Reconnaissance Blind Chess

Provides an MCP server to play RBC against bots, manage game state, and submit senses and moves.
  • python

0

GitHub Stars

python

Language

7 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "firekickerton-rbc-mcp-server": {
      "command": "python",
      "args": [
        "rbc_mcp_server.py"
      ]
    }
  }
}

You run a Reconnaissance Blind Chess (RBC) MCP Server to let LLM agents play RBC games remotely, track ongoing games, view board states, and submit senses and moves. This server provides practical, programmable ways to engage with RBC bots or participate in ranked play, all through MCP-compatible clients.

How to use

You connect to the RBC MCP Server using an MCP client and follow practical workflows to start unranked games against specific bots, participate in ranked games, and manage your gameState and move history. You can start unranked games against bots like random, attacker, trout, Oracle, and StrangeFish2, choose colors (white, black, or random), and then submit sensory information and moves when it’s your turn. You can also listen for ranked invitations, accept them automatically, and monitor your active games.

How to install

Prerequisites: You need Python 3.8 or higher and an account on RBC. Ensure you have network access to RBC so invitations and game data can flow between the server and RBC.

Step 1: Prepare your project directory and environment.

# Create or choose a work directory
mkdir -p ~/projects/rbc_mcp_server
cd ~/projects/rbc_mcp_server

# Create a Python virtual environment
python3 -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

Step 2: Install dependencies.

pip install -r requirements.txt

Step 3: Run the MCP server.

python rbc_mcp_server.py

Step 4: (Optional) Configure your MCP client integration. The server exposes a standard MCP interface for clients to communicate with RBC. You can configure your MCP client to run the server command and point to the local Python process.

Configuration and usage notes

A practical example for integrating with a client like Claude Desktop shows how to register the server in your MCP client config. Use the following as a configuration entry.

{
  "mcpServers": {
    "rbc": {
      "command": "python",
      "args": ["rbc_mcp_server.py"],
      "env": {}
    }
  }
}

Available tools

start_unranked_game

Start an unranked RBC game against a specific bot by sending an invitation to the RBC server, choosing the opponent bot, color, and providing your RBC credentials.

start_ranked_game

Start listening for ranked RBC game invitations. The server accepts invitations automatically and assigns colors as provided by the RBC system.

stop_ranked_listener

Stop listening for new ranked game invitations; ongoing games continue.

get_game_state

Retrieve the current state for a specific game, including board state tracking, turn, and results.

get_board_ascii

Fetch an ASCII representation of the current board and related game information for a given game.

submit_sense

Submit a sense action for the current turn to reveal a 3x3 area around a chosen square.

submit_move

Submit a move for the current turn in UCI format, or indicate a skip with "pass".

list_active_games

List all currently active RBC games associated with your account.

get_game_history

Obtain the detailed move-by-move history for a completed or ongoing game.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational
Reconnaissance Blind Chess MCP Server - firekickerton/rbc-mcp-server | VeilStrat