# MoltBrain for OpenClaw
Long-term memory extension for [OpenClaw](https://github.com/openclaw/openclaw) that learns and recalls your context automatically.
## Installation
### Option 1: As OpenClaw Extension (Recommended)
#### For Bundled Installation
If you're using OpenClaw from source or have cloned the repository:
1. **Copy the extension to OpenClaw's extensions directory:**
```bash
cd your-openclaw-installation/extensions
git clone https://github.com/nhevers/moltbrain.git moltbrain
cd moltbrain/integrations/openclaw
```
2. **Install dependencies and build:**
```bash
npm install && npm run build
```
3. **Enable the plugin:**
```bash
pnpm openclaw plugins enable moltbrain
```
**Important:** Bundled plugins are disabled by default in OpenClaw. You must explicitly enable them.
4. **Configure in OpenClaw config** (optional):
```json
{
"plugins": {
"entries": {
"moltbrain": {
"enabled": true,
"config": {
"dataDir": ".moltbrain",
"maxMemories": 10,
"autoCapture": true
}
}
}
}
}
```
5. **Restart OpenClaw gateway** (if running) to apply changes.
#### For Standalone Installation
If you're installing in a user directory:
1. **Clone to your extensions directory:**
```bash
cd ~/.openclaw/extensions
git clone https://github.com/nhevers/moltbrain.git moltbrain
cd moltbrain/integrations/openclaw
npm install && npm run build
```
2. **Enable the plugin:**
```bash
pnpm openclaw plugins enable moltbrain
```
3. **Restart OpenClaw gateway.**
### Option 2: As OpenClaw Skill
Copy `skill.json` to your OpenClaw skills directory:
```bash
cp integrations/openclaw/skill.json ~/.openclaw/skills/moltbrain.json
```
Configure in your OpenClaw config:
```json
{
"skills": {
"moltbrain": {
"maxMemories": 10,
"channels": ["discord", "slack", "imessage"]
}
}
}
```
### Option 3: Via MCP Server
Start the MCP server:
```bash
npm run mcp:start
```
Add to your OpenClaw MCP config:
```json
{
"mcp": {
"servers": {
"moltbrain": {
"command": "node",
"args": ["path/to/moltbrain/src/mcp/server.js", "--stdio"]
}
}
}
}
```
## Features
### Automatic Memory Capture
The extension hooks into OpenClaw's agent loop to automatically capture:
- **Preferences**: User-stated preferences and likes/dislikes
- **Decisions**: Important decisions made during conversations
- **Learnings**: Technical insights and discoveries
- **Context**: Project-specific information
### Context Injection
Before each response, relevant memories are automatically injected into the context, giving OpenClaw awareness of past interactions.
### Multi-Channel Support
Works across all OpenClaw channels:
- Discord
- Slack
- iMessage
- Microsoft Teams
- Signal
- WebChat
## Available Tools
When installed as an extension, these tools become available to the OpenClaw agent:
### `recall_context`
Retrieve relevant memories based on current context.
**Parameters:**
- `context` (string): The current context to find relevant memories for
- `maxResults` (number, optional): Maximum number of memories to return (default: 10)
**Example:**
```
recall_context("working on the authentication module")
```
### `search_memories`
Search through stored memories.
**Parameters:**
- `query` (string): Search query
- `limit` (number, optional): Maximum results to return (default: 20)
- `types` (array of strings, optional): Filter by memory types (preference, decision, learning, context)
**Example:**
```
search_memories("database schema", limit=10, types=["decision", "learning"])
```
### `save_memory`
Manually save important information.
**Parameters:**
- `content` (string): The information to remember
- `type` (string): Type of memory - one of: "preference", "decision", "learning", "context"
- `metadata` (object, optional): Additional metadata to store
**Example:**
```
save_memory("Always use prepared statements for SQL queries", type="preference")
```
## Configuration
| Option | Type | Default | Description |
|--------|------|---------|-------------|
| `dataDir` | string | `.moltbrain` | Storage directory |
| `maxMemories` | number | `10` | Max memories per context |
| `autoCapture` | boolean | `true` | Auto-capture observations |
| `channels` | array | `[]` | Enabled channels (empty = all) |
## Troubleshooting
### Plugin Not Loading
**Problem:** Plugin shows as "disabled" in `pnpm openclaw plugins list`
**Solution:**
1. Enable the plugin: `pnpm openclaw plugins enable moltbrain`
2. Restart the OpenClaw gateway
3. Verify in config: `plugins.entries.moltbrain.enabled` should be `true`
### Tools Not Appearing
**Problem:** Tools (`recall_context`, `search_memories`, `save_memory`) don't appear in agent's tool list
**Solution:**
1. Verify plugin is enabled (see above)
2. Rebuild TypeScript: `npm run build` or `pnpm tsc`
3. Restart OpenClaw gateway
4. Check logs for registration messages: `[moltbrain] Extension register() called`
### Bundled Plugin Disabled by Default
**Note:** If the extension is installed in OpenClaw's `extensions/` directory (bundled), it will be disabled by default. You must explicitly enable it using `pnpm openclaw plugins enable moltbrain`.
## API
### Hooks
The extension implements these OpenClaw lifecycle hooks:
- `onSessionStart`: Initialize session memory
- `onMessage`: Inject relevant context
- `onResponse`: Extract and save observations
- `onSessionEnd`: Generate session summary
### MCP Methods
When using MCP, these methods are available:
- `memory/search`: Search memories
- `memory/recall`: Get context-relevant memories
- `memory/save`: Store new memory
- `memory/timeline`: Get recent activity
## MoltBook Integration
MoltBrain integrates with [MoltBook](https://moltbook.com) to share memories and learn from other agents. See the [MoltBook integration guide](../../integrations/moltbook/README.md) for details.
## Development
### Building
```bash
npm install
npm run build
```
### Watching for Changes
```bash
npm run watch
```
## License
MIT
