forked from Selig/openclaw-skill
6 custom skills (assign-task, dispatch-webhook, daily-briefing, task-capture, qmd-brain, tts-voice) with technical documentation. Compatible with Claude Code, OpenClaw, Codex CLI, and OpenCode.
48 lines
1.5 KiB
Markdown
48 lines
1.5 KiB
Markdown
# OpenClaw Plugins Documentation
|
|
|
|
## Overview
|
|
|
|
Plugins extend OpenClaw with additional capabilities. A plugin is just a small code module that extends OpenClaw with extra features.
|
|
|
|
## Key Capabilities
|
|
|
|
Plugins can register:
|
|
- Gateway RPC methods
|
|
- Gateway HTTP handlers
|
|
- Agent tools
|
|
- CLI commands
|
|
- Background services
|
|
- Skills (via directories in the plugin manifest)
|
|
- Auto-reply commands
|
|
|
|
## Installation & Management
|
|
|
|
Basic commands for plugin operations:
|
|
|
|
```bash
|
|
openclaw plugins list
|
|
openclaw plugins install @openclaw/voice-call
|
|
openclaw plugins enable <id>
|
|
openclaw plugins disable <id>
|
|
```
|
|
|
|
The system loads plugins from multiple locations: config paths, workspace extensions, global extensions, and bundled extensions.
|
|
|
|
## Official Plugins Available
|
|
|
|
Notable built-in options include voice calling, memory search (Core and LanceDB variants), and messaging channels like Microsoft Teams, Matrix, Nostr, and Zalo.
|
|
|
|
## Configuration
|
|
|
|
Plugins use a declarative config model with JSON Schema validation. Plugin config is validated using the JSON Schema embedded in `openclaw.plugin.json`.
|
|
|
|
Plugin slots enable exclusive categories—only one active simultaneously. For example, memory plugins use the `memory` slot.
|
|
|
|
## Security Consideration
|
|
|
|
Plugins run in-process with the Gateway, so treat them as trusted code. Users should only install plugins from reliable sources.
|
|
|
|
## Development
|
|
|
|
Plugins export either a function or an object with registration logic. The manifest requires an `openclaw.plugin.json` file with metadata, schema, and UI hints for the Control UI.
|