Persistent memory for AI agents — REST API, MCP, OAuth, CLI, dashboard. Semantic search, knowledge graph, automatic consolidation. One self-hosted service, every transport.
Interactive 3D knowledge graph — every memory a glowing node, every relationship a curved edge.
MINOR release: conditional temporal decay, a functional belief derivation pipeline, a consolidation clustering fix, and bootstrap belief injection with task-aware retrieval. ~2,647 tests.
Opt-in temporal decay for search relevance via MEMORY_DECAY_WINDOW_DAYS (default 0 = disabled, zero behavior change for existing users). When enabled, recent memories rank above older ones with equivalent similarity. (#123, @filhocf)
Semantic grouping plus a noise filter make derive_beliefs actually work — the prior exact-content-equality grouping produced zero belief promotions ever on real data. Consolidation clustering also gets an embeddings fix restoring clusters and associations on real databases. (#124, #126, @filhocf)
The bootstrap profile now injects derived beliefs and supports task-aware retrieval via new task_summary and budget_tokens params on get_bootstrap_profile. Belief derivation is gated behind MCP_BELIEFS_ENABLED (default false). (#127, @filhocf)
Claude.ai connects to your memory server through Streamable HTTP with OAuth 2.1 authentication.
Battle-tested with comprehensive testing and optimized for performance.
Talks and coverage from the community.
Heavybit's Humans in the Loop deep dive talks to maintainer Heinrich Krupp about agent amnesia and why persistent memory is the missing infrastructure layer for agentic systems.
Maintainer Heinrich Krupp presents mcp-memory-service at the AI Tinkerers Zürich meetup — persistent memory architecture, semantic search, and multi-agent memory sharing. Watch the full video →
Install from PyPI and connect your AI agents to persistent memory.
I help teams run self-hosted memory in production — OAuth, hybrid sync, multi-client rollout. Free 30-minute intro call.
Get in touch