OPENCLAW_INTEGRATION.md

Engram — OpenClaw Integration Guide

How It Works

Engram integrates with OpenClaw as a native plugin. The plugin registers memory tools and lifecycle hooks that give your agent persistent memory across sessions.

File	Purpose
openclaw.plugin.json	Plugin manifest — tells OpenClaw what tools exist
plugin.py	Plugin entry point — dispatches tool calls to memory scripts
src/index.ts	TypeScript plugin — implements lifecycle hooks, tool handlers, category detection

Tools Registered

Tool	Description
memory_store	Store text with semantic embedding and auto-classification
memory_search	Search stored memories using semantic similarity

Lifecycle Hooks

• before_agent_start — searches Qdrant with the user's message, injects relevant memories as <recalled_memories> context
• after_agent_response — extracts facts from the conversation and auto-stores them

Configuration

Add to ~/.openclaw/openclaw.json:

{
  "plugins": {
    "allow": ["engram"],
    "slots": {
      "memory": "engram"
    },
    "entries": {
      "engram": {
        "enabled": true,
        "config": {
          "qdrantUrl": "http://localhost:6333",
          "embeddingUrl": "http://localhost:11435",
          "collection": "agent-memory",
          "autoRecall": true,
          "autoCapture": true,
          "debug": false
        }
      }
    }
  }
}

For multi-machine setups, replace localhost with the IPs of your Qdrant and FastEmbed hosts.

Verifying It Works

# 1. Check Qdrant is reachable
curl http://localhost:6333/healthz

# 2. Check FastEmbed is reachable
curl http://localhost:11435/health

# 3. Check plugin loaded
openclaw status | grep engram

# 4. Test in an OpenClaw chat:
# memory_store "Testing Engram integration" --category fact
# memory_search "testing"

# 5. Enable debug mode to see hook activity:
# Set "debug": true in config, then:
openclaw gateway logs --follow | grep engram