Repomix Reader MCP

A simple Model Context Protocol (MCP) server for reading and grep-searching Repomix output files. This MCP provides secure file access by restricting operations to a pre-selected set of Repomix output files specified at startup.

Features

• Secure file access - Only allows access to files specified at startup via command line arguments
• File listing - List all available Repomix files with their IDs
• File reading - Read Repomix output files with optional line range selection using file IDs
• Pattern searching - Grep-like pattern searching with regex support and context lines
• Simple interface - File access via numeric IDs instead of complex file paths

Installation

npm install
npm run build

Usage

As an MCP Server

Configure your MCP-compatible client to use this server with the Repomix files you want to make available:

{
  "mcpServers": {
    "repomix-reader": {
      "command": "npx",
      "args": [
        "repomix-reader-mcp",
        "/path/to/first-repomix-output.xml",
        "/path/to/second-repomix-output.xml"
      ]
    }
  }
}

Note: You must specify at least one Repomix output file as a command line argument. Only these files will be accessible through the MCP tools.

Available Tools

list_repomix_files

List all available Repomix files that can be read or searched.

Parameters: None

Returns: A list of available files with their IDs, filenames, and descriptions (extracted from user-provided headers in the XML files).

read_repomix_output

Read content from a Repomix output file with optional line range.

Parameters:

• fileId (string, required): ID of the Repomix output file to read (use list_repomix_files to see available IDs)
• startLine (number, optional): Starting line number (1-based, inclusive)
• endLine (number, optional): Ending line number (1-based, inclusive)

Example:

{
  "fileId": "1",
  "startLine": 1,
  "endLine": 100
}

grep_repomix_output

Search for patterns in a Repomix output file with grep-like functionality.

Parameters:

• fileId (string, required): ID of the Repomix output file to search (use list_repomix_files to see available IDs)
• pattern (string, required): Search pattern (supports regex)
• flags (string, optional): Regex flags (e.g., "i" for case-insensitive, "g" for global)
• contextLines (number, optional): Number of context lines to show around matches (default: 2)
• maxMatches (number, optional): Maximum number of matches to return (default: 100)

Example:

{
  "fileId": "1",
  "pattern": "function.*authenticate",
  "flags": "i",
  "contextLines": 3,
  "maxMatches": 50
}

Development

Install dependencies

npm install

Build the project

npm run build

Run in development mode

npm run dev

Clean build directory

npm run clean

Project Structure

repomix-reader-mcp/
├── .github/
│   └── workflows/               # GitHub Actions CI/CD
│       ├── ci.yml              # Continuous integration
│       ├── release.yml         # Automated releases
│       └── manual-release.yml  # Manual release workflow
├── src/
│   ├── tools/                  # Tool implementations
│   │   ├── index.ts           # Tool exports
│   │   ├── listRepomixFiles.ts # List files tool
│   │   ├── readRepomixOutput.ts # Read tool implementation
│   │   └── grepRepomixOutput.ts # Grep tool implementation
│   ├── fileManager.ts         # File management utility
│   ├── types.ts               # Shared type definitions
│   └── index.ts               # Main MCP server
├── scripts/
│   └── security-check.js      # Security scanner script
├── build/                     # Compiled output (auto-generated)
├── package.json               # Dependencies & scripts
├── tsconfig.json             # TypeScript config
└── README.md                 # Documentation

Dependencies

• @modelcontextprotocol/sdk - MCP SDK for creating MCP servers

How It Works

This MCP server provides secure file operations:

1. File Initialization: Accepts a list of Repomix files as command line arguments at startup
2. File Validation: Verifies file existence and readability during initialization
3. ID-based Access: Assigns numeric IDs to files for secure access without exposing file paths
4. File Reading: Secure file system access limited to pre-approved files
5. Pattern Matching: Uses JavaScript regex for pattern searching
6. Context Display: Shows surrounding lines for grep matches
7. Line Range Selection: Allows reading specific portions of files

Security Features

• Restricted File Access: Only files specified at startup can be accessed
• No Path Traversal: File access is limited to the pre-approved list
• Input Validation: File IDs are validated against the available file list
• Automated Security Scanning: Built-in tools to detect secrets and vulnerabilities
• Pre-publish Checks: Automatic security scanning before npm publish

Supported File Formats

Works with any Repomix output format:

• XML output files
• Markdown output files
• Plain text output files

Example Usage

Starting the MCP server

# Start with specific Repomix files
node build/index.js output1.xml output2.xml /path/to/output3.xml

Listing available files

{
  "tool": "list_repomix_files"
}

Reading a specific section

{
  "tool": "read_repomix_output",
  "fileId": "1",
  "startLine": 1,
  "endLine": 50
}

Searching for patterns

{
  "tool": "grep_repomix_output", 
  "fileId": "1",
  "pattern": "function\\s+\\w+",
  "flags": "gi",
  "contextLines": 2
}

License

MIT

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

Development Workflow

1. Fork and clone the repository
2. Create a feature branch: git checkout -b feature/amazing-feature
3. Make your changes and add tests if applicable
4. Run security checks: npm run security:check
5. Build the project: npm run build
6. Commit your changes: git commit -m "feat: add amazing feature"
7. Push to your branch: git push origin feature/amazing-feature
8. Open a Pull Request

The CI pipeline will automatically run tests and security checks on your PR.

Thanks

Special thanks to the Repomix team for creating an excellent tool for packaging codebases into AI-friendly formats. This MCP server extends Repomix's capabilities by providing used to structured access to Repomix output files.

Without Repomix's innovative approach to codebase analysis and packaging, this project wouldn't exist. We're grateful for their open-source contribution to the AI development community.