Autoclaw Browser

# AutoClaw Browser Automation Skill

## Skill Overview
AutoClaw is a browser automation skill that provides comprehensive control over Chrome browser through MCP (Message Communication Protocol) communication with browser extension.

## Prerequisites
Before starting MCP service, ensure the following files exist in the correct directory:
- `options.js` - Browser extension options page script
- `background.js` - Extension background script handling WebSocket connections

### File Location
```
Download from: https://www.wboke.com/download/autoclaw-chrome-extension.zip
Extract to: ~/.openclaw/browser-extensions/autoclaw/
```

## Available Tools

### ⌨️ Keyboard Operations
| Tool | Description | Parameters |
|------|-------------|------------|
| `claw_press_key` | Press single key | `key: string` |
| `claw_press_combo` | Press key combination (e.g., Ctrl+C) | `keys: string` |
| `claw_type_text` | Type text with optional delay | `text: string, [delay: number]` |

### 📸 Screenshot & Content Extraction
| Tool | Description | Parameters |
|------|-------------|------------|
| `claw_take_screenshot` | Capture screenshot of current page | `[fullPage: boolean]` |
| `claw_get_page_content` | Get page HTML or text content | `[type: html\|text]` |
| `claw_get_text` | Get text content of element | `selector: string` |
| `claw_get_html` | Get HTML content of element | `selector: string` |
| `claw_get_attribute` | Get element attribute value | `selector, attribute` |
| `claw_get_count` | Count matching elements | `selector: string` |
| `claw_is_visible` | Check if element is visible | `selector: string` |

### 🖱️ Mouse & Scroll Operations
| Tool | Description | Parameters |
|------|-------------|------------|
| `claw_mouse_move` | Move mouse to coordinates | `x, y` |
| `claw_mouse_click` | Click at coordinates | `[x, y]` |
| `claw_mouse_right_click` | Right-click at coordinates | `[x, y]` |
| `claw_mouse_double_click` | Double-click at coordinates | `[x, y]` |
| `claw_mouse_wheel` |...

# autoclaw_wboke

Empower OpenClaw with enhanced web page control and bookmark management capabilities.

## Features

| Feature | Description |
|---------|-------------|
| Bookmark Management | Complete CRUD operations, support for folder operations |
| CDP Deep Control | Execute JavaScript, element operations, screenshots |
| Automation Scripts | Reusable script templates |
| One-time Authorization | Take control of all tabs without repeated authorization |

## Two Modes

### Enhanced Mode (Recommended) - Port 30000
- MCP enhanced features
- Complete bookmark management
- One-time authorization to take control of all tabs

### Simple Mode - Port 18792
- Direct connection to Gateway
- Basic web page operations

## Quick Start

### 1. Download Extension

**Recommended: Download from official website**
- Visit: **https://www.wboke.com**
- Download the latest AutoClaw Chrome extension
- Or load from local: `autoclaw-plugin/` directory

### 2. Start MCP Server

1. Open Chrome extension management page `chrome://extensions`
2. Enable "Developer mode" in the top right corner
3. Click "Load unpacked"
4. Select `autoclaw_wboke/autoclaw-plugin`

### 2. Start MCP Server

```bash
cd autoclaw/mcp
npm start
```

### 3. Configure Extension

1. Click extension icon → Settings
2. Select port **30000** (recommended)
3. Click "Take control of all tabs" to complete authorization

## Usage

### Through OpenClaw

1. Copy `autoclaw_wboke` to `~/.openclaw/workspace/skills/`
2. Speak directly in OpenClaw, such as:
   - "Help me bookmark this page"
   - "Open Baidu and take a screenshot"
   - "Search bookmarks for Python"

### Through Command Line

```bash
# Open web page
curl -X POST http://127.0.0.1:30000/mcp -d '{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "claw_navigate",
    "arguments": {"url": "https://www.baidu.com"}
  }
}'

# Take screenshot
curl -X POST http://127.0.0.1:30000/mcp -d '{
  "jsonrpc": "2.0",
  "id": 2,
  "method": "tools/call",...