QuantConnect

QuantConnect is an MCP server that bridges large language models like Claude and OpenAI models to the QuantConnect cloud algorithmic trading platform. Quantitative developers, traders, and engineers use this server to automate strategy development directly through natural language instructions. By exposing over 60 API tools, it enables LLMs to inspect user accounts, manage cloud projects and file structures, compile source code, and trigger backtests. Users can also review detailed backtest metrics, charts, insights, and orders without leaving their MCP client environment. Additionally, the server supports parameter optimization workflows, Object Store management, syntax verification, and live algorithmic trading controls. Developers can initiate live deployments, stream execution logs, manage active portfolios, and send or broadcast commands directly to running strategies. The official implementation is distributed as a multi-platform Docker container compatible with Intel and ARM architectures.

Category: Data & Analytics

Tags: algorithmic-trading, backtesting, finance, market data, quantconnect

Visit QuantConnect

How to install and configure QuantConnect

1. Install and launch Docker Desktop on your machine. 2. Open Claude Desktop, then navigate to File > Settings > Developer > Edit Config to access claude_desktop_config.json. 3. Add the quantconnect server configuration under the mcpServers object: json { "mcpServers": { "quantconnect": { "command": "docker", "args": [ "run", "-i", "--rm", "-e", "QUANTCONNECT_USER_ID", "-e", "QUANTCONNECT_API_TOKEN", "-e", "AGENT_NAME", "--platform", "<your_platform>", "quantconnect/mcp-server" ], "env": { "QUANTCONNECT_USER_ID": "<your_user_id>", "QUANTCONNECT_API_TOKEN": "<your_api_token>", "AGENT_NAME": "MCP Server" } } } } 4. Replace <your_platform> with linux/amd64 or linux/arm64, and fill in your QuantConnect User ID and API Token. 5. Save the file and restart Claude Desktop.

What you can do with QuantConnect

• Creating and editing algorithmic trading code files within QuantConnect cloud projects using natural language instructions. - Triggering asynchronous cloud compilations and checking algorithms for initialization errors or Python PEP8 compliance issues. - Executing strategy backtests and retrieving execution charts, order lists, and quantitative insight metrics directly in chat. - Running parameter optimization jobs to fine-tune strategy inputs and reading execution results or aborting runs. - Deploying trading algorithms to live production environments, checking portfolio states, and stopping or liquidating active live deployments.

Key facts

• https://github.com/QuantConnect/mcp-server
• Data & Analytics, Finance, Crypto & Payments
• algorithmic-trading, backtesting, finance, market data, quantconnect

Part of MCP Servers

How do I install QuantConnect MCP Server?

You can run the official Docker image quantconnect/mcp-server using Docker Desktop. In your MCP client configuration, such as Claude Desktop's config file, add the docker run command configured with your QuantConnect User ID, API Token, and desired system platform architecture.

What can QuantConnect MCP Server do?

It provides 64 tools allowing AI assistants to manage cloud projects, update code files, compile trading scripts, execute backtests, manage parameter optimizations, upload files to the Object Store, check syntax, and deploy or liquidate live trading algorithms.

Which MCP clients work with QuantConnect?

It works with Claude Desktop and any other MCP-compliant client that supports standard input and output server executions via Docker. QuantConnect also maintains an embedded MCP integration directly inside their official VSCode extension.

Where do I get my QuantConnect API credentials?

You can obtain your User ID and API token directly from your QuantConnect Cloud account profile settings under the community profile section by requesting an API token.