Summary

This project provides an experimental collection of specialized AI agents designed to automate Test-Driven Development (TDD) and rapid prototyping. Targeting developers who value rigorous testing and efficient MVP creation, it aims to accelerate development cycles by enforcing TDD methodologies and leveraging real documentation.

How It Works

The system employs a hub-and-spoke architecture, routing requests through a central @task-orchestrator agent. Specialist agents then execute tasks, adhering strictly to the RED → GREEN → REFACTOR TDD cycle. A key differentiator is the integration of Context7, enabling agents to consult real-time library documentation rather than relying on potentially outdated or fabricated knowledge, thereby improving code accuracy and reducing manual lookup effort.

Quick Start & Requirements

Installation is initiated via npx claude-code-collective init. Alternative installation flags (--minimal, --testing-only, --hooks-only, --interactive) allow for customized setups. Prerequisites include Node.js (>= 16.0.0), NPM (>= 8.0.0), and the Claude Code environment with MCP support and its hook system. A restart of Claude Code is necessary post-installation for the hooks to function correctly.

Highlighted Details

• Features over 30 specialized AI agents for diverse development tasks.
• Enforces a strict TDD contract (tests first, then implementation, then refactoring) via automated hooks.
• Integrates Context7 to access current, real-world library documentation.
• Utilizes a central orchestrator for intelligent task delegation and breakdown.
• Includes experimental research agents for complexity analysis and requirement decomposition.

Maintenance & Community

This project is described as a personal endeavor. While issues and pull requests for fixes or prompt improvements are welcomed, rapid responses are not expected. No specific community channels (like Discord/Slack) or formal support structures are detailed.

Licensing & Compatibility

The project is released under the MIT License, permitting broad use, modification, and distribution. Its experimental nature and strong opinion on TDD are primary considerations for adoption, rather than explicit compatibility restrictions beyond standard open-source terms.

Limitations & Caveats

The system is explicitly labeled as experimental and opinionated, particularly regarding its strict adherence to TDD. Users must have Node.js, NPM, and a compatible Claude Code setup. A restart of Claude Code is required for full functionality. Some research agents may exhibit slower performance or over-analyze tasks, and documentation is currently dispersed across project files. It is not intended as a production-ready enterprise solution.