Lichess MCP

The Lichess MCP server turns the popular online chess platform into a first‑class data source for AI assistants. By exposing Lichess’s REST API through the Model Context Protocol, it allows Claude (and any MCP‑compatible client) to query a user’s account, play games, analyze positions, and manage tournaments—all via natural‑language prompts. This eliminates the need for developers to write custom wrappers or handle OAuth flows manually, making it trivial to embed real‑time chess interactions into conversational agents.

At its core, the server provides a rich set of tools that mirror Lichess’s functionality. Developers can authenticate with an API token, then issue commands such as create_challenge, make_move, or get_cloud_eval. Each tool accepts plain parameters (e.g., a FEN string or a move in UCI/SAN) and returns structured JSON, which the AI can parse and present back to the user. This tight coupling means that a conversational agent can, for example, ask the user for a desired time control, automatically generate a challenge, and update the chat with the opponent’s response—all without leaving the dialogue.

Key capabilities include:

• Account management – fetch profile data, trophies, and user statistics.
• Game play – create challenges, make moves, retrieve ongoing games, and export PGNs.
• Position analysis – obtain cloud evaluations for arbitrary FENs or current game positions.
• Tournament handling – list, join, and create arena tournaments, as well as manage team participation.
• Real‑time interaction – the server supports streaming updates for ongoing games, allowing agents to keep users informed of opponent moves or draw offers.

Real‑world scenarios benefit from this integration in several ways. A chess coach could build a tutoring assistant that suggests improvements on the fly, while a game organizer could automate tournament registration and result tracking. Casual players might use an AI companion to challenge friends or explore opening theory without leaving their chat interface. Because the server translates natural language into precise API calls, developers can focus on crafting engaging conversational flows rather than plumbing the platform’s endpoints.

Unique advantages stem from its declarative design and robust error handling. The MCP specification ensures that every tool’s signature is self‑describing, enabling AI assistants to introspect capabilities and guide users with contextual help. Detailed error messages cover authentication failures, rate limits, and invalid inputs, allowing the assistant to surface clear feedback instead of cryptic API errors. Combined with seamless integration into Claude Desktop via Smithery, the Lichess MCP server offers a turnkey solution for embedding chess expertise into conversational AI workflows.