browser-use

Automate navigation, form filling, and data extraction in a live Chromium browser.

6|Updated Jun 25, 2026
One-click install
npx skills add https://github.com/tribgames/mixdog --skill browser-use-tribgames
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: browser-use
Source: https://github.com/tribgames/mixdog/tree/main/src/defaults/skills/browser-use
Command: npx skills add https://github.com/tribgames/mixdog --skill browser-use-tribgames

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? It removes the need to manually click through web pages, fill forms, and copy content by letting the AI drive a live Chromium browser directly, including authenticated and JavaScript-rendered pages that plain URL fetching cannot reach. ## Core Features & Use Cases - Page Interaction: Navigate, click, fill forms, select options, scroll, and handle dialogs using snapshot refs or accessible-name targets. - Content Extraction: Read rendered page text, extract repeated rows by CSS selector, capture screenshots, and print pages to PDF. - Debugging & DevTools: Inspect console logs and network requests, intercept or mock requests, emulate devices and network conditions, and record performance traces. - Use Case: Ask the agent to log into a web app with a saved account, fill out a multi-field form, submit it, and verify the confirmation message — all in one automated flow. ## Quick Start Use the browser tool to open the login page, fill in my saved account credentials, submit the form, and confirm the dashboard loads.

Frequently Asked Questions about browser-use

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

FAQPage Schema
How do I automate form filling in a web browser with an AI agent?▼

Navigate to the page, read the returned snapshot for element refs, then use fill.fields to populate up to 30 fields in one call and submit with click or submit:true. Verify the result with an expect condition on URL or text.

When should I use browser automation instead of fetching a URL?▼

Use web_fetch first for reading or summarizing a known URL. Browser automation is the fallback when content requires rendering, authentication, or interaction, such as JavaScript-heavy pages, sign-in flows, or visual verification.

Can browser automation handle logins with saved passwords?▼

Yes. The fill action supports savedAccount with an email or masked label, and the host fills the username and password itself so the password never appears in replies. CAPTCHA, 2FA, or identity checks are handed back to the user.

Why does my browser automation fail with a stale ref error?▼

Refs become invalid after any page change because every mutation returns a fresh snapshot. Take a new snapshot and use the latest refs, or target elements by accessible name or CSS selector instead of reusing old refs.

Can this tool control an external Chrome or Firefox window?▼

No. It only drives the built-in Chromium browser inside the desktop app and cannot attach to external browser windows or native applications. External windows and OS-level dialogs require a computer-use tool instead.