Self-hosted shared memory, reminders and agent-to-agent messages for teams of AI agents, ranked by trust, with MCP tools, a JSON API and a web UI.
Setup
This project documents no single launch command; follow its repository instructions.
Big Agentic Brain
Enhesa
Related
More- IIrantinfemmanuelShared persistent memory for multi-agent systems: facts stored as entity/key/value triples with hybrid search, checkpoints and conflict resolution.Knowledge & Memory5 Repository stars
- OOpenMemoryMem0OfficialA private, local memory layer from Mem0 that gives AI tools persistent memory with data kept on your machine.Knowledge & Memory67.6k Repository stars
- HHindsightvectorize-ioOfficialLong-term memory for AI agents with retain, recall and reflect operations and semantic, BM25, graph and temporal retrieval; self-hostable.Knowledge & Memory52.6k Repository stars
- CCognee MCPtopoteretesOfficialMemory for AI agents built on graph and vector stores, with ingestion from many sources, recall, deletion and code graph search.Knowledge & Memory32.7k Repository stars
- GGraphiti MCP ServerZepOfficialLets agents build and query temporally aware knowledge graphs, adding interactions and data incrementally and retrieving history.Knowledge & Memory32.1k Repository stars
- SSupermemorySupermemoryOfficialMemory and context engine for AI apps: extracts facts from conversations, keeps user profiles and offers hybrid search; can run locally.Knowledge & Memory31.4k Repository stars