Package Exports

@unforget-ai/openclaw
@unforget-ai/openclaw/dist/index.js

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (@unforget-ai/openclaw) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

Unforget Memory Plugin for OpenClaw

Long-term memory for your OpenClaw agents. Zero LLM on write. Zero config.

Install

openclaw plugins install @unforget-ai/openclaw

Setup

Add to your OpenClaw plugin config:

{
  "unforget-memory": {}
}

That's it. No API keys. No Docker. No database setup.

On first use, the plugin auto-starts an embedded PostgreSQL + Unforget server via unforget-embed.

How It Works

Auto-Recall: Before each prompt, relevant memories are retrieved and injected into context
Auto-Retain: After each response, important conversation turns are stored as memories
4-Channel Retrieval: Semantic + BM25 + entity + temporal search, fused with RRF

Configuration

{
  "unforget-memory": {
    "autoRetain": true,
    "autoRecall": true,
    "autoRecallTopK": 10,
    "orgId": "my-org",
    "agentId": "my-agent"
  }
}

External Server

If you already run Unforget or want to use your own PostgreSQL:

{
  "unforget-memory": {
    "apiUrl": "http://localhost:9077"
  }
}

Requirements

For embedded mode (default):

Python 3.11+ with pip install unforget-embed
Or uvx (Python package runner)

The plugin tries these in order:

uvx unforget-embed (auto-installs from PyPI)
unforget-embed (if pip-installed)
python -m unforget_embed.cli (fallback)

License

Apache 2.0