browse

Control a persistent headless browser via text commands for QA workflows.

Updated Mar 11, 2026
One-click install
npx skills add https://github.com/Liambiswell20-tech/Bolus-Brain --skill browse-liambiswell20-tech
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: browse
Source: https://github.com/Liambiswell20-tech/Bolus-Brain/tree/main/.claude/skills/gstack/browse
Command: npx skills add https://github.com/Liambiswell20-tech/Bolus-Brain --skill browse-liambiswell20-tech

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates QA workflows by providing a persistent headless browser that you can program against to navigate sites, interact with elements, and capture evidence without manual testing.

Core Features & Use Cases

  • Navigate to URLs and maintain session state across commands for reproducible tests.
  • Interact with page elements, verify states, diff before/after actions, and take annotated screenshots for bug reports.
  • Test forms and uploads, handle dialogs, and inspect network/activity traces to dogfood a feature before release.

Quick Start

Launch the skill and issue a command like "goto https://example.com" to begin an automated QA flow.

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?

Automating QA testing with a headless browser involves issuing text commands to navigate URLs, interact with elements, and capture evidence like screenshots. It maintains session state across commands for reproducible feature verification and regression testing.

Can I capture screenshots and diff page states during automated browser testing?

Yes, you can capture screenshots and diff page states during automated browser testing. The browser supports diffing before and after actions, taking annotated screenshots, and inspecting elements to generate evidence for bug reports.

How do I test forms and handle dialogs in a headless browser?

To test forms and handle dialogs in a headless browser, you program text commands against a persistent session. This allows you to interact with page elements, test uploads, handle dialogs, and inspect network traces for site dogfooding.

What's the best way to dogfood a feature before release without manual testing?

The best way to dogfood a feature before release without manual testing is using a persistent headless browser. It enables deterministic actions across multiple pages, handles dialogs, and inspects network traces to verify feature states automatically.

Does the headless browser maintain session state across multiple QA commands?

Yes, the headless browser maintains session state across multiple QA commands. This persistent state enables reproducible automated tests, allowing you to navigate sites, verify states, and interact with elements sequentially without losing context.