Torna MCP

🚀 Torna MCP Server

An MCP (Model Context Protocol) server for interacting with the Torna interface documentation management platform. It provides core tools based on the real Torna OpenAPI specification, allowing LLMs to manage API documentation in Torna through standardized interfaces.

🚀 Quick Start

Installation

Method 1: Install via PyPI (Recommended)

# Use pip
pip install torna-mcp

# Or use uv (Recommended)
uv pip install torna-mcp

Method 2: Install from Source Code

git clone https://github.com/li7hai26/torna-mcp.git
cd torna-mcp
pip install -e .
# Or use uv
uv pip install -e .

Configure Environment Variables

# Set the Torna server address
export TORNA_URL="http://localhost:7700/api"

# Set the module access token
export TORNA_TOKEN="your-module-token-here"

How to get the Token:

1. Log in to the Torna management console.
2. Select a project.
3. Select a module.
4. Click the OpenAPI tab.
5. Copy the token.

Environment variable file: If you use an environment variable file:

cp .env.example .env
# Edit the .env file and set your TORNA_URL and TORNA_TOKEN

Start the MCP Server

torna-mcp

After starting, the server will display connection information in the standard output, which you can configure in the MCP client for use.

✨ Features

Core API Interfaces (2 tools)

Based on the real Torna OpenAPI specification:

• Push Document (torna_push_document) - Push API documentation to Torna

  • Support for creating categories/folders
  • Support for defining request/response parameters
  • Support for error code configuration
  • Support for setting up debugging environments

• Get Document (torna_get_document) - Get detailed information about a single document

  • Retrieve the complete information of the document
  • Include request/response parameters
  • Include error code information

API Specification: Implemented based on Torna Official OpenAPI

🛠️ MCP Client Configuration

Cursor

1. Open the Cursor settings.
2. Find the MCP Servers configuration.
3. Add a new server:

{
  "mcpServers": {
    "torna-mcp": {
      "command": "torna-mcp",
      "env": {
        "TORNA_URL": "http://localhost:7700/api",
        "TORNA_TOKEN": "your-module-token-here"
      }
    }
  }
}

Claude Desktop

1. Edit the Claude configuration file.
2. Add the MCP server configuration:

{
  "mcpServers": {
    "torna-mcp": {
      "command": "torna-mcp"
    }
  }
}

3. Restart Claude Desktop.

IFlow CLI

# Add to the MCP configuration
iflow mcp add torna-mcp

For detailed client configuration instructions, please refer to MCP_CLIENTS.md.

💻 Usage Examples

Push API Documentation

Tool: torna_push_document
Parameters:
{
  "name": "User Login",
  "description": "User login interface",
  "url": "/api/auth/login",
  "http_method": "POST",
  "content_type": "application/json",
  "request_params": [
    {
      "name": "username",
      "type": "string",
      "description": "Username",
      "required": true,
      "example": "john_doe"
    },
    {
      "name": "password", 
      "type": "string",
      "description": "Password",
      "required": true,
      "example": "123456"
    }
  ],
  "response_params": [
    {
      "name": "token",
      "type": "string",
      "description": "Access token"
    },
    {
      "name": "userId",
      "type": "string", 
      "description": "User ID"
    }
  ],
  "author": "Zhang San"
}

Get Document Details

Tool: torna_get_document
Parameters:
{
  "doc_id": "doc_123"
}

Create a Category (Folder)

Tool: torna_push_document
Parameters:
{
  "name": "User Management",
  "description": "User-related API interfaces",
  "url": "",
  "http_method": "GET",
  "is_folder": true
}

Document with Debugging Environment

Tool: torna_push_document
Parameters:
{
  "name": "Product Query",
  "description": "Query product information",
  "url": "/api/products/{id}",
  "http_method": "GET",
  "content_type": "application/json",
  "path_params": [
    {
      "name": "id",
      "type": "int",
      "description": "Product ID",
      "required": true,
      "example": "123"
    }
  ],
  "debug_env_name": "Test Environment",
  "debug_env_url": "http://localhost:8080"
}

🔧 Technical Details

System Requirements

• Python: 3.8 or higher
• Torna: Privately deployed version
• MCP Clients: Cursor, Claude Desktop, VS Code, etc.

Install the Python Environment

• pip (Standard Python installation)
• uv (Recommended - Faster and more modern package manager)

# Install uv
curl -LsSf https://astral.sh/uv/install.sh | sh

📚 Documentation

• Quick Start - Detailed quick start guide
• Installation Guide - Multiple installation methods
• MCP Client Configuration - Configuration methods for various MCP clients
• Deployment Guide - Instructions for deploying in a production environment

🐛 Issue Feedback

If you encounter any issues during use, please:

1. Check GitHub Issues.
2. Submit feedback on the PyPI page.
3. Send an email to: li7hai26@gmail.com

📄 License

This project is licensed under the MIT License. For details, please refer to the LICENSE file.

👨‍💻 Developers

• Author: Aladdin's Lamp
• Email: li7hai26@gmail.com
• GitHub: @li7hai26

🔗 Related Links

• PyPI Package
• GitHub Repository
• Torna Project