What it does
The server provides hybrid retrieval, combining vector embeddings with graph traversal, and its results are citation-aware. It exposes tools for indexing documents, running RAG queries and adding edges to the knowledge graph.
It suits assistants that need to answer from indexed documents and follow relationships between the items in a knowledge graph, not only from vector similarity. Example prompts in the documentation call index_document, rag_query and add_graph_edge.
It can be installed with pip and started with the rag-knowledge-graph-mcp command, which uses stdio transport. It is added to an MCP client configuration, such as the claude_desktop_config.json file used by Claude Desktop. The project is released under the MIT licence.
Limitations and requirements
Requires an MCP-compatible client, such as Claude Desktop, or a configuration file for another MCP client. Installation requires pip, and the server runs over stdio transport.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 13 Nov 2026 · its page on the source
io.github.CSOAI-ORG/rag-knowledge-graph-mcp
Rag Knowledge Graph MCP Server by MEOK AI Labs
v1.0.8As shown on registry.modelcontextprotocol.io when we last read it