chess

This Chess MCP server provides standardized MCP interfaces to access Chess.com public data such as player profiles, statistics, game records, and more. It exposes a suite of tools categorized under Player Information, Games, and Clubs, allowing AI assistants to search, retrieve, and analyze information about players, their games, and club affiliations without requiring authentication. You can enable and configure the available tools via the MCP endpoints exposed by the server, and then integrate these endpoints into your AI workflows to fetch current data or historical records from Chess.com.

To use the server, deploy it (Docker is the recommended option) and point your MCP client configuration to the running instance. The available tools include getting a player's profile and stats, checking online status, listing titled players, retrieving a player's ongoing games or monthly archives, and downloading PGN data for a given month, as well as retrieving club information and members. The tooling is designed to be discoverable through the MCP framework, so your assistant can query by tool names such as get_player_profile, get_player_stats, is_player_online, get_titled_players, get_player_current_games, get_player_games_by_month, get_player_game_archives, download_player_games_pgn, get_club_profile, and get_club_members.

When integrating, ensure your MCP client is configured with the appropriate server name (e.g., chess) and that your tooling requests are shaped according to MCP conventions. Since the server uses Chess.com’s public API, there is no authentication required for access, but rate limits of the upstream API may apply.

Prerequisites

• Docker (recommended) or an alternative MCP transport compatible with your environment
• Access to the internet to pull the Docker image pab1it0/chess-mcp

Installation steps (Docker)

1. Install Docker on your system: https://docs.docker.com/get-started/
2. Pull and run the Chess MCP container:

# Run the MCP container (detached)
docker run --rm -d -p 8000:8000 pab1it0/chess-mcp

3. Verify the container is running and exposing the MCP endpoints at the mapped port (e.g., http://localhost:8000).
4. Update your Claude Desktop or MCP client configuration to point to the running server (see usage notes below).

Alternative – UV (if you prefer running the Python MCP server directly):

1. Install UV and set up a Python virtual environment as per the project instructions.
2. Run the MCP with UV by pointing to the server entrypoint, for example using the provided main.py location in the repository:

# Example (adjust path to your checkout):
uv venv
source .venv/bin/activate  # Unix/macOS
.venv\Scripts\activate     # Windows
uv pip install -e .
uv run src/chess_mcp/main.py

3. Configure your MCP client to use command: uv with the appropriate arguments, as described in the README.

Tips and caveats:

• The MCP uses Chess.com's public API; no authentication is required, but be mindful of upstream rate limits.
• If you deploy via Docker, ensure the container can be reached by your MCP client (adjust ports as needed).
• When using UV, specify the full path to the chess-mcp directory in your client configuration and set NO_UV=1 if you encounter spawn issues.
• Available tools are organized under Player Information, Games, and Clubs. If a tool is not needed, you can omit it from your client configuration to reduce overhead.
• You can customize which tools are exposed via MCP configuration if the container image or deployment supports feature flags or environment overrides.
• For development, install the project’s dev dependencies and run the test suite to validate changes before deployment.