> ## Documentation Index
> Fetch the complete documentation index at: https://codemodcom.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Codemod MCP

export const MCPDemo = () => {
  return <div style={{
    position: 'relative',
    paddingBottom: 'calc(63.109540636042404% + 41px)',
    height: '0',
    width: '100%'
  }}>
      <iframe src="https://demo.arcade.software/J9LMa2GnN8BBG6PpuEiM?embed&amp;embed_mobile=inline&amp;embed_desktop=inline&amp;show_copy_link=true" title="Codemod MCP" frameBorder="0" loading="lazy" allowFullScreen allow="clipboard-write" style={{
    position: 'absolute',
    top: 0,
    left: 0,
    width: '100%',
    height: '100%',
    colorScheme: 'dark'
  }} />
    </div>;
};

Codemod MCP (Model Context Protocol) provides AI assistants with tools for code analysis, AST manipulation, and codemod creation. The MCP server enables AI tools like Claude, Cursor, and others to understand code structure, create codemods, and perform automated code transformations.

## Quick setup

<Steps>
  <Step title="Open IDE settings">
    **For Cursor IDE:**

    * Open Cursor Settings
    * Navigate to **MCP & Integrations**
    * Click **New MCP Server** under MCP Tools
  </Step>

  <Step title="Add MCP server configuration">
    Add the following configuration to your MCP settings:

    ```json theme={null}
    {
      "mcpServers": {
        "codemod": {
          "command": "npx",
          "args": ["codemod@latest", "mcp"]
        }
      }
    }
    ```
  </Step>

  <Step title="Restart your IDE">
    Save the configuration file and restart your IDE to activate the Codemod MCP server.
  </Step>
</Steps>

## Available tools

The Codemod MCP server provides these tools for AI assistants:

* **`dump_ast`** - Dump AST nodes in an AI-friendly format for given source code and language
* **`get_node_types`** - Get compressed tree-sitter node types for specific programming languages
* **`run_jssg_tests`** - Run tests for jssg (TypeScript transformation scripts) codemods with test cases
* **`get_jssg_instructions`** - Get jssg instructions for creating TypeScript transformation scripts
* **`get_ast_grep_instructions`** - Get ast-grep instructions for creating ast-grep rules
* **`get_codemod_cli_instructions`** - Get Codemod CLI instructions for creating codemods

## Building Codemod packages with MCP

<Frame>
  <MCPDemo />
</Frame>

<Steps>
  <Step title="Ask the AI for a specific codemod">
    Describe your transformation [clearly](#best-practices):

    ```
    "Create a jssg codemod that converts all console.log statements to console.info in TypeScript files. 
    The codemod should:
    - Target .ts and .tsx files
    - Only transform console.log calls (not console.error, console.warn, etc.)
    - Preserve all arguments and formatting
    - Include test cases for the transformation"
    ```
  </Step>

  <Step title="Let MCP build the Codemod package">
    The AI will use MCP tools to:

    * Analyze your codebase structure with `dump_ast`
    * Get TypeScript node types with `get_node_types`
    * Create the jssg codemod script
    * Generate test cases with `run_jssg_tests`
    * Set up the [`workflow.yaml` and `codemod.yaml`](/cli/packages/package-structure) files
  </Step>

  <Step title="Validate the workflow">
    ```bash theme={null}
    npx codemod@latest workflow validate -w workflow.yaml
    ```

    This ensures your workflow file is syntactically correct and follows the schema.
  </Step>

  <Step title="Understand and run the output package">
    The MCP will create a complete Codemod package structure:

    ```
    my-codemod/
    ├── codemod.yaml        # Package metadata
    ├── workflow.yaml       # Workflow definition
    ├── scripts/
    │   └── codemod.ts      # Your jssg transformation
    └── tests/              # Test cases
    ```

    Run your codemod:

    ```bash theme={null}
    npx codemod@latest workflow run -w ./my-codemod/
    ```
  </Step>
</Steps>

## Why use Codemod MCP?

Codemod MCP democratizes code transformation by integrating robust deterministic engines into AI-powered development workflows:

* **Individual Developers**: Take care of repetitive refactors without manual effort
* **Accessibility**: Makes tackling tech debt more approachable for developers of all skill levels
* **AI Integration**: Seamlessly integrates deterministic code transformation engines into AI IDE workflows

## Codemod MCP vs Codemod Studio

|               | Codemod MCP                                        | Codemod Studio                           |
| ------------- | -------------------------------------------------- | ---------------------------------------- |
| **Interface** | Integrated into your IDE                           | Web-based interface                      |
| **AI Model**  | Your choice of AI model                            | Built-in AI assistant                    |
| **Control**   | Granular control over codemod creation             | Quick prototyping and sharing            |
| **Iteration** | Direct, instant control in your IDE                | Web-based iteration                      |
| **Sharing**   | Publishing to Codemod Registry and version control | Built-in sharing                         |
| **Best For**  | Power users, framework maintainers, pro codemods   | Quick codemod creation and collaboration |

## Best practices

When working with AI assistants through Codemod MCP, follow these guidelines for better results:

<Steps>
  <Step title="Be specific about output format">
    Tell the AI exactly what type of codemod you want:

    * **jssg:** TypeScript-based transformation scripts (codemods) that operate on ASTs generated by ast-grep, allowing you to transform code in any [supported language](https://ast-grep.github.io/reference/languages.html).
    * **ast-grep YAML rule:** general-purpose transformations
    * **shell workflow:** complex multi-step migrations
  </Step>

  <Step title="Provide clear requirements">
    Describe your transformation in plain language with concrete examples:

    * Show before/after code snippets
    * Specify edge cases to handle or avoid
    * Include any constraints or special conditions
  </Step>

  <Step title="Break down complex changes">
    Instead of one large transformation, create focused codemods for specific patterns:

    * One codemod per logical change
    * Simpler codemods are more maintainable
    * Easier to test and debug individual transformations
  </Step>
</Steps>

## Troubleshooting

<AccordionGroup>
  <Accordion title="Server won't start">
    If the MCP server fails to start:

    * Ensure you have Node.js 16.0.0 or higher installed
    * Try running `npx codemod@latest mcp` manually to test
  </Accordion>

  <Accordion title="Tools not available">
    If MCP tools aren't showing up:

    * Restart your AI client after adding the MCP server
    * Check the MCP server logs for error messages
    * Verify the JSON configuration syntax is correct
    * Ensure your AI client supports Model Context Protocol
  </Accordion>

  <Accordion title="Performance issues">
    For slow performance:

    * More complex transformations may take longer to build
    * Consider using specific file patterns to limit analysis scope
    * Check system resources and network connectivity
    * Use faster models if available
  </Accordion>
</AccordionGroup>