PageIndex MCP for Developers

Connect PageIndex to your agent framework or LLM via MCP (Model Context Protocol) . Works with Claude Agent SDK, Vercel AI SDK, LangChain, OpenAI Agents SDK, and any MCP-compatible client.

PageIndex MCP uses API Key authentication for reliable deployment in production. It shares the same PageIndex API under the hood, so your existing API key, documents, and plan work seamlessly across both REST and MCP access.

Integration

1. Go to the PageIndex Developer Dashboard  to create an API Key.
2. Add to your MCP client configuration:

{
  "mcpServers": {
    "pageindex": {
      "type": "http",
      "url": "https://api.pageindex.ai/mcp",
      "headers": {
        "Authorization": "Bearer your_api_key"
      }
    }
  }
}

Cookbook: MCP Integration Examples

Step-by-step Jupyter notebooks showing how to connect PageIndex MCP with popular agent frameworks and LLM APIs. Each notebook walks through performing agentic, reasoning-based retrieval with PageIndex from an uploaded PDF.

See more notebooks and details in the MCP Integration Cookbook.

Open Source

The PageIndex MCP server is open-source. For more details, visit the PageIndex MCP GitHub repo .

Community & Support

• Star us on GitHub 
• Join our Discord 
• Leave us a message