clawbrowser
Use when the agent needs to drive a browser through the Microsoft Playwright CLI (`playwright-cli`) for navigation, form interactions, screenshots, recordings, data extraction, session management, or debugging without loading a full MCP browser. It trains the agent on the CLI commands, snapshots, and session/config habits that make Playwright CLI reliable for scripted browsing.
Why use this skill?
Master browser automation with the clawbrowser skill for OpenClaw. Streamline navigation, form interactions, and data extraction using Playwright CLI commands.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/tezatezaz/clawbrowserWhat This Skill Does
The clawbrowser skill provides OpenClaw agents with robust browser automation capabilities by leveraging the official Microsoft Playwright CLI. Unlike complex browser-integration layers that require full MCP browser setups, this skill is optimized for lightweight, command-driven interactions. It allows the agent to navigate, interact with form elements, capture visual evidence, and handle session states via a standardized terminal interface. By utilizing snapshots to generate dynamic element references (e.g., e1, e2), the agent maintains high accuracy when interacting with DOM elements, effectively minimizing errors caused by page structural changes. It is designed for agents that need to perform complex web-based tasks such as automated form submission, testing, data scraping, and session-managed browsing.
Installation
To integrate this skill into your OpenClaw environment, execute the following command in your terminal:
clawhub install openclaw/skills/skills/tezatezaz/clawbrowser
Ensure that you have Node.js and npm installed on your system, as the skill requires the global installation of @playwright/cli to function. After installation, verify the configuration by running playwright-cli --help to ensure the tool is available in your PATH.
Use Cases
- Automated Authentication: Navigating to login portals, filling credentials, and verifying successful entry.
- Data Extraction: Iteratively visiting pages, snapshotting the DOM to locate data points, and using
evalto scrape structured information. - Visual Regression Testing: Taking screenshots or PDF captures of pages after specific user actions to ensure UI consistency.
- Workflow Verification: Scripting multi-step interactions where an agent needs to verify system behavior by checking console logs or network traffic during the session.
Example Prompts
- "Open https://github.com/login, fill in my credentials from the vault, click the sign-in button, and take a screenshot of the dashboard to confirm success."
- "Navigate to the latest news section on example-blog.com, extract all headline links, and save the data to a local file."
- "Launch the browser in headed mode to debug why the submit button on the contact form is not triggering correctly on the site https://client-portal.test."
Tips & Limitations
Always call snapshot immediately after any navigation or dynamic DOM update. Stale references are the primary cause of interaction failures; re-snapshotting ensures your element references (e1, e2, etc.) are mapped to the current page state. For long-running sessions, use the --session flag to maintain state and avoid repeated authentication overhead. Be mindful that headless mode is the default; if your target site relies on sophisticated user-agent sniffing or complex JS animations, use the playwright-cli.json configuration file to adjust viewport settings, launch arguments, or browser engine types. This skill is strictly for automation and should not be used as a primary daily-driver browser due to the lack of a graphical UI interaction layer for human-agent collaboration.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-tezatezaz-clawbrowser": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: network-access, file-write, file-read, code-execution
Related Skills
audit-badge-demo
Demo skill showcasing the audit badge workflow; still experimental.
clawcast
Skill for managing EVM wallets, transactions, and network helpers via cast; covers onboarding, checks, and operating procedures.
clawcast
Skill for managing EVM wallets, transactions, and network helpers via cast; covers onboarding, checks, and operating procedures.
clawcast-wallet
Skill for managing EVM wallets, transactions, and network helpers via cast; covers onboarding, checks, and operating procedures.
test-audit-badge
Test skill to demonstrate the audit badge; do not use it for real workflows.