mcp-ragdocs
96
MCP-Ragdocs is a server that facilitates semantic search through documentation using a vector database like Qdrant. It supports adding and querying documents using natural language, providing flexibility and ease of retrieval from various documentation sources.
MCP-Ragdocs
A Model Context Protocol (MCP) server that enables semantic search and retrieval of documentation using a vector database (Qdrant). The server allows adding documentation from URLs or local files and searching through them using natural language queries.
Features
- Add documentation from URLs or local files
- Store documentation in a vector database for semantic search
- Search through documentation using natural language
- List all documentation sources
Requirements
- Node.js 16 or higher
- Qdrant (local or cloud)
- Embedding options: Ollama (default), OpenAI (optional)
Available Tools
- add_documentation: Add documentation from a URL
- search_documentation: Search stored documentation
- list_sources: List all documentation sources stored