Caura
by Community
Caura (formerly MemClaw) — governed shared memory for AI agent fleets. Multi-agent, multi-tenant, MCP-native. Trust tiers, keystone policies, audit trails, knowledge graph, self-im
OSS
Caura
Added 1 Oct 2026
Overview
Caura is an open-source shared memory system for AI agent fleets. It provides multi-tenant memory with trust tiers, policy controls, audit trails, and a knowledge graph. Built as an MCP-native service in Python, it is designed for governed access to agent memory.
Best for
Best for
Teams building governed multi-agent systems in Python.
Use cases
- Enforce access policies across multi-agent memory
- Audit agent memory reads and writes
- Share context between agents via a knowledge graph
Notes
544 stars on GitHub. Last updated 2026-10-01. Licensed Apache-2.0.
Pros
- Apache 2.0 license
- MCP-native integration
- Includes audit trails and trust tiers
Cons
- Community project with modest adoption (544 stars)
- Python-only codebase
- Self-improving retrieval adds operational complexity
Get the free Developer’s Field Guide
A 27-page field guide to the AI coding workflow with Claude. Claude Code, MCP servers, the prompt patterns that work, and what to delegate. Free.
Enter your work email. We send it straight over, plus a few short notes worth knowing. Unsubscribe any time.