\n\n# PageIndex MCP\n\n[](https://pageindex.ai/mcp) [](https://vectify.ai/pageindex)\n\n[**PageIndex**](https://github.com/VectifyAI/PageIndex) is a vectorless, reasoning-based RAG system that represents documents as hierarchical **tree structures**. It enables LLMs to navigate and retrieve information through structure and **reasoning**, not vector similarity — much like a human would retrieve information using a book's index.\n\n[**PageIndex MCP**](https://pageindex.ai/mcp) exposes this **LLM-native tree index** directly to LLMs via MCP, allowing platforms like **Claude**, **Cursor**, and other MCP-compatible agents or LLMs to reason over document structure and retrieve the right information — without vector databases.\n\nWant to chat with long PDFs but hit context limit reached errors? Add your file to PageIndex to seamlessly chat with long PDFs on any agent/LLM platforms.\n\n✨ Chat to long PDFs the **human-like, reasoning-based way** ✨\n\n- Support local and online PDFs \n- Free 1000 pages \n- Unlimited conversations \n\nFor more information about PageIndex MCP, check out the [PageIndex MCP](https://pageindex.ai/mcp) project page.\n\n
\n\nPageIndex is a vectorless, **reasoning-based RAG** system that generates hierarchical **tree structures** of documents and uses multi-step **reasoning** and tree search to retrieve information like a human expert would. It has the following key properties:\n\n- **Higher Accuracy**: Relevance beyond similarity\n- **Better Transparency**: Clear reasoning trajectory with traceable search paths\n- **Like A Human**: Retrieve information like a human expert navigates documents\n- **No Vector DB**: No extra infrastructure overhead\n- **No Chunking**: Preserve full document context and structure\n- **No Top-K**: Retrieve all relevant passages automatically\n\n\n---\n# PageIndex MCP Setup \nSee [PageIndex MCP](https://pageindex.ai/mcp) for full video guidances.\n\n### 1. For Claude Desktop (Recommended)\n\n**One-Click Installation with Desktop Extension (MCPB):**\n\n1. Download the latest `.mcpb` file from [Releases](https://github.com/VectifyAI/pageindex-mcp/releases)\n2. Double-click the `.mcpb` file to install automatically in Claude Desktop\n3. The OAuth authentication will be handled automatically when you first use the extension\n\n> **Note**: Claude Desktop Extensions now use the `.mcpb` (MCP Bundle) file extension. Existing `.dxt` extensions will continue to work, but we recommend using `.mcpb` for new installations.\n\nThis is the easiest way to get started with PageIndex's reasoning-based RAG capabilities.\n\n### 2. For Other MCP-Compatible Clients\n\n#### Option 1: Local MCP Server (with local PDF upload)\n\n**Requirements:** Node.js ≥18.0.0\n\nAdd to your MCP configuration:\n\n```json\n{\n \"mcpServers\": {\n \"pageindex\": {\n \"command\": \"npx\",\n \"args\": [\"-y\", \"pageindex-mcp\"]\n }\n }\n}\n```\n\n> **Note**: This local server provides full PDF upload capabilities and handles all authentication automatically.\n\n#### Option 2: Direct Connection to PageIndex\n\nConnect directly to the PageIndex OAuth-enabled MCP server:\n\n```json\n{\n \"mcpServers\": {\n \"pageindex\": {\n \"type\": \"http\",\n \"url\": \"https://chat.pageindex.ai/mcp\"\n }\n }\n}\n```\n\n\n**For clients that don't support HTTP MCP servers:**\n\nIf your MCP client doesn't support HTTP servers directly, you can use [mcp-remote](https://github.com/geelen/mcp-remote) as a bridge:\n\n```json\n{\n \"mcpServers\": {\n \"pageindex\": {\n \"command\": \"npx\",\n \"args\": [\"-y\", \"mcp-remote\", \"https://chat.pageindex.ai/mcp\"]\n }\n }\n}\n```\n\n\n\n> **Note**: Option 1 provides local PDF upload capabilities, while Option 2 only supports PDF processing via URLs (no local file uploads).\n\n# Related Links\n\n[](https://vectify.ai/pageindex) \n[](https://github.com/VectifyAI/PageIndex)\n\n## License\n\nThis project is licensed under the terms of the MIT open source license. Please refer to [MIT](./LICENSE) for the full terms.\n","readmeFilename":"README.md"}
![](\"https://docs.pageindex.ai/images/general/mcp_banner.jpg\")
\n \n![](\"https://docs.pageindex.ai/images/cookbook/vectorless-rag.png\")
\n \n