browser-automation

Automates multi-step web page control with the OpenClaw browser tool.

1|Updated Jun 9, 2026
One-click install
npx skills add https://github.com/JEROME-PRAKASH-L/openclaw --skill browser-automation-jerome-prakash-l
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: browser-automation
Source: https://github.com/JEROME-PRAKASH-L/openclaw/tree/main/extensions/browser/skills/browser-automation
Command: npx skills add https://github.com/JEROME-PRAKASH-L/openclaw --skill browser-automation-jerome-prakash-l

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Multi-step browser automation often breaks due to stale element refs, duplicate tabs, lost login state, and silent timeouts. This Skill provides a disciplined operating loop for the OpenClaw browser tool so agents can navigate, click, and recover reliably without getting stuck. ## Core Features & Use Cases - Stable Tab Management: Open tabs with labels, reuse existing tabs, and target them via suggestedTargetId instead of fragile raw DevTools IDs. - Snapshot-Driven Actions: Read the page with action="snapshot" and refs="aria" before every click, then re-snapshot after navigation or modal changes. - Stale Ref Recovery: Detect failed refs, re-snapshot the same tab, retry once, and report real blockers like logins, captchas, or 2FA instead of looping. - Use Case: Joining a Google Meet call — open a labeled meet tab, handle camera/microphone permission screens as progress, and report any sign-in or 2FA step that needs the user. ## Quick Start Use the browser tool to open a labeled tab for the target site, snapshot it, and complete the login check and form submission step by step.

Frequently Asked Questions about browser-automation

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate multi-step browser tasks with the OpenClaw browser tool?

Check browser status first, open tabs with labels, snapshot the page before every action, and act using refs from the latest snapshot. After navigation or modal changes, snapshot again before the next step.

How to fix stale element ref errors in browser automation?

Snapshot the same targetId again, locate the current visible control, and retry once with the new ref. If the UI has moved to a blocker state like a login or captcha, report the blocker instead of retrying in a loop.

Can I use my existing browser session and cookies with OpenClaw?

Yes, use profile="user" to attach to your running Chromium-based browser with existing cookies and login state. For these sessions, omit timeoutMs on actions like type, fill, hover, and scrollIntoView since that driver rejects per-call timeouts.

Why does browser automation fail with duplicate tabs after retries?

Retries after timeouts can leave extra windows behind. List tabs with action="tabs" before opening new ones, reuse matching labels or URLs, and close duplicates by their tabId.

What targetId should I pass to OpenClaw browser tool actions?

Pass the suggestedTargetId returned by tabs or open calls, which is the label if one exists or a stable tabId like t1. Avoid bare numbers and raw DevTools target IDs, which can change under Chromium target replacement.