RepoRadar
Leave a Tip
Item detail

caura-ai/caura-memclaw

MemClaw is an Apache-2.0 'fleet memory' platform for multi-tenant, multi-agent AI fleets that adds trust tiers, keystone policies, audit trails, a knowledge graph, and self-improving retrieval on top of a standard Postgres + pgvector + Redis stack. It runs in three modes (managed cloud, self-hosted Docker, OpenClaw plugin) and ships an MCP server out of the box; the README claims 300+ AI agents in

Open source →Back to radar
Score7.7
Popularity70.0
Risknone
TierSilver
Score breakdown
Usefulness7.0
Novelty8.0
Momentum7.0
Maturity7.1
Open-source/build8.4
Evidence7.2
Workflow potential9.2
Setup ease6.4

Popularity is tracked separately. Support, ads, sponsorships, and tips never affect these signals.

Why it matters

Useful for teams running more than ~10 AI agents in production who need governed shared memory (not just per-agent memory). Self-host with Docker Compose (~5 min), point the MCP server at Claude Code or OpenHands, and confirm a write from one agent is searchable by another before adopting it for production.

Who should use it

teams running more than ~10 AI agents in production who need governed shared memory (not just per-agent memory)compliance-sensitive workloads that need trust tiers, keystone policies, and audit trails on every memory writeengineers who want a Postgres + pgvector + Redis stack under the agent memory instead of a vendor-specific cloudOpenClaw users who want a first-party fleet-memory plugin without writing custom integration codeany team that needs an MCP server so Claude Code / OpenHands / Codex can write to and read from the same fleet memory

Who should skip it

Skip if the source link, docs, or setup requirements do not match your workflow.

Risk explanation

No inherent user-impacting risk is flagged from the captured evidence.

Evidence links

Closest alternatives / related signals

agent-memoryfleet-memorymulti-agentmcppostgrespgvectorredistrust-tiers