A comprehensive browser automation skill that enables AI agents to test websites, fill forms, and validate responsive designs using Playwright.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install playwright-skill
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 playwright-skill 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 Playwright Browser Automation?
This skill brings the full power of Playwright to your development workflow, allowing your AI agent to interact with websites exactly like a human user. By utilizing Openclaw Skills, you can automate complex browser-based tasks such as end-to-end testing, visual regression checks, and performance validation without manually writing boilerplate code.
The skill is designed for maximum efficiency, featuring automatic detection of local development servers to streamline the testing of localhost environments. It prioritizes a clean workspace by generating ephemeral test scripts in the system's temporary directory, ensuring your main project files remain uncluttered while providing a visible browser window for real-time debugging and observation.
Playwright Browser Automation Use Cases
- Automatically testing local development servers for regressions after code changes.
- Validating website responsiveness across desktop, tablet, and mobile viewports.
- Automating repetitive browser tasks such as filling out complex multi-step forms.
- Testing authentication flows including login, redirect logic, and session persistence.
- Performing SEO and accessibility audits by checking for broken links and metadata.
How Playwright Browser Automation Works
- The AI agent receives a request to perform a browser-related task or website test.
- The skill initiates a server detection sequence to identify any running development environments on localhost.
- Based on the target URL, the skill generates a custom Node.js script utilizing the Playwright library.
- This script is saved to a temporary directory (/tmp) to keep the user's project folder clean.
- The skill executes the script using a dedicated runner that launches a visible Chromium instance for transparency.
- Real-time feedback, including console logs and saved screenshots, is returned to the agent and user.
Playwright Browser Automation Setup
To initialize the Playwright Browser Automation capabilities within Openclaw Skills, run the following setup commands in the skill directory:
cd $SKILL_DIR
npm run setup
This command installs the necessary dependencies and the Chromium browser binary required for automation.
Playwright Browser Automation Data Schema & Taxonomy
The skill manages data primarily through the system's temporary storage to maintain a lightweight footprint. Below is how the skill organizes its output:
| Data Type | Storage Location | Purpose |
|---|---|---|
| Test Scripts | /tmp/playwright-test-*.js |
Custom logic generated for specific automation tasks. |
| Visual Assets | /tmp/*.png |
Screenshots captured during testing (e.g., desktop, mobile). |
| Configuration | $SKILL_DIR/lib/helpers.js |
Core utility functions and server detection logic. |
| Execution Logs | Standard Output | Detailed logs providing status updates on browser actions. |
Playwright Browser Automation Advanced Features
- Smart Dev Server Detection: Automatically finds active ports on localhost to eliminate manual URL entry.
- Custom HTTP Headers: Support for identifying automated traffic or bypassing bot detection via environment variables.
- Multi-Viewport Testing: Built-in patterns for testing responsive design across several device sizes in a single pass.
- Safe Interaction Helpers: Integrated utilities for resilient clicking, typing, and handling of cookie banners.
- Inline Script Execution: Capability to run quick, one-off browser commands directly from the CLI without creating persistent files.
SKILL.md
Loading
Related Openclaw Skills
Proactive Messages
A proactive AI skill that silently schedules and sends context-aware follow-ups and reminders based on conversation, calendar, and email triggers.
Proactive Messages
An AI agent skill designed to detect significant life events in conversation and silently schedule warm, personalized follow-ups.
X Monitor
An automated monitoring tool that surfaces high-value technology and trend content from X while filtering out political noise and engagement farming.
probtrade
A comprehensive toolset for Polymarket prediction market intelligence and automated trading execution.
Kiro Creator Monitor Daily Brief
An automated intelligence loop that monitors social and developer platforms to generate daily summaries and ready-to-post social media drafts.
Kiro Search Aggregator
A multi-source search utility that aggregates, ranks, and summarizes data from Google, Scholar, YouTube, and X into a single research brief.
Featured*
