A lightweight browser automation skill that uses opencli and a Chrome/Chromium bridge to control web pages without expensive APIs.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install opencli-browser-automation
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install opencli-browser-automation using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is OpenCLI Browser Automation?
OpenCLI Browser Automation is a developer-centric tool designed to bridge the gap between AI agents and the web browser. By utilizing opencli in conjunction with a specialized Chrome/Chromium extension, this skill enables high-fidelity automation across 79+ websites. It is built to work with existing browser profiles, meaning it inherits your active login sessions for platforms like Zhihu, Bilibili, and Doubao, making it significantly more efficient than standard headless browser solutions.
This approach is a cornerstone for users of Openclaw Skills who need to perform complex web interactions without the overhead of heavy infrastructure. It supports 502 built-in commands and offers an AI-driven discovery mode to automatically map new website structures, synthesize commands, and record manual browser actions into reproducible YAML definitions.
OpenCLI Browser Automation Use Cases
- Automating continuous dialogues with AI assistants like Doubao or Yuanbao for advanced prompting workflows.
- Scraping real-time trending data and rankings from social platforms like Zhihu, Bilibili, and Hacker News.
- Performing automated search and data extraction on technical sites like GitHub, V2EX, and 36kr.
- Executing authenticated web tasks that require pre-existing browser cookies and session states.
- Generating automated visual audits and screenshots for web-based debugging and monitoring.
How OpenCLI Browser Automation Works
- The user establishes a link between the CLI and the browser by loading a specific extension in Developer Mode.
- The skill uses the
opencli doctorcommand to verify the active bridge connection to the Chrome/Chromium instance. - Navigation is handled via the
operate opencommand, which directs the browser to the target URL within the user's active profile. - The
operate statecommand retrieves a map of interactive elements, assigning unique indices to buttons, inputs, and textareas. - Interaction commands (click, type, scroll) are executed based on these indices or via direct JavaScript injection using the
evalfeature. - For new or unmapped sites, the
exploreandgeneratefunctions use AI to analyze the DOM and register new automation commands automatically.
OpenCLI Browser Automation Setup
- Download the
opencli-extension.zipand extract it to a local directory. - Open your browser's extension management page (e.g.,
edge://extensions/orchrome://extensions/). - Enable Developer Mode and click 'Load unpacked', then select the extracted folder.
- Ensure the browser (Chrome Beta or Edge recommended) is running.
- Validate the setup using the following command in your terminal:
opencli doctor
- Confirm the output displays
Extension: connectedto begin using the skill.
OpenCLI Browser Automation Data Schema & Taxonomy
| Component | Description |
|---|---|
| Element Index | Numerical identifiers (e.g., [162]) mapped to DOM elements for easy CLI targeting. |
| Site Map | Definitions for 79+ pre-configured websites including specific selectors for hot lists and search bars. |
| Command Registry | A collection of 502 built-in commands covering navigation, interaction, and data retrieval. |
| Output Data | Supports raw innerText extraction, JSON-serialized DOM structures, and binary screenshot files (PNG/JPG). |
| Daemon State | Metadata containing PID, port assignments, and memory usage for the bridge process. |
OpenCLI Browser Automation Advanced Features
- AI-Driven Command Generation: Use
opencli generateto automatically explore a site and synthesize new interaction commands on the fly. - Cascade Logic: Employs an intelligent strategy to find the simplest and most reliable interaction path for a given web target.
- Session Persistence: Inherits all browser cookies and local storage, allowing Openclaw Skills to bypass many login-related hurdles.
- JavaScript Evaluation: Execute complex custom logic directly in the page context via the
operate evalcommand for advanced data parsing. - Action Recording: Record manual browser steps and convert them into reusable YAML command sets using
opencli record.
SKILL.md
Loading
Related Openclaw Skills
Unraid XML Generator
Automatically generate valid Unraid DockerMan user template XML files for seamless container deployment and management.
Memory Auto-Manager
An automated memory lifecycle manager that extracts key insights from AI conversations and syncs them to permanent storage and vector indexes.
Gate Exchange Dual Investment
A specialized AI agent skill for managing non-principal-protected dual investment products and yield strategies on the Gate exchange.
Amazon Search
A powerful automation tool for extracting structured product data and images directly from Amazon using headless browser technology.
Gate Coupon Assistant
A specialized tool for querying and managing Gate Exchange coupons, vouchers, and trial funds through an AI agent.
Gate Unified Account Assistant
A specialized AI skill for managing Gate.io unified account equity, margin borrowing, and leverage configurations with high-precision data handling.
Featured*
