🚀 ClaudeCraft
Multi-Agent Development Framework for High-Quality, Secure Software Development with Claude AI
Transform your development workflow with AI-powered agents that ensure security, quality, and velocity at every step.
🎯 What is ClaudeCraft?
ClaudeCraft is a revolutionary multi-agent development framework that brings structure, security, and quality to AI-assisted software development. Instead of chaotic back-and-forth coding with AI, you get specialized AI agents working together in a coordinated workflow - each optimized for specific development tasks.
🏆 Key Benefits
- 🛡️ Security-First: OWASP Top 10 compliance built into every feature
- 🎯 Quality-Focused: 80%+ test coverage, code reviews, performance monitoring
- ⚡ High Velocity: Structured agent handoffs eliminate context loss
- 📚 Self-Documenting: Comprehensive memory and decision tracking
- 🔧 Production-Ready: GitHub Actions, deployment configs, monitoring included
- 🚀 Easy Setup: Get started in under 5 minutes
📦 Installation
Via npm (Recommended)
npm install -g claudecraft
claudecraft-init my-awesome-app react-python
Via pip
pip install claudecraft
claudecraft-init my-awesome-app react-python
Via GitHub Template
- Click "Use this template" on GitHub
- Clone your new repository
- Run
./setup-project.sh "my-app" react-python
🤖 Meet Your Development Team
👨💼 Planner Agent
- Creates detailed Product Requirements Documents (PRDs)
- Breaks down complex features into implementable tasks
- Makes architectural decisions and manages project vision
- Long-running sessions, rich context
👨💻 Coder Agent
- Implements single features with security and quality focus
- Follows established patterns and coding standards
- Creates comprehensive tests and documentation
- Task-focused sessions, implementation expertise
🧪 Tester Agent
- Develops comprehensive test strategies
- Finds bugs and performance issues
- Validates security measures and edge cases
- Verification-focused, quality assurance
📝 Documenter Agent
- Maintains user guides, API docs, and README files
- Updates documentation with every feature change
- Manages GitHub repository files and templates
- Documentation expertise, user experience focus
🔍 Reviewer Agent
- Conducts thorough code quality assessments
- Performs security audits and vulnerability scanning
- Provides improvement recommendations and best practices
- Quality gates, security validation
⚡ Quick Start
1. Install ClaudeCraft
# Choose your preferred method:
npm install -g claudecraft # Node.js users
pip install claudecraft # Python users
# OR use GitHub template
2. Initialize Your Project
# Create new project with your preferred tech stack:
claudecraft-init "my-awesome-app" react-python # React + Python/FastAPI
claudecraft-init "my-awesome-app" react-node # React + Node.js/Express
claudecraft-init "my-awesome-app" react-only # React frontend only
3. Start Development with Claude
# Your first prompt to Claude (Claude Code, ChatGPT, etc.):
"Initialize project development with ClaudeCraft.
Create PRD for [your app idea] and begin feature implementation."
That's it! Your multi-agent development environment is ready. 🎉
🏗️ What You Get Out of the Box
📋 Project Management
- PRD Template: Comprehensive requirements documentation
- Feature Tracking: todo.md with prioritized development tasks
- Memory System: Persistent project knowledge and decisions
- Bug Tracking: Systematic issue management and resolution
🛡️ Security & Quality
- Security Rules: OWASP Top 10 compliance built-in
- Quality Gates: Code review, testing, and performance standards
- Agent Handoffs: Structured transitions between development phases
- CI/CD Pipelines: GitHub Actions for testing and deployment
🔧 Development Tools
- Multiple Tech Stacks: React+Python, React+Node.js, React-only
- Environment Setup: Automated dependency management
- Testing Framework: Unit, integration, and E2E testing
- Documentation: Auto-generated and maintained docs
📁 Project Structure
your-project/
├── frontend/ # React/Next.js application
├── backend/ # Python/FastAPI or Node.js/Express
├── docs/ # Project documentation
├── workspace/ # ClaudeCraft agent workspace
│ ├── prd.md # Product Requirements Document
│ ├── todo.md # Feature implementation checklist
│ ├── memory-bank.md # Project knowledge base
│ ├── bug-tracker.md # Issue tracking
│ └── agent-instructions/ # AI agent guidance
├── .github/ # CI/CD workflows and templates
├── CLAUDE.md # AI agent context and rules
└── framework-workflow.md # Complete development workflow
🌟 Framework Advantages
Traditional AI Development Problems
❌ Context loss between sessions
❌ Inconsistent code quality
❌ Security vulnerabilities
❌ Poor documentation
❌ No systematic testing
❌ Chaotic development process
ClaudeCraft Solutions
✅ Persistent Memory: All decisions and context preserved
✅ Quality Gates: Automated quality and security checks
✅ Structured Workflow: Clear agent roles and handoff procedures
✅ Documentation-Driven: Self-updating, comprehensive docs
✅ Test-Driven Development: Built-in testing at every level
✅ Production-Ready: Deployment and monitoring included
🎯 Supported Technology Stacks
React + Python/FastAPI (Recommended)
- Frontend: React, Next.js, TypeScript, Tailwind CSS
- Backend: Python, FastAPI, SQLAlchemy, Alembic
- Database: PostgreSQL (production), SQLite (development)
- Testing: Jest, pytest, Playwright
- Deployment: Vercel + Railway
React + Node.js/Express
- Frontend: React, Next.js, TypeScript, Tailwind CSS
- Backend: Node.js, Express, TypeScript
- Database: PostgreSQL, MongoDB
- Testing: Jest, Supertest, Playwright
- Deployment: Vercel + Railway
React Only
- Frontend: React, Next.js, TypeScript, Tailwind CSS
- Backend: External APIs or serverless functions
- Testing: Jest, Playwright
- Deployment: Vercel, Netlify
🚀 Real-World Example
# 1. Create project
claudecraft-init "task-manager" react-python
# 2. Start with Planner Agent (in Claude)
"Create PRD for a team task management app with user auth,
project boards, task assignment, and real-time collaboration."
