Dune Analytics MCP Server

A mcp server that bridges Dune Analytics data to AI agents.

Features

• Tools:
  • get_latest_result: Fetch the latest results of a Dune query by ID.
  • run_query: Execute a Dune query by ID and retrieve results.
• CSV Output: All results are returned as CSV-formatted strings for easy processing.

Prerequisites

• Python 3.10+
• A valid Dune Analytics API key (get one from Dune Analytics)

Installation

Installing via Smithery

To install Dune Analytics for Claude Desktop automatically via Smithery:

npx -y @smithery/cli install @kukapay/dune-analytics-mcp --client claude

Manual Installation

1. Clone the Repository:

   bash

   Copy

   git clone https://github.com/kukapay/dune-analytics-mcp.git
   cd dune-analytics-mcp
   

2. Set Up Environment Variables: Create a .env file in the project root:

   Code

   Copy

   DUNE_API_KEY=your_api_key_here
   

   Alternatively, set it as a system environment variable:

   server.ts

   Copy

   export DUNE_API_KEY="your_api_key_here"
   

Usage

Running the Server

• Development Mode:

  bash

  Copy

  mcp dev main.py
  

  This starts the server with hot reloading for development.

• Install for Claude Desktop:

  bash

  Copy

  mcp install main.py --name "Dune Analytics"
  

  Installs the server as a service for use with Claude Desktop.

Tool Usage

1. get_latest_result(query_id)

   • Description: Retrieves the latest results of a specified Dune query.
   • Input: query_id (int) - The ID of the Dune query.
   • Output: CSV-formatted string of the query results.
   • Example:
     Code

     Copy

     get_latest_result(query_id=4853921)
     

2. run_query(query_id)

   • Description: Executes a Dune query and returns the results.
   • Input: query_id (int) - The ID of the Dune query to run.
   • Output: CSV-formatted string of the query results.
   • Example:
     Code

     Copy

     run_query(query_id=1215383)
     

Example Commands in Claude Desktop

• "Get latest results for dune query 1215383"
• "Run dune query 1215383"

License

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