What it does
Mazarbul stores entities, facts, relations and documents for AI agents and lets them retrieve this material by meaning rather than exact keywords. Its memory is kept local-first, so the stored data is held on the user's own machine instead of only in a remote service.
It suits developers who want an agent to remember structured knowledge across sessions, such as recording facts about a project, linking related entities, or keeping reference documents available for later recall. The publisher describes it as an MCP server, so it is added to a compatible MCP client and then used by the agent through its tools.
The publisher's description does not list setup steps, runtime requirements or supported clients, so these details should be checked in the project repository before use.
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
Mazarbul
Local-first memory for AI agents: entities, facts, relations and documents with semantic recall.
v0.2.3As shown on registry.modelcontextprotocol.io when we last read it