Model Context Protocol (MCP)

The Model Context Protocol (MCP) is an open standard that enables AI applications to securely connect to external data sources and tools. Codegen provides an MCP server that allows AI editors like Cursor to access Codegen’s powerful codemod generation and API capabilities directly within your development workflow.

What is MCP?

MCP acts as a bridge between AI assistants and external tools, allowing them to:

Access external data sources and APIs
Execute tools and commands safely
Provide rich context to improve AI responses
Maintain security through standardized protocols

With Codegen’s MCP integration, your AI editor can:

Generate codemods using the Codegen CLI
Create and manage agent runs
Access organization and user information
Get expert guidance on Codegen best practices
Streamline your code transformation workflows

Prerequisites

Before setting up the MCP server, ensure you have:

Codegen CLI installed: Install via pip or your preferred package manager
```
pip install codegen
```

Authentication: Set up your Codegen API credentials

# Set your API key as an environment variable
export CODEGEN_API_KEY="your-api-key-here"

# Or initialize a Codegen project
codegen init

AI Editor with MCP support: Currently supported editors include:
- Cursor (recommended)
- Claude Desktop
- Cline (VS Code extension)

Quick Start

1. Start the MCP Server

The easiest way to start the Codegen MCP server is using the built-in CLI command:

codegen mcp

This starts the server with default settings (stdio transport). The server will display:

🚀 Starting Codegen MCP server...
📡 Using stdio transport

For additional options:

# Start with HTTP transport on a specific port
codegen mcp --transport http --port 8080

# Start with custom host
codegen mcp --host 0.0.0.0 --port 3000

# View all available options
codegen mcp --help

Note: HTTP transport is currently not fully implemented and will fall back to stdio transport.

2. Configure Your AI Editor

The configuration depends on your AI editor. See the sections below for specific setup instructions.

Cursor Configuration

Cursor provides the most seamless MCP integration experience. Follow these steps:

Method 1: Using Cursor Settings (Recommended)

Open Cursor Settings
- Press Cmd/Ctrl + , to open settings
- Navigate to Features → Model Context Protocol
Add New MCP Server
- Click “Add Server”
- Fill in the following details:
```
Name: Codegen
Command: codegen mcp
```
Save and Restart
- Click “Save”
- Restart Cursor to activate the MCP server

Method 2: Manual Configuration File

If you prefer manual configuration, you can edit Cursor’s MCP configuration file directly:

Locate the configuration file:
- macOS: ~/Library/Application Support/Cursor/User/globalStorage/mcp.json
- Windows: %APPDATA%\Cursor\User\globalStorage\mcp.json
- Linux: ~/.config/Cursor/User/globalStorage/mcp.json

Add the Codegen MCP server:

{
  "mcpServers": {
    "codegen": {
      "command": "codegen",
      "args": ["mcp"],
      "env": {
        "CODEGEN_API_KEY": "your-api-key-here"
      }
    }
  }
}

Restart Cursor to load the new configuration.

Verification

To verify the MCP server is working in Cursor:

Open a new chat in Cursor
Look for “Available Tools” in the chat interface
You should see Codegen MCP tools listed, such as:
- generate_codemod
- create_agent_run
- get_organizations
- get_users

Alternative Editors

Claude Desktop

For Claude Desktop, add this configuration to your claude_desktop_config.json:

{
  "mcpServers": {
    "codegen": {
      "command": "codegen",
      "args": ["mcp"],
      "env": {
        "CODEGEN_API_KEY": "your-api-key-here"
      }
    }
  }
}

Configuration file locations:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json

Cline (VS Code Extension)

For the Cline VS Code extension, add this to your MCP settings:

{
  "mcpServers": {
    "codegen": {
      "command": "codegen",
      "args": ["mcp"],
      "env": {
        "CODEGEN_API_KEY": "your-api-key-here"
      }
    }
  }
}

Available Tools

Once configured, the Codegen MCP server provides these tools to your AI editor:

`generate_codemod`

Generate a new codemod for a specific task and codebase.

Parameters:

title: The title of the codemod (hyphenated format)
task: Description of what the codemod should accomplish
codebase_path: Absolute path to your codebase directory

Example usage in chat:

Generate a codemod called "add-logging" that adds console.log statements to all functions in my React components at /path/to/my/project

What it does: Provides instructions to use the Codegen CLI to create a codemod with the specified title and task description.

`create_agent_run`

Create a new agent run in your Codegen organization.

Parameters:

org_id: Your organization ID
prompt: The task/prompt for the agent to execute
repo_name: Repository name (optional)
branch_name: Branch name (optional)

Example usage in chat:

Create an agent run in organization 123 with the prompt "Fix all TypeScript errors in the src directory"

`get_agent_run`

Get details of a specific agent run.

Parameters:

org_id: Organization ID
agent_run_id: Agent run ID

Example usage in chat:

Get the status of agent run 456 in organization 123

`get_organizations`

Get list of organizations you have access to.

Parameters:

page: Page number (default: 1)
limit: Number of organizations per page (default: 10)

Example usage in chat:

Show me all the organizations I have access to

`get_users`

Get list of users in an organization.

Parameters:

org_id: Organization ID
page: Page number (default: 1)
limit: Number of users per page (default: 10)

Example usage in chat:

List all users in organization 123

`get_user`

Get details of a specific user in an organization.

