Doc Indexer

This tool serves as a local document indexer, implementing the Model Context Protocol (MCP) server for robust semantic search capabilities. Built with Rust, it efficiently parses and extracts content from a wide array of file formats, including PDF, Excel, SQL/PL-SQL, Markdown, and HTML. By leveraging Qdrant for vector storage and Voyage AI (or OpenAI-compatible) for embeddings, it transforms unstructured documents into searchable vectors, making it ideal for enhancing Retrieval-Augmented Generation (RAG) applications, integrating with AI code assistants like Claude Code CLI, and providing intelligent knowledge retrieval over diverse internal documentation.

Key Features

Use Cases