Crypto Whitepapers MCP Server

An MCP server serving as a structured knowledge base of crypto whitepapers for AI agents to access, analyze, and learn from.

Features

• Search Whitepapers: Use DuckDuckGo to find whitepaper PDFs for cryptocurrency projects.
• Load Whitepapers: Download and index whitepaper PDFs into the knowledge base.
• Query Knowledge Base: Query whitepaper content with optional project filtering.
• List Projects: View all projects available in the knowledge base.
• Claude Desktop Integration: Access tools and prompts via MCP in Claude Desktop.

Prerequisites

• Python 3.10+
• uv for dependency management and running
• Internet access.
• Claude Desktop for MCP integration (optional)

Setup

1. Clone the Repository:

   hljs language-bash

   Copy

   git clone https://github.com/kukapay/crypto-whitepapers-mcp.git
   cd crypto-whitepapers-mcp
   

2. Install Dependencies with uv:

   hljs language-bash

   Copy

   uv sync
   

3. Integrate with Claude Desktop (Optional):

   • Edit the Claude Desktop configuration file:
     • MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json
     • Windows: %APPDATA%\Claude\claude_desktop_config.json
   • Add the following configuration:
     hljs language-json

     Copy

     {
         "mcpServers": {
             "crypto-whitepapers": {
                 "command": "uv",
                 "args": [
                     "--directory",
                     "/absolute/path/to/crypto-whitepapers-mcp"   
                     "run",
                     "crypto-whitepapers-mcp"
                 ]
             }
         }
     }
     

     Replace /absolute/path/to/crypto-whitepapers-mcp with the absolute path to the repository.
   • Restart Claude Desktop and verify the server is loaded (look for the hammer icon in the input box).

Usage

Tools

The following tools are available via the MCP server:

• list_available_projects(): Lists all projects in the knowledge base (derived from PDF filenames).

  • Example: list_available_projects()
  • Returns: JSON list of project names.

• search_whitepaper(project_name: str): Searches for a project's whitepaper PDF using DuckDuckGo.

  • Example: search_whitepaper("bitcoin")
  • Returns: JSON list of up to 5 results with title, URL, and snippet.

• load_whitepaper(project_name: str, url: str): Downloads a whitepaper PDF from a URL and loads it into the knowledge base.

  • Example: load_whitepaper("bitcoin", "https://bitcoin.org/bitcoin.pdf")
  • Returns: Success or error message.

• ask_whitepapers(query: str, project_name: str = None): Searches the knowledge base for a query, optionally filtered by project.

  • Example: ask_whitepapers("blockchain technology", "bitcoin")
  • Returns: Up to 5 matching text snippets.

Prompts

• analyze_tokenomics(project_name: str): Analyzes tokenomics (distribution, supply, incentives) in a project's whitepaper using the ask_whitepapers tool.
  • Example: In Claude Desktop, run "Analyze the tokenomics of Ethereum."

Examples

1. List available projects:
   hljs language-arduino

   Copy

   List all available projects.
   

2. Search for a whitepaper:
   hljs language-rust

   Copy

   Search for the Bitcoin whitepaper PDF.
   

3. Load a whitepaper:
   hljs language-csharp

   Copy

   Load the Bitcoin whitepaper from https://bitcoin.org/bitcoin.pdf.
   

4. Query the knowledge base:
   hljs language-csharp

   Copy

   Ask the knowledge base about blockchain technology in the Bitcoin whitepaper.
   

License

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