idan-rubinidan-rubinCommunityยท1 Agent Skills Included

browserclaw

Deterministic browser automation with numbered element refs

Controls real browsers through text snapshots where every interactive element gets a numbered ref for exact clicking, typing, and form filling. Eliminates fragile CSS selectors, XPath queries, and costly vision-model screenshots from web automation workflows. Handles iframes, dialogs, multi-tab sessions, and data extraction so your own code keeps full control of the reasoning loop.
npx skills add idan-rubin/browserclaw --all -g -y

All Skills in This Repository (1)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install browserclaw?โ–ผ

Run `npx skills add idan-rubin/browserclaw --all -g -y` in your terminal to install the skill globally.

How does browserclaw automate browsers without selectors?โ–ผ

It takes an accessibility snapshot where every interactive element gets a numbered ref like e3, then resolves that ref to one exact element via Playwright locators. No CSS selectors, XPath, or coordinate guessing needed.

Does browserclaw include its own AI agent?โ–ผ

No. It is deliberately just the eyes and hands โ€” you bring your own model and agent loop, keeping full control over reasoning and decisions.

Can browserclaw handle iframes and popups?โ–ผ

Yes. Elements inside cross-origin iframes get frame-prefixed refs, and it includes built-in handlers for native dialogs, multi-tab sessions, and page state inspection.

Do I need coding experience to use browserclaw?โ–ผ

Basic TypeScript or JavaScript knowledge helps since it is an embeddable library, but the included skill file lets your agent drive it from plain-English requests.

Related Repositories in Software Engineering

View All in Software Engineeringโ†’