AVS Docs

소개

This powerful system facilitates efficient document retrieval by leveraging vector search. It intelligently ingests and chunks markdown documents, preserving their hierarchical structure, then generates high-quality contextual embeddings using Voyage AI's API. These documents and their corresponding embeddings are stored in MongoDB with parent-child relationships, enabling advanced semantic search. The tool offers a FastMCP server, allowing seamless integration with MCP clients for comprehensive document search capabilities, and supports flexible configuration of vector dimensions and chunking strategies.

주요 기능

• Ingests and chunks markdown documents with hierarchical headers
• Generates embeddings using Voyage AI's contextual embeddings API
• Stores documents and embeddings in MongoDB with parent-child relationships
• 1 GitHub stars
• Supports configurable vector dimensions and chunking strategies
• Provides a FastMCP server for semantic document search

사용 사례

• Performing semantic document search for information retrieval
• Integrating vector search capabilities into MCP clients like Claude Desktop
• Building knowledge bases for AI applications requiring contextual document access