agent-browser

Capture, upload, and compare web page screenshots for UI verification.

Updated Apr 22, 2026
One-click install
npx skills add https://github.com/Zeyuzhao/background-agents --skill agent-browser-zeyuzhao
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/Zeyuzhao/background-agents/tree/main/packages/sandbox-runtime/src/sandbox_runtime/skills/agent-browser
Command: npx skills add https://github.com/Zeyuzhao/background-agents --skill agent-browser-zeyuzhao

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill addresses the need for visual verification and UI inspection by enabling users to take, upload, and compare screenshots within the sandbox environment.

Core Features & Use Cases

  • Visual verification: Capture screenshots of web pages before and after changes.
  • UI comparison: Generate visual diffs to identify differences between states.
  • Use Case: A developer verifies UI updates by capturing screenshots before and after a change, then compares them to ensure accuracy.

Quick Start

Use the agent-browser skill to take a screenshot of the current webpage and upload it for review.

Frequently Asked Questions about agent-browser

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

FAQPage Schema
How do I capture UI screenshots for visual verification during web development?

Capture UI screenshots for visual verification by using browser automation to take full-page images within a sandbox environment. This allows you to record the webpage state before and after changes.

What is a visual diff and how does it help with UI inspection?

A visual diff is a generated image comparison that highlights pixel-level differences between two UI screenshots. It helps identify visual changes between webpage states to ensure UI updates are accurate.

How can I compare web page screenshots before and after a code change?

Compare web page screenshots by capturing full-page images before and after your update, then generating a visual diff. This process highlights discrepancies to verify UI accuracy without manual inspection.

Can I take annotated screenshots of web pages in a sandbox?

Yes, you can capture annotated screenshots in a sandbox environment. The skill supports taking full-page images and applying annotations to facilitate quick and effective UI validation and review.

Do I need browser automation techniques to generate visual diffs?

Yes, generating visual diffs requires browser automation techniques to capture screenshots and image processing libraries to compare them. These dependencies are necessary to perform visual verification tasks.

What is the best way to automate UI inspection without manual review?

Automate UI inspection by capturing full-page screenshots in a sandbox and generating visual diffs to compare states. This approach replaces manual review by automatically highlighting differences between webpage versions.