Memory Bank MCP
JSON →A production-ready Model Context Protocol (MCP) server that provides a powerful, vector-native memory bank for AI agents. Built with the Protocol-Lattice Go Agent Framework, this server offers persistent, searchable, and shareable memory with multiple database backends.
Tools · 15
- health.ping Check if the server is running.
- memory.embed Get the vector embedding for a piece of text.
- memory.add_short Add a memory to a session's short-term buffer.
- memory.flush Persist a session's short-term buffer to the long-term vector store.
- memory.store_long Directly embed and store a memory in the long-term store.
- memory.retrieve_context Retrieve relevant memories for a query from a session.
- spaces.upsert Create or update a shared space with a TTL and ACL.
- spaces.grant Grant a role (reader, writer, admin) to a principal for a space.
- spaces.revoke Revoke a principal's access to a space.
- spaces.list List all spaces a principal has access to.
- shared.join Make a principal's session view include a specific space.
- shared.leave Remove a space from a principal's session view.
- shared.add_short_to Add a short-term memory directly to a shared space.
- shared.retrieve Retrieve memories from a principal's merged view (local + joined spaces).
- engine.metrics Get a snapshot of the memory engine's performance metrics.
Environment variables
QDRANT_URLQDRANT_API_KEYGEMINI_API_KEYYOUR_GEMINI_API_KEY
Links
★ 14 GitHub stars