# 3. Coder Agent implements features
"Implement user authentication with JWT tokens and role-based access control."
# 4. Tester Agent validates
"Create comprehensive test suite for authentication system including edge cases."
# 5. Documenter Agent updates docs
"Update API documentation and user guide for authentication features."
# 6. Reviewer Agent ensures quality
"Perform security audit and code quality review of authentication system."
# Result: Production-ready feature with documentation, tests, and security validation!
📚 Documentation
| Document | Purpose |
|---|---|
| Setup Guide | Step-by-step setup and usage |
| Agent Workflows | How agents work together |
| Best Practices | Development guidelines |
| Complete Workflow | Full development process |
| Security Rules | Security requirements |
| Handoff Procedures | Agent transition protocols |
🔧 Advanced Features
Agent Handoff System
Structured transitions between development phases with complete context preservation.
Quality Metrics
- Code coverage tracking
- Performance monitoring
- Security vulnerability scanning
- Technical debt assessment
Memory Management
- Persistent project knowledge
- Architectural decision records
- Bug pattern recognition
- Success pattern replication
Integration Support
- Crystal multi-session management
- CrewAI agent orchestration
- GitHub Actions CI/CD
- Monitoring and alerting
🌍 Community & Support
- GitHub: https://github.com/stack-junkie/ClaudeCraft
- Issues: Report bugs or request features
- Discussions: Community discussions
- npm: https://www.npmjs.com/package/claudecraft
- PyPI: https://pypi.org/project/claudecraft/
🤝 Contributing
ClaudeCraft follows its own multi-agent development principles:
- Planner Agent: Review and plan improvements
- Coder Agent: Implement changes with tests
- Tester Agent: Validate all functionality
- Documenter Agent: Update documentation
- Reviewer Agent: Quality and security review
See CONTRIBUTING.md for detailed guidelines.
📄 License
MIT License - see LICENSE for details.
⭐ Why Choose ClaudeCraft?
"Finally, a development framework that brings order to AI-assisted development. The multi-agent approach eliminates the chaos and ensures every feature is secure, tested, and documented."
- 🎯 Structured Development: Clear roles and responsibilities
- 🛡️ Security Built-In: OWASP compliance from day one
- 📈 Quality Assured: Comprehensive testing and code review
- 📚 Self-Documenting: Always up-to-date documentation
- ⚡ High Velocity: Efficient agent handoffs
- 🔧 Production-Ready: Deployment and monitoring included
🌟 Get Started Today
Ready to transform your development workflow?
- Install ClaudeCraft:
npm install -g claudecraftorpip install claudecraft - Initialize your project:
claudecraft-init "my-app" react-python - Start building with structured, secure, quality-focused development
Join the revolution in AI-assisted software development! 🚀
Built by developers, for developers. Designed for quality, security, and velocity.
Metadata
Release files for claudecraft 1.0.2
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| claudecraft-1.0.2.tar.gz | 9.2 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| claudecraft-1.0.2-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 16.5 kB
Release files / claudecraft-1.0.2.tar.gz
| Download URL | claudecraft-1.0.2.tar.gz |
|---|---|
| Size | 9.2 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
1d216bf6a6f9ec8ce354d186bde41412e7130acc3a35022b5f0a8e5b34ef8590
|
|
BLAKE2b-256 checksum How to use checksums |
6dc0c8ff0d6d278e32a436b524d1b761b18f756fcb0fbebb2c3613a1c2feece6
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.13.5
|
Release files / claudecraft-1.0.2-py3-none-any.whl
| Download URL | claudecraft-1.0.2-py3-none-any.whl |
|---|---|
| Size | 7.3 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
ef2a6a8854d2a946b720fb48d7ea50518de4c245cee78568a75663eedf7d3ab1
|
|
BLAKE2b-256 checksum How to use checksums |
1f1c846b956f7b2675709932231f7cc9087c29fe567802e164cda3bd166e158d
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.13.5
|