PageIndex MCP

v1.6.2Knowledge & Memorystable

Enables LLMs to chat with long PDFs using a reasoning-based, tree-structured document index that navigates content like a human would, without requiring vector databases or hitting context limits.

pageindex-mcpmcpai-integration
Share:
345
Stars
0
Downloads
0
Weekly
0/5

What is PageIndex MCP?

PageIndex MCP is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to enables llms to chat with long pdfs using a reasoning-based, tree-structured document index that navigates content like a human would, without requiring vector databases or hitting context limits.

Enables LLMs to chat with long PDFs using a reasoning-based, tree-structured document index that navigates content like a human would, without requiring vector databases or hitting context limits.

This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Enables LLMs to chat with long PDFs using a reasoning-based,

Use Cases

Chat with long PDFs without context limits.
Navigate documents with reasoning index.
Perform tree-structured document search.
VectifyAI

Maintainer

LicenseMIT License
Languagetypescript
Versionv1.6.2
UpdatedMay 20, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

NPM

npx -y pageindex-mcp

Manual Installation

npx -y pageindex-mcp

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use PageIndex MCP

PageIndex MCP enables AI assistants to chat with long PDFs by building a reasoning-based, tree-structured document index that navigates content the way a human would — section by section. Unlike vector-database approaches, it preserves full document context without chunking, eliminating context-limit errors on large documents. Developers can connect it via a remote HTTP endpoint using an API key from the PageIndex dashboard, or run it locally with npx. It integrates with Claude, OpenAI Agents SDK, Vercel AI SDK, and LangChain.

Prerequisites

  • Node.js 18.0.0 or higher installed
  • A PageIndex account and API key from https://dash.pageindex.ai/api-keys
  • An MCP-compatible client such as Claude Desktop or Cursor
  • The PDF files you want to query (local or online URLs)
1

Create a PageIndex account and get an API key

Sign up at pageindex.ai and navigate to the API Keys section of the dashboard at https://dash.pageindex.ai/api-keys. Generate a new API key and copy it for use in your MCP configuration.

2

Choose your connection method

PageIndex MCP can run as a remote HTTP MCP server (recommended, no local install) by pointing your client to https://api.pageindex.ai/mcp with your Bearer token, or as a local server via npx.

npx -y @pageindex/mcp
3

Configure your MCP client for the remote API

Add the PageIndex MCP server to your MCP client configuration file using the HTTP transport with your API key as a Bearer authorization header.

{
  "mcpServers": {
    "pageindex": {
      "type": "http",
      "url": "https://api.pageindex.ai/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY_HERE"
      }
    }
  }
}
4

Alternatively, configure for local npx usage

If you prefer to run the server locally, use the stdio transport in your Claude Desktop config file (claude_desktop_config.json).

{
  "mcpServers": {
    "pageindex": {
      "command": "npx",
      "args": ["-y", "@pageindex/mcp"],
      "env": {
        "PAGEINDEX_API_KEY": "YOUR_API_KEY_HERE"
      }
    }
  }
}
5

Restart your MCP client and verify the connection

Restart Claude Desktop or your chosen MCP client. The PageIndex tools should appear in the tools list. You can verify by asking your AI assistant to list available tools.

PageIndex MCP Examples

Client configuration (Claude Desktop)

Full claude_desktop_config.json entry for PageIndex MCP using the remote HTTP endpoint.

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

Prompts to try

Example prompts that work well once PageIndex MCP is connected to your AI assistant.

- "Index this PDF and summarize the key findings: https://example.com/report.pdf"
- "What does section 3.2 say about data retention in this contract?"
- "Find all mentions of risk factors in this 200-page annual report"
- "Compare the methodology sections of these two research papers"

Troubleshooting PageIndex MCP

Authorization error when connecting to the remote API

Double-check that your API key is copied correctly from https://dash.pageindex.ai/api-keys and is placed in the Authorization header as 'Bearer YOUR_KEY' with no extra spaces or quotes.

npx command fails with Node.js version error

PageIndex MCP requires Node.js 18 or higher. Run 'node --version' to check your version and upgrade via https://nodejs.org or using nvm.

Large PDFs time out or return partial results

For very large documents, the tree-structured indexing may take extra time on first load. Try splitting your question into more specific queries targeting particular sections rather than asking for a full document summary.

Frequently Asked Questions about PageIndex MCP

What is PageIndex MCP?

PageIndex MCP is a Model Context Protocol (MCP) server that enables llms to chat with long pdfs using a reasoning-based, tree-structured document index that navigates content like a human would, without requiring vector databases or hitting context limits. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install PageIndex MCP?

Install via npm with the command: npx -y pageindex-mcp. Then add the server configuration to your AI client's JSON config file (e.g., claude_desktop_config.json or .cursor/mcp.json).

Which AI clients work with PageIndex MCP?

PageIndex MCP works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is PageIndex MCP free to use?

Yes, PageIndex MCP is open source and available under the MIT License license. You can use it freely in both personal and commercial projects.

Browse More Knowledge & Memory MCP Servers

Explore all knowledge & memory servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "pageindex-mcp": { "command": "npx", "args": ["-y", "pageindex-mcp"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use PageIndex MCP?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides