Mengram — OpenClaw Memory Plugin

Human-like long-term memory for your OpenClaw agent. Three memory types that work together, with automatic recall and capture on every turn.

What It Does

| Without Mengram | With Mengram | |---|---| | "Which restaurant?" | "Booking Kaganat at 7pm for 2. Vegan menu for Anya?" | | New session = blank slate | Knows your preferences, history, workflows | | Same as day 1 after 100 chats | Deep understanding of who you are |

Memory types:

• Semantic — facts: preferences, relationships, habits
• Episodic — events with timestamps and outcomes
• Procedural — learned workflows that self-improve from failures
• Graph RAG — 2-hop knowledge graph traversal connects related memories

Auto-recall: Before every agent turn, relevant memories are injected into context. No manual tool calls needed.

Auto-capture: After every turn, new information is automatically extracted and stored. Nothing is lost.

Install

openclaw plugins install openclaw-mengram

Setup

1. Get a free API key at mengram.io

2. Add to ~/.openclaw/openclaw.json:

{
  "plugins": {
    "entries": {
      "openclaw-mengram": {
        "enabled": true,
        "config": {
          "apiKey": "${MENGRAM_API_KEY}"
        }
      }
    },
    "slots": {
      "memory": "openclaw-mengram"
    }
  }
}

3. Set your API key:

export MENGRAM_API_KEY="om-your-key-here"

4. Restart OpenClaw. Memory works automatically.

Configuration

| Option | Default | Description | |---|---|---| | apiKey | $MENGRAM_API_KEY | API key from mengram.io | | baseUrl | https://mengram.io | Custom URL for self-hosted | | autoRecall | true | Inject memories before each turn | | autoCapture | true | Store memories after each turn | | topK | 5 | Max results per search | | graphDepth | 2 | Knowledge graph hops (0=off, 1, 2) | | injectProfile | false | Include cognitive profile periodically | | profileFrequency | 25 | Profile injection every N turns | | maxFactsPerEntity | 5 | Max facts shown per entity in context | | maxRelationsPerEntity | 5 | Max relationships shown per entity | | maxEpisodes | 5 | Max episodic memories in context | | maxProcedures | 3 | Max procedures in context | | maxStepsPerProcedure | 8 | Max steps shown per procedure | | captureMessageCount | 10 | Messages to capture after each turn | | requestTimeout | 15000 | HTTP timeout in milliseconds | | debug | false | Verbose logging |

Tools

The agent can also use these tools explicitly:

| Tool | Purpose | |---|---| | memory_search | Search all 3 memory types | | memory_store | Save text to memory | | memory_forget | Delete a memory entity | | memory_profile | Get cognitive profile | | memory_procedures | List learned workflows | | memory_feedback | Record workflow success/failure (triggers evolution) |

Slash Commands

| Command | Action | |---|---| | /remember <text> | Save to memory | | /recall <query> | Search memory | | /forget <entity> | Delete from memory |

CLI

openclaw mengram search "coffee preferences"
openclaw mengram stats
openclaw mengram profile
openclaw mengram procedures

Experience-Driven Procedures

Workflows learn from experience:

Day 1: Agent figures out deploy steps manually
Day 2: Agent finds the saved workflow, follows it (v1)
Day 3: Deploy fails — agent reports failure with context
Day 4: Procedure auto-evolved to v2 with fixed steps

Record outcomes with memory_feedback. On failure with context, the procedure automatically evolves.

vs mem0

| Feature | mem0 | Mengram | |---|---|---| | Memory types | 1 (flat facts) | 3 (semantic + episodic + procedural) | | Knowledge graph | Optional | Built-in Graph RAG (2-hop) | | Self-improving workflows | No | Yes (auto-evolution) | | Cognitive profile | No | Yes | | Price | $99/mo+ | Free (open-source) |

Links

• mengram.io — Get API key
• GitHub — Source code
• API Docs — Full API reference

License

Apache-2.0