A technical configuration skill to connect OpenClaw to remote Chrome instances via CDP for automated web interactions.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install openclaw-browser-auto
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 openclaw-browser-auto 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 OpenClaw Browser Automation?
This skill enables OpenClaw to interface with remote Chrome or CDP-compatible browsers, allowing AI agents to perform sophisticated browser automation tasks. By leveraging this capability within Openclaw Skills, developers can run headless browsers in Docker containers or connect to cloud services like browserless.io, effectively separating the automation logic from the browser execution environment.
The skill provides a standardized way to manage browser profiles, ensuring that Openclaw Skills can interact with web content regardless of where the browser is actually hosted. This is particularly useful for server-side automation where a local GUI browser is unavailable or undesirable.
OpenClaw Browser Automation Use Cases
- Automating web interactions on a remote server without a local UI.
- Running headless browser tasks inside Docker containers using specialized images like chromedp/headless-shell.
- Integrating cloud-based browser services such as browserless.io for scalable automation.
- Executing web-based workflows in CI/CD pipelines via Openclaw Skills.
- Managing persistent browser sessions for complex, multi-step AI agent tasks.
How OpenClaw Browser Automation Works
- Deploy a browser instance with Chrome DevTools Protocol (CDP) enabled, typically via a Docker container.
- Verify the CDP endpoint is accessible via a JSON version check.
- Configure the OpenClaw gateway by defining a specific profile with the correct cdpUrl in the openclaw.json configuration file.
- Enable the browser module and set the attachment mode to remote-only to prevent local browser initialization.
- Restart the OpenClaw gateway to initialize the connection and apply the new configuration.
- Use the OpenClaw CLI to verify the browser status and begin automation tasks using Openclaw Skills.
OpenClaw Browser Automation Setup
1. Launch Browser Container
It is recommended to use chromedp/headless-shell for a lightweight and persistent session. Run the following command to start the container:
docker run -d --name browser-auto -p 9222:9222 --shm-size=512m chromedp/headless-shell:latest
Verify that the CDP is available:
curl http://127.0.0.1:9222/json/version
2. Configure OpenClaw
Edit your configuration file located at ~/.openclaw/openclaw.json to include the browser settings:
{
"browser": {
"enabled": true,
"defaultProfile": "remote-chrome",
"attachOnly": true,
"profiles": {
"remote-chrome": {
"cdpUrl": "http://127.0.0.1:9222",
"color": "#00AA00"
}
}
}
}
3. Restart and Verify
Apply the changes by restarting the gateway and checking the status:
systemctl --user restart openclaw-gateway
openclaw browser status
OpenClaw Browser Automation Data Schema & Taxonomy
The configuration is managed through the standard OpenClaw configuration schema. Below are the primary configuration fields utilized by this skill within Openclaw Skills:
| Configuration Key | Description |
|---|---|
browser.enabled |
Boolean flag to activate the browser automation module. |
browser.defaultProfile |
The name of the profile to use by default for automation. |
browser.attachOnly |
If set to true, the system will not try to launch a local browser, only connect via CDP. |
profiles.<name>.cdpUrl |
The network address (HTTP/HTTPS) of the remote Chrome DevTools Protocol. |
profiles.<name>.color |
A UI metadata field used to visually distinguish sessions. |
OpenClaw Browser Automation Advanced Features
- Multi-profile management allowing Openclaw Skills to switch between different remote browser environments.
- Support for authenticated cloud providers like browserless.io using API tokens embedded in the CDP URL.
- Optimization for headless session persistence to prevent tab loss during automation cycles.
- Environment-specific service configuration via systemd for robust production deployments.
- Support for custom SHM sizes in Docker to handle memory-intensive web pages.
SKILL.md
Loading
Related Openclaw Skills
GitHub Passwordless Setup
A comprehensive skill to automate GitHub authentication using SSH keys and Personal Access Tokens for zero-password operations.
rBitcoin Agent Skill
A comprehensive automation toolkit for deploying, verifying, and managing rBitcoin nodes with integrated CPU mining.
Zan Diary System
A comprehensive personal management system for tracking daily logs, task lists, and annual goal progress within a structured Markdown environment.
PersonalDataHub
PersonalDataHub acts as a secure access control gateway for AI agents to interact with private data from Gmail and GitHub while maintaining strict owner-defined privacy policies.
AI News BNB Trader
A TypeScript-based trading bot that executes BSC swaps based on news sentiment and strict risk management.
EVM BNB Band Trader
A specialized automation tool for executing price-triggered swing trades on the BNB Chain with built-in risk management.
Featured*
