RAG Anything

关于

This FastAPI application serves as a comprehensive wrapper for the RAG-Anything library, offering both a REST API and an MCP server for seamless integration with tools like Claude Desktop. Engineered with a hexagonal architecture for exceptional maintainability and testability, it specializes in multi-modal document processing, allowing users to index diverse file types such as PDFs, DOCX, and images. With advanced query controls and a PostgreSQL backend for vector and graph storage, it enables granular context retrieval, making it a powerful solution for building intelligent RAG pipelines.

主要功能

• Batch folder indexing with recursive directory support
• MCP server for Claude Desktop integration
• PostgreSQL backend with vector and graph storage
• 0 GitHub stars
• Granular context retrieval (chunks, entities, relationships)
• Multi-modal document processing (PDF, DOCX, PPTX, Images, etc.)

使用案例

• Integrating RAG capabilities into applications via a REST API
• Retrieval-only RAG for building custom prompts or integrating with external LLMs
• Enhancing Claude Desktop with document indexing and query functionality