Cisco Catalyst Center MCP Server

MCP Server for Cisco Catalyst Center APIs

This project implements a Model Context Protocol (MCP) server that wraps the Cisco Catalyst Center APIs defined in a Swagger JSON file. The MCP server allows you to use these APIs as tools within AI agents built with LangChain or any other framework supporting the Model Context Protocol.

Features

• Automatically parses Swagger/OpenAPI JSON to extract API endpoints as tools
• Implements the Model Context Protocol for tool discovery and execution
• Provides a FastAPI server to expose the tools via MCP
• Includes authentication with the Cisco Catalyst Center API
• Example client to demonstrate using the MCP server with LangChain

Prerequisites

• Python 3.9 or higher
• Cisco Catalyst Center with API access
• OpenAI API key

Setup

1. Clone the repository and navigate to the project directory:

2. Install the required dependencies:

3. Configure the environment variables:

Edit the .env file with your configuration details:

Running the MCP Server

Option 1: Running Directly

Start the MCP server directly with:

Option 2: Running with Docker (Recommended)

The server can also be run as a Docker container, which ensures consistent environments and easier deployment.

1. Make sure Docker and Docker Compose are installed on your system.
2. Run the Docker container:

This script will build the Docker image and start a container. The server will be available at http://localhost:8000.

API Endpoints

• GET /: Health check endpoint
• POST /mcp/tools: Get available tools in MCP format
• POST /mcp/execute_tool/{tool_name}: Execute a specific tool with parameters

Using with LangChain

The project includes an example client (example_client.py) that demonstrates how to use the MCP server with LangChain.

Run the example client with:

The client will fetch the available tools from the MCP server and create a LangChain agent that can use these tools.

Project Structure

• server.py: Main FastAPI server implementing the MCP
• swagger_parser.py: Parser for Swagger/OpenAPI JSON files
• api_client.py: Client for making requests to Cisco APIs
• tool_handler.py: Handler for MCP tools that wrap API endpoints
• custom_tools.py: Implementation of custom tools that combine multiple API calls
• config.py: Configuration module
• example_client.py: Example client using LangChain
• Dockerfile: Configuration for building the Docker image
• docker-compose.yml: Docker Compose configuration for services
• run_docker.sh: Script to build and run the Docker container

MCP Integration

This server implements the Model Context Protocol (MCP) for AI agents. The MCP defines a standardized way for AI models to discover and use tools.

The key components of MCP integration are:

1. Tool Discovery: The /mcp/tools endpoint returns tool schemas in the MCP format.
2. Tool Execution: The /mcp/execute_tool/{tool_name} endpoint executes tools and returns results.

Customization

You can customize the MCP server by:

1. Modifying the SwaggerParser to handle different Swagger/OpenAPI formats
2. Extending the ApiClient to add authentication methods or error handling
3. Adding new endpoints to the FastAPI server
4. Creating custom tools in the custom_tools.py file that combine multiple API calls for advanced functionality

Custom Tools

The server includes a custom_tools.py module that demonstrates how to create custom tools that combine multiple API calls. These tools can provide higher-level functionality that is not directly exposed by the API.

Currently implemented custom tools:

• network_health_report: Generates a comprehensive network health report by combining data from multiple API endpoints

Troubleshooting

• If authentication fails, check your Cisco API credentials in the .env file
• If tool execution fails, check the logs for more detailed error messages
• If parsing fails, ensure your Swagger JSON is valid and follows the OpenAPI 3.0.x format

Docker-specific troubleshooting

• If the Docker container fails to start, check Docker logs: docker-compose logs -f
• If the container can't access the Swagger file, ensure the context path in docker-compose.yml is correct
• To rebuild the Docker image completely: docker-compose down && docker-compose build --no-cache && docker-compose up -d
• To enter the running container for debugging: docker exec -it mcp-server bash

License

MIT