insightfinder-mcp-server

insightfinder/insightfinder-mcp-server

3.3

If you are the rightful owner of insightfinder-mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

The InsightFinder MCP Server is a tool that enables Large Language Models (LLMs) to access incident data from the InsightFinder platform using the Model Context Protocol (MCP).

Tools
  1. fetch_incidents

    Enables LLMs to retrieve incident data based on parameters like systemName, startTime, etc.

InsightFinder MCP Server

This project provides a Model Context Protocol (MCP) server that allows Large Language Models (LLMs) to fetch incident data from the InsightFinder platform by acting as a tool.

Features

  • fetch_incidents Tool: Enables LLMs to retrieve incident data based on parameters like systemName, startTime, etc.
  • stdio Transport: Communicates with MCP clients via standard I/O.
  • Configuration via Environment Variables: Securely manage API credentials.

Setup

  1. Clone the repository:

    git clone <your-repo-url>
    cd insightfinder-mcp-server
    
  2. Create and activate a virtual environment:

    python -m venv .venv
    source .venv/bin/activate
    
  3. Install dependencies:

    pip install -e .
    
  4. Configure environment variables: Create a .env file in the project root and add your InsightFinder credentials:

    INSIGHTFINDER_API_URL="[https://app.insightfinder.com](https://app.insightfinder.com)"
    INSIGHTFINDER_JWT_TOKEN="your_jwt_token_here"
    INSIGHTFINDER_SYSTEM_NAME="system_name"
    INSIGHTFINDER_USER_NAME="user"
    

Running the Server

Using Python (Local Development)

Use the provided script to run the server:

./scripts/run_server.sh

Using Docker

You can run the MCP server using Docker without needing to install Python or dependencies locally:

docker run -i --rm \
  -e INSIGHTFINDER_API_URL=your_api_url \
  -e INSIGHTFINDER_JWT_TOKEN=your_jwt_token \
  -e INSIGHTFINDER_SYSTEM_NAME=your_system_name \
  -e INSIGHTFINDER_USER_NAME=your_user_name \
  docker.io/insightfinder/insightfinder-mcp-server:latest

Environment Variables:

  • INSIGHTFINDER_API_URL: Your InsightFinder API endpoint (e.g., https://app.insightfinder.com)
  • INSIGHTFINDER_JWT_TOKEN: Your JWT authentication token
  • INSIGHTFINDER_SYSTEM_NAME: The system name to query incidents for
  • INSIGHTFINDER_USER_NAME: Your InsightFinder username

MCP Client Configuration: When configuring your MCP client, use the following configuration:

{
  "insightfinder": {
    "command": "docker",
    "args": [
      "run",
      "-i",
      "--rm",
      "-e", "INSIGHTFINDER_API_URL=your_api_url",
      "-e", "INSIGHTFINDER_JWT_TOKEN=your_jwt_token",
      "-e", "INSIGHTFINDER_SYSTEM_NAME=your_system_name",
      "-e", "INSIGHTFINDER_USER_NAME=your_user_name",
      "docker.io/insightfinder/insightfinder-mcp-server:latest"
    ],
    "transport": "stdio",
  }
}

Replace the environment variable values with your actual InsightFinder credentials and configuration.