QuantConnect Docs

QuantConnect Docs is an MCP server that provides intelligent search and retrieval capabilities across QuantConnect PDF documentation. It connects local PDF files to Model Context Protocol clients, converting algorithmic trading manuals and platform documentation into structured markdown files. Built for quantitative developers, algorithmic traders, and researchers who use the QuantConnect platform, it enables assistants like Claude to parse complex financial development material. The server indexes documents using an inverted index with TF-IDF scoring and proximity matching, ensuring precise excerpts are returned alongside page references. Users can look up documentation for local platform setup, algorithm initialization, universe selection, and custom indicators without leaving their chat interface. By maintaining an MD5-based file cache and persisting search indices to disk, the server avoids redundant processing on startup, offering efficient, context-grounded technical reference directly within AI workflows.

Category: Files, Documents & PDFs

Tags: algorithmic-trading, documentation, finance, pdf, quantconnect

Visit QuantConnect Docs

How to install and configure QuantConnect Docs

1. Clone the repository and navigate to the project directory. 2. Create and activate a Python virtual environment: bash python -m venv env source env/bin/activate 3. Install the required dependencies: bash pip install -r requirements.txt 4. Configure Claude Desktop by opening your configuration file (claude_desktop_config.json) and adding the server under mcpServers: json { "mcpServers": { "quantconnect-pdf-server": { "command": "/path/to/your/project/env/bin/python3", "args": ["/path/to/your/project/server.py"], "env": { "QUANTCONNECT_PDF_FOLDER": "/path/to/your/project/quantconnect-docs", "QUANTCONNECT_MARKDOWN_FOLDER": "/path/to/your/project/quantconnect-docs/markdown" } } } } 5. Replace the file paths with your actual project locations and restart Claude Desktop.

What you can do with QuantConnect Docs

• Searching QuantConnect documentation for specific Python API methods such as the OnData event handler or Initialize routine. * Retrieving detailed code examples and explanations for implementing custom technical indicators within trading strategies. * Looking up parameter definitions and execution behaviors across various available QuantConnect order types. * Comparing deployment workflows by searching guidance on local platform installations versus QuantConnect cloud backtesting. * Accessing exact page excerpts from algorithm design guides to reference universe selection methods directly in chat.

Key facts

• https://github.com/lhstorm/mcp_server_quantconnect_docs
• Files, Documents & PDFs, Finance, Crypto & Payments, Web Search & Research
• algorithmic-trading, documentation, finance, pdf, quantconnect

Part of MCP Servers

How do I install QuantConnect Docs?

To install the server, set up a Python 3.8 or higher virtual environment and install dependencies using pip install -r requirements.txt. Then configure Claude Desktop by referencing the Python executable and server.py path in your claude_desktop_config.json file alongside your PDF folder paths.

What can QuantConnect Docs do?

The server converts QuantConnect PDF files into markdown, builds an inverted TF-IDF search index, and provides tools to search terms, list documents, and extract full page contents with highlighted text matches.

Which tools are provided by QuantConnect Docs?

It provides three tools: search_quantconnect_docs for TF-IDF querying with excerpts, list_quantconnect_docs to view available processed files, and get_document_content to pull specific pages or complete markdown content.

Which MCP clients work with QuantConnect Docs?

It is configured for Claude Desktop via its standard JSON configuration file, but it also works with any MCP-compliant client capable of running local Python stdio processes.

How does QuantConnect Docs handle PDF processing?

It automatically converts PDFs into markdown files using PyPDF2, stores processing state via MD5 hashes to prevent duplicate work, and saves the search index to disk for fast startup times.