QuantConnect

QuantConnect Algorithmic Trading Platform Orchestration MCP - Agentic LLM Driven Trading Strategy Design, Research & Implementation

python

GitHub Stars

python

Language

6 months ago

First Indexed

2 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": {
    "taylorwilsdon-quantconnect-mcp": {
      "command": "uvx",
      "args": [
        "quantconnect-mcp"
      ],
      "env": {
        "QUANTCONNECT_USER_ID": "YOUR_USER_ID",
        "QUANTCONNECT_API_TOKEN": "YOUR_API_TOKEN",
        "QUANTCONNECT_ORGANIZATION_ID": "YOUR_ORG_ID"
      }
    }
  }
}

You can run and integrate QuantConnect’s MCP server locally to automate project management, backtesting, live trading, and advanced analytics. It’s designed for fast, secure, and asynchronous processing, so you can drive QuantConnect workflows from your own AI or automation clients.

How to use

You use an MCP client to communicate with the QuantConnect MCP Server. Start a local or connected MCP instance, authenticate with your credentials, and then issue commands or natural language requests to manage projects, run backtests, deploy live strategies, access historical data, and run analytics. The server supports end-to-end workflows from project creation to backtesting and live deployment, all through an MCP client that speaks the MCP protocol.

How to install

Prerequisites: You need a working Python environment and the MCP client runtime. You also need your QuantConnect credentials (User ID and API Token) ready for authentication.

# Install and run directly from the MCP distribution using the recommended runtime
uvx quantconnect-mcp

# Or install the package with uvx/pip and then run
uv pip install quantconnect-mcp
pip install quantconnect-mcp

Authentication

Credentials are required for the MCP server to function. You must provide a user ID and an API token, with an optional organization ID for scoped access.

export QUANTCONNECT_USER_ID="your_user_id"        # Required
export QUANTCONNECT_API_TOKEN="your_api_token"    # Required
export QUANTCONNECT_ORGANIZATION_ID="your_org_id" # Optional

Environment and startup

Configure the server to use your credentials and start the MCP runtime. The recommended default transport is STDIO for MCP clients that run locally.

# Start using the default STDIO transport
uvx quantconnect-mcp

# Optional HTTP transport (if you configure a remote endpoint in your setup)
MCP_TRANSPORT=streamable-http MCP_PORT=8000 uvx quantconnect-mcp

Natural language interaction

After you start the server, interact with it through an MCP-compatible AI client. Use natural language to describe the workflow you want, such as building a global portfolio, performing PCA analysis, or deploying a live algorithm.

Additional notes

You can adjust transport, port, and host settings via environment variables as needed. Ensure you keep credentials secure and rotate tokens periodically. The server is designed for asynchronous, high-performance processing to support complex financial workflows.

Testing

Run the test suite to verify authentication, server core, and tool integrations. Use the provided test commands to ensure all components work together before production use.

Contributing

Contribute by adding features, improving tests, and maintaining type safety. Follow the project’s guidelines for development, testing, and code quality to keep the MCP server reliable and secure.

License

This project is licensed under the MIT License. See the license file for details.

Built by

VeilStrat

AI signals for GTM teams