAuth2 System" |
⏸️ /workflow:session:pause |
Pause current session |
/workflow:session:pause |
▶️ /workflow:session:resume |
Resume session |
/workflow:session:resume "OAuth2 System" |
📋 /workflow:session:list |
List all sessions |
/workflow:session:list --active |
🔄 /workflow:session:switch |
Switch sessions |
/workflow:session:switch "Payment Fix" |
🎯 Workflow Operations
| Command |
Function |
Usage |
💭 /workflow:brainstorm |
Multi-agent planning |
/workflow:brainstorm "microservices architecture" |
📋 /workflow:plan |
Convert to executable plans |
/workflow:plan --from-brainstorming |
🔍 /workflow:plan-deep |
Deep architectural planning |
/workflow:plan-deep "API redesign" --complexity=high |
⚡ /workflow:execute |
Implementation phase |
/workflow:execute --type=complex |
✅ /workflow:review |
Quality assurance |
/workflow:review --auto-fix |
🏷️ Task Management
| Command |
Function |
Usage |
➕ /task:create |
Create implementation task |
/task:create "User Authentication" |
🔄 /task:breakdown |
Decompose into subtasks |
/task:breakdown IMPL-1 --depth=2 |
⚡ /task:execute |
Execute specific task |
/task:execute IMPL-1.1 --mode=auto |
📋 /task:replan |
Adapt to changes |
/task:replan IMPL-1 --strategy=adjust |
🎯 Complete Development Workflows
🚀 Complex Feature Development
🔥 Quick Development Examples
🚀 Full Stack Feature Implementation
⚡ Rapid Bug Resolution
📊 Architecture Analysis & Refactoring
🏗️ Project Structure
⚡ Performance & Technical Specs
📊 Performance Metrics
| Metric |
Performance |
Details |
| 🔄 Session Switching |
<10ms |
Atomic marker file operations |
| 📊 JSON Queries |
<1ms |
Direct JSON access, no parsing overhead |
| 📝 Doc Updates |
<30s |
Medium projects, intelligent targeting |
| 🔍 Context Loading |
<5s |
Complex codebases with caching |
| ⚡ Task Execution |
10min timeout |
Complex operations with error handling |
🛠️ System Requirements
- 🖥️ OS: Windows 10+, Ubuntu 18.04+, macOS 10.15+
- 📦 Dependencies: Git, Node.js (Gemini), Python 3.8+ (Codex)
- 💾 Storage: ~50MB core + variable project data
- 🧠 Memory: 512MB minimum, 2GB recommended
🔗 Integration Requirements
- 🔍 Gemini CLI: Required for analysis workflows
- 🤖 Codex CLI: Required for autonomous development
- 📂 Git Repository: Required for change tracking
- 🎯 Claude Code IDE: Recommended for optimal experience
⚙️ Installation & Configuration
🚀 Quick Installation
✅ Verify Installation
⚙️ Essential Configuration
Gemini CLI Setup
Optimized .geminiignore
🤝 Contributing
🛠️ Development Setup
- 🍴 Fork the repository
- 🌿 Create feature branch:
git checkout -b feature/enhancement-name
- 📦 Install dependencies
- ✅ Test with sample projects
- 📤 Submit detailed pull request
📏 Code Standards
- ✅ Follow existing command patterns
- 🔄 Maintain backward compatibility
- 🧪 Add tests for new functionality
- 📚 Update documentation
- 🏷️ Use semantic versioning
📞 Support & Resources
📄 License
This project is licensed under the MIT License - see the LICENSE file for details.
🚀 Claude Code Workflow (CCW)
Professional software development workflow automation through intelligent multi-agent coordination and autonomous execution capabilities.
