MCP Server Overview
Connect IndexMind to your IDE with the Model Context Protocol server.
The Model Context Protocol (MCP) is an open standard that lets AI-powered coding tools communicate with external services. IndexMind ships an MCP server that exposes your project data, including recommendations, visibility queries, sentiment analyses, competitor narratives, and agent actions, directly inside your IDE.
Instead of switching between your editor and the IndexMind dashboard, you can browse recommendations, review visibility data, and approve agent actions without leaving your development environment.
What you can do
- Browse recommendations: List open, in-progress, and resolved recommendations for any project, then drill into full fix guidance with code samples.
- View visibility data: Pull AI visibility queries and copy-ready fixes into your editor context.
- Review sentiment and narratives: Access sentiment analysis results and competitor narrative actions alongside your code.
- Approve agent actions: Review and approve pending agent actions from your IDE, keeping your workflow uninterrupted.
- Update statuses: Mark recommendations and narrative actions as done without opening a browser.
Prerequisites
Before setting up the MCP server, make sure you have:
- Node.js 20+ installed on your machine
- An IndexMind account on the Starter plan or above
- An org-scoped API key created in your project settings
MCP keys are created in Project Settings > MCP. Each key is scoped to your organization and grants access to all projects within it.
Quick start
Getting the MCP server running takes three steps:
- Install the server: Clone the repo, install dependencies, and build the server.
- Configure your IDE: Add the MCP server to Cursor, VS Code, or Claude Desktop.
- Set environment variables: Provide your API key, endpoint, and project ID.
Once connected, your IDE will discover all 15 MCP tools automatically. See the Tools Reference for a full list of available tools, and Health Checks to verify everything is working.