Point this server at a folder of markdown notes and it becomes searchable by meaning, not just by keyword — from your AI client, or from your shell.
Knowledge Base MCP Server is an MCP project maintained in the jeanibarz/knowledge-base-mcp-server repository. The maintainer describes its purpose as “Point this server at a folder of markdown notes and it becomes searchable by meaning, not just by keyword — from your AI client, or…”. Read the project documentation for the complete scope.
Project documentation · Repository · Discovery listing. This profile records the public documentation checked on 2026-10-08.
Check the linked publisher documentation and your connected client's tool list. Names and availability can vary with the server version and account permissions.
| Tool | What it does |
|---|---|
| list_knowledge_bases | Discover which shelves exist, so the agent can scope a later search instead of guessing a name. |
| retrieve_knowledge | The core semantic search — returns the passages most relevant to a query so an agent can ground its answer in your documents rather than in what the model… |
| ask_knowledge | Retrieve, then have a configured local or OpenAI-compatible LLM write an answer with citations — one call instead of retrieve-then-prompt. |
| list_models | List the registered embedding models and which one is active. |
| kb_stats | Read-only corpus, index, model, cache, and transport statistics — useful for a health check without touching the index. |
| diff_index | Compare retrieval results across two persisted index versions, to see what a rebuild actually changed. |
| add_document | Write a text document into a KB through the guarded mutation path. |
| delete_document | Delete a KB-relative document through the guarded mutation path. |
| reindex_knowledge_base | Force a global rebuild, optionally validating a named KB first. |
Store entities, relations and observations in a local knowledge graph with the reference Memory MCP server.
A structured scratchpad for hard problems — stepwise reasoning with revisions, branches and visible logic.
Academic research search across PubMed and arXiv
Agent memory with git-like version control. Custom LLMs turn conversations into structured facts with automatic conflict detection - your agent sees
Interact with StarRocks
Cross-tool persistent memory and context for AI assistants over MCP.