Parameters:

org_id: Organization ID
user_id: User ID

Example usage in chat:

Get details for user 789 in organization 123

Authentication

The MCP server uses API key authentication. You can provide your API key in several ways:

Environment Variable (Recommended)

export CODEGEN_API_KEY="your-api-key-here"

MCP Configuration

Include the API key in your MCP server configuration:

{
  "mcpServers": {
    "codegen": {
      "command": "codegen",
      "args": ["mcp"],
      "env": {
        "CODEGEN_API_KEY": "your-api-key-here"
      }
    }
  }
}

Getting Your API Key

Visit the Codegen dashboard
Navigate to your account settings
Generate or copy your API key
Set it as an environment variable or in your MCP configuration

Troubleshooting

Common Issues

MCP Server Not Starting

Verify Codegen is properly installed: codegen --version
Check that the codegen mcp command works: codegen mcp --help
Ensure your API key is set correctly

Tools Not Appearing in Editor

Restart your AI editor after configuration changes
Check the MCP server configuration syntax
Verify the codegen command is in your system PATH
Look for error messages in the editor’s developer console

Authentication Errors

Verify your API key is correct and not expired
Check that the CODEGEN_API_KEY environment variable is set
Ensure you have the necessary permissions in your organization

Permission Errors

Ensure the codegen command is executable
Check file permissions on configuration files
Try running the editor with appropriate permissions

Path Resolution Issues

Use absolute paths in configuration when possible
Verify the codegen command location: which codegen
Consider using the full path to the codegen executable

Advanced Configuration

Custom Server Path If you need to specify a custom path to the Codegen installation:

{
  "mcpServers": {
    "codegen": {
      "command": "/path/to/your/codegen",
      "args": ["mcp"]
    }
  }
}

Environment Variables You can set additional environment variables for the MCP server:

{
  "mcpServers": {
    "codegen": {
      "command": "codegen",
      "args": ["mcp"],
      "env": {
        "CODEGEN_API_KEY": "your-api-key-here",
        "CODEGEN_API_BASE_URL": "https://api.codegen.com"
      }
    }
  }
}

HTTP Transport (Future) For remote or multi-user setups, HTTP transport will be available:

# Start server with HTTP transport (when implemented)
codegen mcp --transport http --port 8080

Best Practices

Secure Your API Key: Never commit API keys to version control. Use environment variables or secure configuration management.
Use Descriptive Codemod Names: Use clear, hyphenated names for generated codemods
```
Good: "add-error-handling", "migrate-to-hooks"
Avoid: "fix", "update", "change"
```

Provide Clear Task Descriptions: Be specific about what you want the codemod to accomplish

Good: "Add try-catch blocks around all async function calls in React components"
Avoid: "Add error handling"

Test Generated Codemods: Always review and test generated codemods before applying them to production code
Use Version Control: Commit your code before running codemods to easily revert if needed
Monitor Agent Runs: Use the get_agent_run tool to monitor the progress and results of your agent runs

Examples

Example 1: Generate a React Hook Migration Codemod

In your AI editor chat:

Use the generate_codemod tool to create a codemod called "migrate-class-to-hooks" that converts React class components to functional components with hooks. The codebase is located at /Users/myname/projects/my-react-app

Example 2: Create an Agent Run

Create an agent run in my organization to fix all TypeScript errors in the src directory of my project

Example 3: Monitor Agent Progress

Check the status of agent run 123 in organization 456 to see if it's completed

Example 4: List Team Members

Show me all users in organization 789 so I can see who's on my team

Resources

The MCP server also provides these resources for additional context:

system://agent_prompt: Information about the Codegen SDK and platform
system://setup_instructions: Environment setup instructions
system://manifest: Server metadata and version information

Support

If you encounter issues with the MCP integration:

Check the documentation: Review this guide and the official MCP documentation
Test the CLI: Ensure codegen mcp works independently
Check logs: Look for error messages in your editor’s developer console
Community support: Join our Discord community for help
Report bugs: Create an issue on our GitHub repository

What’s Next?

With Codegen MCP integration set up, you can:

Generate codemods directly from your AI editor
Create and manage agent runs without leaving your IDE
Access your organization’s resources and team information
Streamline your code transformation workflows
Integrate Codegen into your existing development process

Explore our API documentation to learn more about Codegen’s capabilities, or check out our examples repository for inspiration.

Overview

Capabilities

Integrations

Sandboxes

Settings

​Model Context Protocol (MCP)

​What is MCP?

​Prerequisites

​Quick Start

​1. Start the MCP Server

​2. Configure Your AI Editor

​Cursor Configuration

​Method 1: Using Cursor Settings (Recommended)

​Method 2: Manual Configuration File

​Verification

​Alternative Editors

​Claude Desktop

​Cline (VS Code Extension)

​Available Tools

​generate_codemod

​create_agent_run

​get_agent_run

​get_organizations

​get_users

​get_user

​Authentication

​Environment Variable (Recommended)

​MCP Configuration

​Getting Your API Key

​Troubleshooting

​Common Issues

​Advanced Configuration

​Best Practices

​Examples

​Example 1: Generate a React Hook Migration Codemod

​Example 2: Create an Agent Run

​Example 3: Monitor Agent Progress

​Example 4: List Team Members

​Resources

​Support

​What’s Next?

Model Context Protocol (MCP)

What is MCP?

Prerequisites

Quick Start

1. Start the MCP Server

2. Configure Your AI Editor

Cursor Configuration

Method 1: Using Cursor Settings (Recommended)

Method 2: Manual Configuration File

Verification

Alternative Editors

Claude Desktop

Cline (VS Code Extension)

Available Tools

`generate_codemod`

`create_agent_run`

`get_agent_run`

`get_organizations`

`get_users`

`get_user`

Authentication

Environment Variable (Recommended)

MCP Configuration

Getting Your API Key

Troubleshooting

Common Issues

Advanced Configuration

Best Practices

Examples

Example 1: Generate a React Hook Migration Codemod

Example 2: Create an Agent Run

Example 3: Monitor Agent Progress

Example 4: List Team Members

Resources

Support

What’s Next?