Standalone long-term memory plugin for MiniClaw OS.
AI agents start every session from zero. No memory of what worked. No memory of what failed. No memory of you.
miniclaw-memory gives agents hybrid vector + keyword search across everything they've ever learned:
- Facts — things the agent knows about you and your world
- Lessons — what worked, what didn't, and why
- Postmortems — detailed analysis of failures to avoid repeating them
- Reflections — nightly self-review that extracts patterns
Agent asks: "What do I know about Mike's blog writing preferences?"
mc-kb search "blog writing style"
→ 'Mike prefers conversational tone, short paragraphs'
→ 'Last blog post got 2.4k views with technical hook'
→ 'Always include code examples — readers skip theory'
Memory persists across sessions, restarts, and updates. The agent gets smarter over time.
This is one module in the MiniClaw OS cognitive architecture. Install the full system for memory + planning + continuity + self-repair.
Apache 2.0