Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@metamind-vault-rag search my vault for notes about neural networks"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A retrieval engine for a directory of markdown. It watches files, indexes them incrementally, and answers hybrid search queries over them.
Vectors live in sqlite-vec, keywords in SQLite FTS5, and the two are fused with reciprocal rank fusion. Embeddings run in-process through fastembed's ONNX models, so there is no server to stand up and no API key to hold. An optional cross-encoder rescore tier is available through the rerank extra.
uv tool install metamind-vault-ragRelated MCP server: mnemonic
Command | Purpose |
| Watch a directory and index changes |
| One-shot full reindex |
| Loopback HTTP search API |
| stdio MCP server |
| Environment and index diagnostics |
Variable | Meaning |
| Directory to index |
| Collection name, which scopes the index files |
| Port for the loopback search API |
| Where indexes, caches and logs are written. Defaults to |
Indexes are written to the state directory, named after the collection, and are never placed inside the corpus. Two clients pointed at different collections, or different state directories, coexist on one machine without either knowing about the other.
Installed by any client that wants retrieval without running a service. The engine holds no opinion about who is asking: it names no client in its output, its environment variables are all VAULT_-prefixed, and it writes nothing outside the state directory.
uv run --extra dev pytestA client can be pointed at a working copy instead of a release with uv tool install --from /path/to/this/repo metamind-vault-rag.
MIT
search_vault + wikilinks discovered in source files.
Inputs: query (required), k.
Semantic search over the Obsidian Knowledge vault.
Inputs: query (required), k. Returns: result
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 3 tools
Each tool has a clearly distinct purpose: semantic search, finding related notes, and expanding search via wikilinks. There is no overlap or ambiguity.
All tools use snake_case and follow a verb-like pattern, but 'related_notes' uses an adjective rather than a verb, creating minor inconsistency with the verb_noun style of the other two.
With only 3 tools, the set is minimal but well-scoped for the server's focus on vault search and navigation. It avoids superfluous tools while covering essential actions.
The server covers the core needs of a RAG vault assistant: searching semantically, exploring relationships, and expanding through linked notes. This is a complete set for its intended purpose.
Serve a folder of Markdown notes as an MCP server: hybrid search, reading, and sourced answers.
An MCP server for search and retrieval over a local LanceDB database
Token-efficient MCP memory for Markdown vaults. Tiered search, GraphRAG, AI memories.
Self-hosted AI workspace: semantic RAG document search, long-term memory, and AI chat via MCP.
