llms.txt@web/public git:20260627.76887e6A

Immutable. This exact content is served forever at /api/v1/blob/d0d588ace43a69ba.

# AI DevKit

> AI DevKit is an open-source control plane for AI coding agents. It works alongside Cursor, Claude Code, GitHub Copilot, Gemini CLI, Codex, opencode, Pi, and other tools.

AI DevKit solves common multi-agent coding problems: scattered terminals, duplicated setup files, copy-pasted context, forgotten project decisions, and unverified completion claims. It provides one project-local config, one console, cross-agent communication, local-first SQLite memory, workflow skills, and verification.

## Core Features

- **One Local Config**: Project-local setup for supported coding agents from `.ai-devkit.json`
- **Agent Console and Communication**: Inspect running sessions and send prompts, logs, or stdin to the right agent
- **Long-Term Memory**: 100% local SQLite-based knowledge storage for coding standards, patterns, and decisions across sessions
- **Structured Development Workflows**: Phase-based guidance through requirements, design, planning, implementation, testing, and code review
- **Skills System**: Community-driven plugins that teach AI new capabilities and domain-specific patterns
- **Multi-Agent Support**: Works with a broad set of AI coding environments including Cursor, Claude Code, GitHub Copilot, Gemini CLI, Codex, opencode, and others

## Quick Start

Install via npm:

```
npm install -g ai-devkit
```

Initialize in your project:

```
npx ai-devkit@latest init
```

## Key Skills

- `dev-lifecycle` - Start a new feature, validate requirements and design, execute plans, and run pre-commit reviews
- `dev-commit` - Commit only intended, verified changes with a conventional message
- `tdd` - Add test coverage with a test-first workflow
- `structured-debug` - Debug systematically with reproduction, hypotheses, fixes, and verification
- `memory` - Store important guidelines and patterns in memory
- `document-code` - Analyze and document existing code

## Documentation

- [What is AI DevKit?](https://ai-devkit.com/docs/0-what-is-ai-devkit): Overview, problem statement, and how AI DevKit works
- [Getting Started](https://ai-devkit.com/docs/1-getting-started): Installation, project initialization, and first steps
- [Supported AI Agents](https://ai-devkit.com/docs/2-supported-agents): Reference for supported AI coding environments and setup capabilities
- [Development with AI DevKit](https://ai-devkit.com/docs/3-development-with-ai-devkit): Full development lifecycle workflow
- [Debugging with AI DevKit](https://ai-devkit.com/docs/4-debugging-with-ai-devkit): Structured debugging methodology
- [Understanding Existing Code](https://ai-devkit.com/docs/5-understand-existing-code-with-ai-devkit): Code analysis and knowledge capture
- [Memory](https://ai-devkit.com/docs/6-memory): Local memory system for persistent knowledge
- [Skills](https://ai-devkit.com/docs/7-skills): Community-driven skill management system
- [Agent Management](https://ai-devkit.com/docs/8-agent-management): Experimental agent detection and switching

## Other Pages

- [Home](https://ai-devkit.com): Landing page with features overview and quick start
- [Vision](https://ai-devkit.com/vision): Project philosophy and alignment with "The New Engineering Workflow"
- [Roadmap](https://ai-devkit.com/roadmap): Development roadmap and feature status
- [Skills Registry](https://ai-devkit.com/skills): Browse and search available skills
- [FAQ](https://ai-devkit.com/faq): Frequently asked questions and getting started guides

## Project Links

- GitHub: https://github.com/Codeaholicguy/ai-devkit
- npm: https://www.npmjs.com/package/ai-devkit
- Website: https://ai-devkit.com