browse

Drive a headless browser to load URLs, interact with elements, and capture page state.

Updated Mar 30, 2026
One-click install
npx skills add https://github.com/nhattran998/personal-athlete-agent --skill browse-nhattran998
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: browse
Source: https://github.com/nhattran998/personal-athlete-agent/tree/main/.agents/skills/gstack-browse
Command: npx skills add https://github.com/nhattran998/personal-athlete-agent --skill browse-nhattran998

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Fast headless browser for QA testing and site dogfooding. Navigate any URL, interact with elements, verify page state, diff before/after actions, take annotated screenshots, check responsive layouts, test forms and uploads, handle dialogs, and assert element states. ~100ms per command. Use when you need to test a feature, verify a deployment, dogfood a user flow, or file a bug with evidence. Use when asked to "open in browser", "test the site", "take a screenshot", or "dogfood this". (gstack)

Core Features & Use Cases

  • Fast headless browser for QA testing and site dogfooding.
  • Navigate, interact, verify, diff, and screenshot pages across devices.
  • Use in bug reproduction, deployment verification, and user-flow dogfooding.

Quick Start

Open a URL in a headless browser and verify a key element is visible.

Frequently Asked Questions about browse

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

FAQPage Schema
How do I automate QA testing with a headless browser to verify page state?

Automate QA testing by driving a headless browser to load URLs, interact with elements, and assert element states. It captures page state and takes annotated screenshots at roughly 100ms per command to validate feature deployments.

What is the best way to take annotated screenshots for bug reproduction?

The best way to take annotated screenshots for bug reproduction is using a headless browser that navigates URLs, interacts with elements, and captures visual evidence. This approach provides deterministic commands and state checks for filing bugs with proof.

Can I test responsive layouts and forms across different devices in a headless browser?

Yes, you can test responsive layouts and forms across devices in a headless browser. It supports navigating pages, handling dialogs, testing uploads, and verifying element states across responsive layouts to ensure cross-device compatibility.

Does headless browser testing support diff before and after actions for deployment verification?

Headless browser testing supports diff before and after actions for deployment verification. It applies deterministic commands to interact with elements, capture page state, and assert element states to verify deployments and dogfood user flows.

Why use a headless browser for automated user flow dogfooding instead of manual checks?

Use a headless browser for automated user flow dogfooding because it executes deterministic commands at ~100ms each, interacting with elements and asserting states. This provides faster, more reliable feature verification than manual checks.

What are the limitations of using headless browser automation for QA validation?

Limitations of headless browser automation for QA validation include its focus on deterministic commands, element-state assertions, and page-state diffs. It is designed for feature verification, deployment checks, and bug reproduction rather than performance testing or network analysis.