agent-browser

Automate browser interactions and capture screenshots for UI verification.

2|Updated Mar 29, 2026
One-click install
npx skills add https://github.com/NagyVikt/codex-plugins --skill agent-browser-nagyvikt
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-browser
Source: https://github.com/NagyVikt/codex-plugins/tree/main/plugins/vercel/skills/agent-browser
Command: npx skills add https://github.com/NagyVikt/codex-plugins --skill agent-browser-nagyvikt

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill eliminates the manual effort required to test, verify, and interact with web applications, ensuring that your development server output matches expected behavior without human intervention.

Core Features & Use Cases

  • Browser Automation: Programmatically navigate, click, fill forms, and scroll through web pages.
  • Visual Verification: Capture screenshots and perform visual diffs to ensure UI consistency across deployments.
  • Data Extraction: Scrape text and structured data from dynamic web pages for analysis or testing.
  • Use Case: Automatically verify that a new feature is correctly rendered on a local development server by opening the page, interacting with the UI, and capturing an annotated screenshot.

Quick Start

Use the agent-browser skill to open the local development server at localhost:3000, wait for the network to be idle, and capture a full-page screenshot for verification.

Frequently Asked Questions about agent-browser

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

FAQPage Schema
How do I automate browser interactions to verify web apps on a local development server?

Browser automation for web app verification uses programmatic navigation and form manipulation on a local development server. You navigate pages, interact with the UI, and capture screenshots to verify expected behavior without manual human intervention.

Can I use Playwright or Puppeteer for headless browser visual verification?

Playwright and Puppeteer support headless browser visual verification by executing deterministic UI workflows. They capture full-page screenshots and perform visual diffs to ensure UI consistency across deployments within headless or headed environments.

What is the best way to scrape structured data from dynamic web pages during testing?

Scraping structured data from dynamic web pages during testing is done through automated browser data extraction. You programmatically navigate to target URLs, wait for network idle, and extract text or structured data for analysis or verification.

Does browser automation support waiting for network idle before capturing screenshots?

Browser automation supports waiting for network idle before capturing screenshots. This ensures dynamic web pages fully render their content, allowing deterministic UI workflows and accurate visual verification when testing a local development server.

How do I capture a full-page screenshot for UI consistency verification?

To capture a full-page screenshot for UI consistency verification, you use browser automation to open the target URL, wait for the network to be idle, and trigger the screenshot capture command to save the rendered page output.