Aegis Memory – Open-source memory layer for multi-agent AI systems

[Arulnidhi (Arulnidhi Karunanidhi)]

Built this to solve memory problems in multi-agent setups.

What it does:

- Scoped memory (private/shared/global per agent)

- Cross-agent queries and handoffs

- ACE patterns: voting, reflections, session tracking

- Integrations: LangChain, LangGraph, CrewAI

Self-hostable with Postgres + pgvector.

GitHub: GitHub - quantifylabs/aegis-memory: Memory engine for multi-agent AI systems. ACE patterns for persistent learning. Production-ready, self-hostable, Apache 2.0.

PyPI: `pip install aegis-memory`

Would love feedback from anyone building agent systems, and what’s missing? What would make this useful for your workflow?