browse

Controls a persistent headless Chromium browser for QA testing and site verification.

Updated Jun 16, 2026
One-click install
npx skills add https://github.com/adamtpang/summon.company --skill browse-adamtpang
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: browse
Source: https://github.com/adamtpang/summon.company/tree/main/.claude/skills/browse
Command: npx skills add https://github.com/adamtpang/summon.company --skill browse-adamtpang

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires playwright, bun.

What problem does it solve? Manually verifying that a web page loads, a form submits, or a deployment works requires opening a browser and clicking through flows by hand. This Skill gives an AI agent a persistent headless Chromium instance it can drive with fast commands to navigate pages, interact with elements, capture screenshots, and assert page state. ## Core Features & Use Cases - Page interaction and inspection: Navigate URLs, click, fill, hover, and scroll elements, then read text, HTML, links, forms, console logs, and network requests. - Snapshots and visual evidence: Capture ARIA-tree snapshots with interactive-element refs, diff before/after states, and take annotated or responsive screenshots for bug reports. - State assertions and flows: Assert visibility, enabled, checked, and focused states; test dialogs, file uploads, and responsive layouts across viewports. - Use Case: After deploying a login flow, ask the agent to open the site, fill credentials, submit the form, diff the page state, and screenshot the resulting dashboard as proof the flow works. ## Quick Start Ask the agent to open your staging site in the browser, fill in the login form, submit it, and take an annotated screenshot of the result.

Frequently Asked Questions about browse

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

FAQPage Schema
How do I test a login flow with a headless browser?

Navigate to the login page with goto, run snapshot -i to list interactive elements, fill the email and password fields by their refs, click submit, then use snapshot -D to diff the resulting page and confirm the dashboard appears.

How do I take a screenshot of a web page from the command line?

Run the screenshot command with an output path after navigating to the page, or use snapshot -a -o for an annotated screenshot labeling interactive elements. Responsive mode captures mobile, tablet, and desktop viewports in one call.

Does the headless browser keep cookies and login sessions between commands?

Yes. The browser is persistent, so cookies, tabs, and login sessions survive between commands. The first call auto-starts Chromium in about 3 seconds, and subsequent commands run in roughly 100ms.

Why does the snapshot miss dropdown or autocomplete items?

Dropdown items rendered in portals or popovers often lack ARIA roles and are absent from the accessibility tree. Use the -C flag or snapshot -i, which runs a cursor-interactive scan that detects clickable divs and popover children.

Can I test file uploads and dialogs in headless Chromium?

Yes. Use the upload command with a file input selector and local file path, and set dialog handlers with dialog-accept before triggering alerts or confirms. Dialog contents can be inspected afterward with the dialog command.