skills.rest
timikalo7
timikalo7COMMUNITY
Explore all skills by this author →
GitHub

browser-drive

Drive a real Chromium browser with Playwright to automate interactive web tasks.

Updated Jun 9, 2026
Search Tags:#browser automation#web scraping#form filling#playwright#chromium#headless browser#interaction capture
One-click install
npx skills add https://github.com/timikalo7/Execute --skill browser-drive-timikalo7
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: browser-drive
Source: https://github.com/timikalo7/Execute/tree/main/.claude/skills/browser-drive
Command: npx skills add https://github.com/timikalo7/Execute --skill browser-drive-timikalo7

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires playwright.

What problem does it solve? Many web tasks require a browser that actually does things — filling forms, clicking through flows, logging in, scraping JavaScript-rendered single-page apps, or reproducing a user-reported bug — and plain HTTP fetching or static screenshots cannot reach that content. This Skill provides a proven Playwright recipe plus container-specific fixes so browser automation works reliably in this environment. ## Core Features & Use Cases - General browser driving: Launch the container's pre-installed Chromium via Playwright to fill forms, click elements, read rendered text, and take screenshots of any live page. - External site access via Node transport: Intercept every browser request and serve it through Node's fetch, bypassing the container's blocked direct HTTPS connections, with video recording and periodic screenshots. - Interaction and design-system capture: Photograph interactive states (menus, accordions, forms, preloaders) and extract a reference site's real type scale, colors, spacing, and motion directly from the DOM. - Use Case: You need to reproduce a bug a user reported on a live web app — launch Chromium, log in, click through the exact flow, and capture screenshots of each state as evidence. ## Quick Start Ask the agent to use the browser-drive skill to open a target URL in Chromium, fill in the login form, click through the flow, and save a screenshot of the result.

Frequently Asked Questions about browser-drive

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

FAQPage Schema
How do I automate browser interactions with Playwright?▼

Launch Chromium with playwright's chromium.launch, create a page with a set viewport, then use page.goto, page.fill, page.click, and page.locator to drive the flow. Run the script with node and point executablePath at the pre-installed browser binary.

How to scrape a JavaScript-rendered SPA page?▼

Load the page in a real Chromium instance with Playwright and wait for domcontentloaded, then read rendered content via page.locator(...).innerText() or page.evaluate. Scroll the page first so lazy-mounted sections enter the DOM before extraction.

Why does Playwright fail with 'Executable doesn't exist' for chromium_headless_shell?▼

The project's pinned Playwright version expects a browser build absent from /opt/pw-browsers. Fix it by passing executablePath pointing to the installed Chromium binary instead of running npx playwright install.

Can Chromium in this container reach external websites directly?▼

No, direct external HTTPS connections from Chromium are blocked below the proxy layer. The workaround intercepts every browser request with Playwright's route handler and fulfills it through Node's fetch, which can reach external sites.

When should I use browser driving instead of screenshot QA tools?▼

Use browser driving when the browser must perform actions — clicking, filling forms, logging in, downloading behind a click. Use visual-qa or headless-screenshot-verify when you only need accessibility checks or pixel-level screenshot verification.

Related Skills

Explore More Software Engineering Skills→
jleechanorgjleechanorg

dogfood

Identify, document, and report bugs in web applications through exploratory QA workflows.

Official
Advanced
KhaledMKhaledKhaledMKhaled

recreate-screenshot

Recreate screenshot references into pixel-perfect UI implementations with exact colors and typography.

Community
Advanced
SirTheOracleSirTheOracle

adversarial-verify

Re-run QA manifest findings and generate a verification-report.yaml with pixel-diff evidence.

Official
Advanced
ginlix-aiginlix-ai
1.7k

web-scraping

Scrape web pages as markdown, HTML, or text using MCP tools and the Scrapling Python API.

Official
Intermediate