GXtract MCP Server

Example Use Cases

• 1

  Search and retrieve relevant documents from a GroundX account programmatically within an editor workflow.

• 2

  Ask focused questions about the contents of a specific GroundX document and receive context-aware answers.

• 3

  Extract or explain semantic objects (diagrams, tables) from documents and manage cached GroundX metadata to reduce API usage.

Description

GXtract is a Model Context Protocol (MCP) server that integrates with the GroundX platform to provide document search, question-answering over documents, and semantic object explanations directly from editors like VS Code. It offers a set of tools (searchDocuments, queryDocument, explainSemanticObject) and cache management endpoints to reduce API calls and improve performance. The server is built with modern Python (3.12+), supports stdio and HTTP transport, and is designed for easy VS Code integration via MCP settings. It simplifies secure API key usage and local development with UV and Hatch tooling.