MCP AI Developer Agent

# MCP AI Developer Agent

An AI-powered developer assistant built with the Model Context Protocol (MCP), OpenRouter, and TypeScript.

MCP AI Developer Agent is a command-line AI coding assistant that combines Large Language Models with MCP tools to understand, analyze, and modify real codebases.

Instead of being a simple chatbot, it acts as an AI development agent that can inspect projects, search code, read files, and eventually edit code, execute commands, and automate development workflows.

Features

• Multiple AI models via OpenRouter

• MCP Client & MCP Server

• Interactive CLI interface

• Project selection

• AI model selection

• Filesystem tools

• TypeScript implementation

• Modular architecture

• Ready for future tool extensions

Related MCP server: File MCP Server

Current MCP Tools

Planned Features

• AI Chat

• Automatic MCP tool calling

• Git integration

• Terminal command execution

• Project indexing

• Code generation

• Code editing

• Multi-agent workflows

• Plugin system

• Desktop application (Electron/Tauri)

Project Structure

src/
│
├── agent/
│   ├── chat.ts
│   ├── model-selector.ts
│   └── project-selector.ts
│
├── config/
│   ├── config.ts
│   └── openrouter.ts
│
├── mcp/
│   ├── client.ts
│   └── server.ts
│
├── openrouter/
│   └── models.ts
│
├── tools/
│   ├── filesystem.ts
│   └── ...
│
└── cli.ts

Tech Stack

• TypeScript

• Node.js

• Model Context Protocol (MCP)

• OpenRouter API

• Zod

• Inquirer

• TSX

Installation

Clone the repository:

git clone https://github.com/<username>/mcp-dev-agent.git
cd mcp-dev-agent

Install dependencies:

npm install

Create a .env file:

OPENROUTER_API_KEY=your_openrouter_api_key
OPENROUTER_BASE_URL=https://openrouter.ai/api/v1

Development

Start the development server:

npm run dev

Build the project:

npm run build

Run the compiled version:

npm start

Example Workflow

MCP AI Developer Agent

✔ Select AI Model
  Claude
  GPT
  Gemini

✔ Select Project

E:\my-project

You >

Analyze this project

Find authentication

Fix TypeScript errors

Generate tests

Explain this component

Vision

The goal of this project is to build an open, extensible AI developer assistant powered by the Model Context Protocol.

The long-term roadmap includes:

• Understanding entire codebases

• Autonomous tool usage

• Safe code modifications

• Git-aware workflows

• Multi-model support

• Extensible plugin ecosystem

• Desktop application

• Local and cloud model support

Roadmap

• MCP Server

• MCP Client

• Interactive CLI

• Filesystem tools

• OpenRouter integration

• AI chat

• Automatic tool calling

• Git tools

• Terminal tools

• Project indexing

• Code editing agent

• Multi-agent workflows

• Desktop application

Contributing

Contributions, feature requests, and bug reports are welcome.

If you'd like to contribute:

1. Fork the repository.

2. Create a feature branch.

3. Commit your changes.

4. Open a Pull Request.

License

This project is licensed under the MIT License.

Acknowledgements

• Model Context Protocol (MCP)

• OpenRouter

• TypeScript

• Node.js

• Open Source AI Community