MemClawz

MemClawz is an open-source shared memory system for AI agent fleets, described as Redis for AI agents. It provides a centralized memory layer where multiple AI agents can store, search, and learn from collective experiences. Unlike traditional AI agents that operate in isolation and lose context between conversations, MemClawz enables collective intelligence by allowing agents to share memories. The system features hybrid search combining BM25 keyword and vector similarity, an 8-state memory lifecycle from active to deleted, automatic entity and temporal extraction, composite scoring based on relevance, recency, importance and access frequency, and smart compaction with 3-tier deduplication. Additional capabilities include multi-agent federation for distributed fleets, sleep-time reflection for pattern detection, and 15 structured memory types. The architecture combines Qdrant vector database, Neo4j graph database, and Redis cache with a REST API, MCP server, and WebSocket interfaces.

Use this tool if you:
• Run multiple AI agents that need to share context and learn from each other
• Want persistent memory across agent conversations and sessions
• Need hybrid search combining keyword and semantic similarity
• Want automatic memory enrichment and lifecycle management
• Need to route tasks to the most qualified agent based on past experience
• Want to implement sleep-time reflection for pattern detection
• Need a memory system that scales across distributed agent fleets