agent-browser

Drive a controlled browser to navigate pages, capture screenshots, and inspect console and network activity.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Static HTTP fetches cannot reach JavaScript-rendered pages, login-gated content, or dynamic UI states, making it impossible to verify UI changes, reproduce reported bugs, or capture visual evidence from modern web apps. ## Core Features & Use Cases - Supervised Browser Verification: Navigate pages, wait for selectors or network-idle, and capture viewport or full-page screenshots as evidence. - Console and Network Inspection: Read console warnings/errors and non-2xx network responses to diagnose UI bugs and failing requests. - Interactive Flow Walkthroughs: Click through short flows step by step, re-screenshotting after each state change to confirm acceptance criteria. - Use Case: A user reports a broken checkout button. Launch the browser, reproduce the flow at a 1366x768 viewport, capture the console error and the failing 500 response, and deliver reproduction steps with screenshots. ## Quick Start Use the agent-browser skill to open the staging site, wait for the dashboard to render, and capture a full-page screenshot plus any console errors.

Frequently Asked Questions about agent-browser

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

FAQPage Schema
How do I verify a UI change on a live web page?

Launch a controlled browser, navigate to the page, wait for a specific selector or network-idle rather than arbitrary sleeps, then capture a viewport or full-page screenshot immediately. Record console errors and any non-2xx network responses as supporting evidence.

When should I use a browser instead of curl for web pages?

Use a browser only when the content is JavaScript-rendered, login-gated, or requires interaction, since curl cannot see SPA content. For static HTML pages, curl or HTTP fetch is cheaper, faster, and more reliable.

Can I automate login-gated pages with a browser?

Yes, but prefer programmatic auth such as API tokens or magic links over UI login. If UI login is required, the user must provide credentials explicitly for that session, and credentials must never appear in logs, screenshots, or output.

Why does my browser automation fail with selector not found?

The page may have changed or the selector is checked before rendering completes. Take a screenshot to see the actual page state, then adjust the selector or wait condition accordingly.

What are the limitations of browser-based verification?

It is unsuited for unattended large-scale scraping, sites behind authentication you do not own, and sensitive accounts like banking where automation risks lockout. Cross-origin iframes you do not own also cannot be inspected.