kraken

Kraken Pro MCP for Gemini CLI exposes a set of trading and market data tools that wrap the Kraken Pro REST API for use inside the Gemini CLI. Once running, Gemini will automatically select the appropriate tool based on your natural-language prompt, enabling you to fetch market data, place orders, and manage your Kraken account from the terminal. Available trading tools include adding orders, canceling or amending orders, listing open orders, and setting a dead-man switch to cancel all orders if needed. Market data tools provide access to server time, system status, asset info, tradable pairs, tickers, OHLC data, order books, and recent trades or spreads. You can invoke these tools by prompting Gemini with natural language, for example asking for the current price of a pair or placing an order with specific parameters, and the CLI will route to the corresponding tool with the appropriate arguments.

Prerequisites:

• Gemini CLI configured on your system
• FastMCP CLI installed (pip install fastmcp-cli)
• Kraken Pro account with API keys

Installation steps:

1. Clone the repository: git clone https://github.com/oilst/kraken-mcp.git cd kraken-mcp

2. Prepare environment variables: Create a .env file (or export in your shell) with your Kraken API credentials:

   KRAKEN_API_KEY="your_public_key" KRAKEN_API_SECRET="your_private_base64_secret"

3. Install the MCP server via FastMCP: fastmcp install gemini-cli kraken-server.py

4. Run the server (example): python kraken-server.py or rely on FastMCP to manage the server as part of the Gemini CLI workflow

Note:

• Ensure kraken-server.py is present in the project root and accessible.
• The environment variables KRAKEN_API_KEY and KRAKEN_API_SECRET must be set for authentication with Kraken Pro.
• If you encounter permission or network issues, verify Python availability and network access to Kraken endpoints.

Tips and caveats:

• The tools mirror common Kraken Pro API actions; for exact parameter details, refer to the docstrings in kraken-server.py.
• Use validate=True for trading operations to test before placing real orders.
• Store your API keys securely; avoid committing .env files to version control.
• If running behind proxies or corporate networks, ensure outbound access to Kraken endpoints is allowed.
• The MCP server relies on the Gemini CLI to route natural language prompts to the appropriate tool; ensure Gemini is properly installed and configured.