memory-lancedb-pro: Give Your OpenClaw Agents a Brain That Remembers

Most AI agents suffer from amnesia – they forget everything between sessions. memory-lancedb-pro changes that with production-grade long-term memory for OpenClaw agents.

🚀 Why You Need This

Without memory: Every chat starts from zero. "Use tabs, not spaces" → next session: "I told you TABS!" 😤

With memory-lancedb-pro: Agent auto-recalls your preferences, past decisions, and project context:

You: "PostgreSQL over MongoDB?"
Agent: "Feb 12 discussion: PostgreSQL chosen for [reasons]..." ✅

✨ Core Features

1. Smart Auto-Capture

• LLM-powered 6-category extraction: profiles, preferences, entities, events, cases, patterns
• L0/L1/L2 layered storage with two-stage deduplication
• Triggers after 2+ messages, no manual memory_store needed

2. Hybrid Retrieval Engine

Query → [Vector Search + BM25 FTS] → Cross-Encoder Rerank → Weibull Decay → Filter

3. Production Ready

• Multi-scope isolation: global, agent:, project:, user:
• Any OpenAI-compatible embedding (Jina, OpenAI, Ollama, Gemini)
• Full CLI: list/search/stats/delete/export/import/upgrade
• One-click setup script handles upgrades, repairs, config validation

🛠️ One-Click Install (Recommended)

curl -fsSL https://raw.githubusercontent.com/CortexReach/toolbox/main/memory-lancedb-pro-setup/setup-memory.sh -o setup-memory.sh
bash setup-memory.sh

Or via OpenClaw CLI:

openclaw plugins install memory-lancedb-pro@beta

⚙️ Quick Config (add to openclaw.json)

{
  "plugins": {
    "slots": { "memory": "memory-lancedb-pro" },
    "entries": {
      "memory-lancedb-pro": {
        "config": {
          "embedding": {
            "provider": "openai-compatible",
            "apiKey": "${OPENAI_API_KEY}",
            "model": "text-embedding-3-small"
          },
          "autoCapture": true,
          "autoRecall": true,
          "smartExtraction": true
        }
      }
    }
  }
}

Restart: openclaw gateway restart

🎯 Pro Features (v1.1.0-beta.10)

✅ OpenClaw 2026.3+ native hook support ✅ Smart extraction with 6 semantic categories ✅ Lifecycle decay – important memories persist longer ✅ Multi-provider reranking (Jina, SiliconFlow free tier) ✅ OAuth auth for LLM extraction ✅ /lesson skill integration

📊 Compared to Built-in memory-lancedb

🌐 Multi-Language Docs

English | 简体中文 | 日本語 | Français | Español | Deutsch...

🚀 Get Started Today

Star the repo: https://github.com/CortexReach/memory-lancedb-pro Install: bash setup-memory.sh Watch demo: YouTube

Your agents will thank you – with memories that actually stick. 🧠✨