Persistent key-value memory for AI agents — like Redis, but simpler. Built for the MCP ecosystem.
AI agents lose all context between sessions. Every conversation starts from zero. Agent Memory MCP fixes this by giving agents a persistent, searchable memory that survives restarts, crashes, and context-window limits.
Auto-expiring memories with second-level precision. Set it and forget it.
Organize memories by project, user, or domain. Isolated storage per namespace.
Find memories by keyword across all namespaces. Case-insensitive, ranked by relevance.
Every memory tracks create time, last access, and access count for analytics.
POSIX file locking ensures safe concurrent access from multiple agents.
1,000 entries free. Need more? Go Pro for unlimited storage at $19/month.
pip install agent-memory-mcp
memory_remember — Store a value with optional TTL
memory_recall — Retrieve a value + metadata
memory_forget — Delete a key permanently
memory_search — Search all namespaces by keyword
memory_list_namespaces — List namespaces with counts
memory_clear_namespace — Wipe a namespace
memory_stats — Global storage statistics