playwright-browser-automation

What This Skill Does

The playwright-browser-automation skill provides direct access to the Playwright API, enabling sophisticated browser automation capabilities. Unlike approaches that might rely on intermediate layers, this skill allows for precise control over web browser interactions. Users can navigate websites, interact with virtually any element on a page (buttons, forms, links), and extract data from web pages. It also supports advanced features such as taking screenshots, generating PDF documents from web pages, and recording video of the automation process. This method is designed to be more reliable and robust than other potential automation methods, offering a powerful tool for automating complex web-based workflows.

Installation

To install the Playwright browser automation skill, you will need to use the clawhub command-line tool.

clawhub install openclaw/skills/skills/spiceman161/playwright-browser-automation

Additionally, Playwright requires its own browser binaries. After installing the skill, you may need to install these separately. For example, to install the Chromium browser (which is often used by default):

npx playwright install chromium

Refer to the Playwright documentation for installing other browsers like Firefox or WebKit if needed.

Use Cases

This skill is ideal for a wide range of tasks involving web browser interaction:

• Web Scraping: Extracting structured data from websites, such as product prices, contact information, or news articles.
• Automated Testing: Performing end-to-end testing of web applications by simulating user interactions.
• Data Entry & Form Submission: Automating the process of filling out and submitting online forms.
• Content Generation: Creating reports or documentation by capturing screenshots or generating PDFs of web pages.
• Workflow Automation: Automating multi-step processes that occur within a web browser, such as logging into multiple systems or processing data across different web applications.
• Monitoring: Regularly checking websites for changes or specific content.

Example Prompts

1. "Automate logging into the admin panel at https://example.com/admin using username admin and password password123, then navigate to the 'Users' section and extract all user emails into a CSV file."
2. "Take a screenshot of the main page of https://www.nasa.gov and save it as nasa_homepage.png. Then, generate a PDF of the page titled 'NASA Homepage'."
3. "Navigate to https://github.com/openclaw/skills and extract the names and descriptions of the top 5 repositories, then output this data as a JSON array."

Tips & Limitations

• Reliability: Playwright's auto-waiting and retry mechanisms significantly enhance reliability compared to simple setTimeout or fixed delays. Leverage getByRole, getByText, and getByPlaceholder for more resilient selectors.
• Dynamic Content: Websites with heavy JavaScript or single-page applications (SPAs) might require explicit waits for elements to appear or for network activity to subside (waitUntil: 'networkidle').
• Headless Mode: By default, Playwright runs in headless: true mode (no visible browser window). For debugging, you can set headless: false in your script to observe the automation in real-time.
• Context Isolation: Use browser.newContext() to create isolated browser sessions, ensuring cookies, local storage, and cache do not interfere between different automation tasks.
• Error Handling: Implement robust error handling (e.g., using try...catch blocks) to gracefully manage unexpected page states or network issues.
• Resource Intensive: Browser automation can be resource-intensive, especially when running multiple instances or complex tasks. Ensure your system has sufficient RAM and CPU.
• Terms of Service: Be mindful of website robots.txt files and Terms of Service when scraping or automating interactions, as excessive or aggressive automation can lead to being blocked.