Skip to main content

AI Assistants

The Codegen Project provides a Model Context Protocol (MCP) server that enables AI assistants to help you create and manage The Codegen Project configurations and interactions. This allows tools like Claude Code, Cursor, Windsurf, and other MCP-compatible assistants to understand The Codegen Project and assist you more effectively.

Why Use the MCP Server?​

When connected to an AI assistant, the MCP server provides:

  • Configuration Creation: AI can generate complete codegen configuration files based on your requirements
  • Configuration Adaption: AI can help adapt your codegen configuration based on your requirements
  • Interacting with generated code: Can help inject the right context for interacting with the generated code
  • Documentation Access: AI has direct access to all project documentation.

Quick Setup​

Claude Code​

Add to ~/.claude/mcp.json:

{
  "mcpServers": {
    "the-codegen-project": {
      "url": "https://the-codegen-project.org/api/mcp"
    }
  }
}

Cursor​

Add to .cursor/mcp.json in your project root:

{
  "mcpServers": {
    "the-codegen-project": {
      "url": "https://the-codegen-project.org/api/mcp"
    }
  }
}

Windsurf​

Add to your Windsurf MCP configuration:

{
  "mcpServers": {
    "the-codegen-project": {
      "url": "https://the-codegen-project.org/api/mcp"
    }
  }
}

Other MCP-Compatible Tools​

Any tool that supports the Model Context Protocol can connect using the URL:

https://the-codegen-project.org/api/mcp

Self-Hosting​

You can run your own instance of the MCP server for development or private use. See the MCP server repository for setup instructions.

Q&A​

Q: If AI can generate code, why use The Codegen Project?​

The generator gives you deterministic, repeatable output from a stable input (spec + config). That makes CI consistent, supports large-team conventions, and lets you regenerate code without drift across runs or models.

Q: Won’t AI output be maintained by developers anyway?​

Yes, but generators keep a traceable “source of truth” in the configuration. That means you can explain why code exists, regenerate it reliably, and keep updates consistent across many services.

Q: When should I prefer AI over a generator?​

Use AI for exploration, prototypes, or one-off scripts. Use the generator when you want consistent output, shared conventions, automated regeneration.

Q: Does this project compete with AI assistants?​

It complements them. The MCP server gives assistants a deterministic interface to create and adjust configs, while the generator produces the exact code your repo expects.

Q: What’s the biggest advantage over “just using AI”?​

Repeatability. The same input produces the same output every time, which is critical for CI, refactors, and multi-team consistency.