AI DevKit
Make AI coding agents follow a repeatable engineering workflow with planning, memory, verification, skills, and review. One config. All agents. Same workflow.
What is AI DevKit?
AI DevKit is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to make ai coding agents follow a repeatable engineering workflow with planning, memory, verification, skills, and review. one config. all agents. same workflow.
Make AI coding agents follow a repeatable engineering workflow with planning, memory, verification, skills, and review. One config. All agents. Same workflow.
This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Make AI coding agents follow a repeatable engineering workfl
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx ai-devkitConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use AI DevKit
AI DevKit is a CLI and MCP server that enforces a repeatable, structured engineering workflow across AI coding agents including Claude Code, Cursor, and OpenAI Codex CLI. It bundles eight built-in skills covering requirements, design, planning, implementation, TDD, debugging, documentation, and code simplification, and persists agent memory in a SQLite-backed knowledge base so context survives across sessions. A single configuration file synchronizes settings across all agents, and additional skills can be pulled from a community registry with contributions from Anthropic, Vercel, Supabase, and others.
Prerequisites
- Node.js 18 or later installed
- At least one AI coding agent installed: Claude Code, Cursor, or OpenAI Codex CLI
- An MCP-compatible client that can run npx-based servers
- Git initialized in the project directory where you run ai-devkit init
Initialize AI DevKit in your project
Run the init command from your project root. This creates the .ai-devkit.json master config, generates agent-specific folders (.claude/, .cursor/, .codex/), and scaffolds the docs/ai/ documentation structure.
npx ai-devkit@latest initReview and customize the generated configuration
Open .ai-devkit.json to review the eight pre-installed skills (dev-lifecycle, verify, memory, tdd, structured-debug, document-code, simplify-implementation, technical-writer). Enable or disable skills to match your team's workflow.
Add the MCP server to your AI agent configuration
The init command populates each agent's settings.json automatically. For manual addition to Claude Code, add the server under mcpServers in your Claude configuration.
Install additional skills from the registry
Browse the community skill registry and install specialized skills from Anthropic, Vercel, Supabase, and other contributors. Skills are yaml files that define prompting patterns and tool sequences.
npx ai-devkit@latest init --template ./senior-engineer.yamlControl running agent sessions
Use the agent subcommand to list active sessions, send instructions to a running agent, or pipe test output directly into an agent session for automated verification.
npx ai-devkit agent list
npx ai-devkit agent send "run tests and report" --id <agent-name> --wait
npm test 2>&1 | npx ai-devkit agent send --id <agent-name> --stdinAI DevKit Examples
Client configuration
Add AI DevKit to Claude Desktop's mcpServers configuration to give Claude access to the skill and workflow enforcement tools.
{
"mcpServers": {
"ai-devkit": {
"command": "npx",
"args": ["ai-devkit"]
}
}
}Prompts to try
Example prompts that leverage AI DevKit's structured workflow skills inside an AI coding agent session.
- "Follow the dev-lifecycle skill to gather requirements for a user authentication feature before writing any code"
- "Use the tdd skill to write failing tests for the payment module first, then implement until they pass"
- "Run the verify skill to confirm the build and test suite pass before marking this task complete"
- "Search the memory knowledge base for what we decided about the database schema last week"
- "Apply the structured-debug skill to reproduce and fix this null pointer exception"Troubleshooting AI DevKit
The init command fails with a permissions error when creating .claude/ or .cursor/ directories
Run the command from your project root where you have write permissions. If running in a managed environment, check that the directories don't already exist with conflicting ownership.
Skills are not being followed by the agent — it ignores the dev-lifecycle workflow
Ensure the MCP server is listed in the agent's active settings.json and that the agent session was started after the config was written. Some agents require a restart to pick up new MCP servers.
The memory skill does not persist knowledge between sessions
AI DevKit stores memory in a SQLite file in the project directory. Verify the .ai-devkit.json config points to a writable path and that the file hasn't been deleted by a clean script.
Frequently Asked Questions about AI DevKit
What is AI DevKit?
AI DevKit is a Model Context Protocol (MCP) server that make ai coding agents follow a repeatable engineering workflow with planning, memory, verification, skills, and review. one config. all agents. same workflow. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install AI DevKit?
Follow the installation instructions on the AI DevKit GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with AI DevKit?
AI DevKit works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is AI DevKit free to use?
Yes, AI DevKit is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
AI DevKit Alternatives — Similar Coding Agents Servers
Looking for alternatives to AI DevKit? Here are other popular coding agents servers you can use with Claude, Cursor, and VS Code.
Dify
★ 142.2kProduction-ready platform for agentic workflow development.
Ruflo
★ 54.0k🌊 The leading agent orchestration platform for Claude. Deploy intelligent multi-agent swarms, coordinate autonomous workflows, and build conversational AI systems. Features enterprise-grade architecture, self-learning swarm intelligence, RAG integrat
Goose
★ 45.7kan open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
Antigravity Awesome Skills
★ 38.3kInstallable GitHub library of 1,400+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, Antigravity, and more. Includes installer CLI, bundles, workflows, and official/community skill collections.
AgentScope
★ 25.5kBuild and run agents you can see, understand and trust.
Serena
★ 24.5kA coding agent toolkit that provides IDE-like semantic code retrieval and editing tools, enabling LLMs to efficiently navigate and modify codebases using symbol-level operations instead of basic file reading and string replacements.
Browse More Coding Agents MCP Servers
Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up AI DevKit in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use AI DevKit?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.