robipop22

robipop22 / openclaw-memory

Public

Triple-layer memory system for AI agents — SQLite + Qdrant + Postgres/AGE

20
2
100% credibility
Found Feb 17, 2026 at 17 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

openclaw-memory is a self-hosted memory system for AI agents offering tiered storage from local quick-access cache to semantic search and knowledge graph for persistent recall across sessions.

How It Works

1
📚 Discover OpenClaw Memory

You hear about a helpful tool that gives your AI helpers a real memory so they remember conversations and facts between chats.

2
🛠️ Set it up easily

With a quick download and simple start command, everything is ready to use right away, no fuss.

3
Choose your memory power
🔹
Basic mode

Perfect for simple local notes that pop up instantly.

🔍
Smart search

Finds memories by what they mean, not just exact words.

🕸️
Full connections

Links people, projects, and decisions into a knowledge web.

4
💾 Save your first memories

You jot down facts, chat summaries, or decisions, and it stores them safely for later.

5
🔍 Search like magic

Ask any question about past info, and it pulls up exactly what you need, even if phrased differently.

6
🤖 Power up your AI

Connect it to your AI helper, and now it remembers users, projects, and choices across every conversation.

🧠 AI remembers forever

Your agents feel smart and personal, recalling details perfectly every time you chat.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 17 to 20 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is openclaw-memory?

openclaw-memory is a TypeScript triple-layer memory system for AI agents, stacking SQLite for instant local cache, Qdrant for semantic vector search, and Postgres/AGE for knowledge graph queries. It lets agents store facts, conversations, and decisions that persist across sessions, with a unified API for storing, searching, and listing memories by agent ID, scope, or tags. Start in lite mode with zero dependencies, then scale to full without rewriting code.

Why is it gaining traction?

Its tiered setup—lite for prototyping, standard for semantic recall, full for entity relationships—means you add power as needed, with graceful fallback if Qdrant or Postgres/AGE flakes out. The CLI handles init, store, search, and infra spin-up via Docker, while the HTTP API supports smart search endpoints like /api/search and graph traversal. Auto entity extraction from memories saves manual LLM prompting.

Who should use this?

AI agent builders needing persistent recall beyond in-memory hacks, especially in multi-agent systems where scopes isolate user vs. global data. Devs prototyping OpenClaw-style frameworks or integrating memory into TypeScript/Bun apps for decisions, org charts, or session summaries.

Verdict

With 17 stars and 1.0% credibility, it's early-stage but battle-ready for experiments—strong README, CLI, and Docker templates make onboarding fast. Skip for production until more adoption; otherwise, solid for agent memory prototypes.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.