TSP MCP Server

What is TSP MCP Server?

TSP MCP Server is a Model Context Protocol server for solving the Traveling Salesman Problem (TSP). It provides optimized algorithms to find the shortest route through a set of cities and integrates with Claude Desktop for natural language interactions.

How to use TSP MCP Server?

Install via npm or GitHub, then configure Claude Desktop by editing its claude_desktop_config.json to point to the server’s built dist/index.js. After restarting Claude Desktop, you can ask natural language questions like “Solve TSP for cities at (0,0), (1,1), (2,0).”

Key features of TSP MCP Server

• Multiple algorithms: Dynamic Programming (≤10 cities) and Nearest Neighbor + 2‑opt.
• Supports both coordinate-only and named city inputs.
• Calculates total distance for custom routes.
• Tests multiple starting points to find better solutions.
• Generates high‑quality PNG visualizations of routes.
• Customizable visualization themes (modern, dark, minimal, colorful).

Use cases of TSP MCP Server

• Find the shortest delivery route for a set of logistics points.
• Optimize travel itineraries for field service or sales visits.
• Visualize and compare TSP solutions for educational demonstrations.

FAQ from TSP MCP Server

What algorithms does TSP MCP Server use?

Dynamic Programming (Held‑Karp) for ≤10 cities (guaranteed optimal), and Nearest Neighbor + 2‑opt heuristic for larger instances (high‑quality solutions).

What are the runtime dependencies?

Node.js version 18 or higher and npm or yarn are required. The server is designed for use with the Claude Desktop application.

Where does data live?

All data (city coordinates, routes) is processed locally – the server runs on your machine and does not send data externally.

How is the server invoked?

It uses the Model Context Protocol’s tool‑based transport. Communication happens via JSON‑RPC over standard I/O.

Is authentication required?

No authentication or API keys are mentioned in the README.