Playwright Local
Run local Playwright browser automation for E2E tests, scraping, screenshots, PDFs, and anti-bot workflows without cloud browser farms.
Overview
playwright-local is an agent skill most often used in Ship testing (also Build integrations) that runs local Playwright automation for E2E tests, scraping, and browser artifacts.
Install
npx skills add https://github.com/jezweb/claude-skills --skill playwright-localWhat is this skill?
- Documents remedies for 10 named failure modes including CI timeout hangs and Ubuntu compatibility issues
- Stealth and session patterns for anti-bot and authenticated flows on protected sites
- Headless Chromium, real Chrome, and cloud-adjacent local setups for faithful IP and extension behavior
- Screenshot, PDF export, infinite scroll, and form or login automation recipes
- Playwright v1.57 Speedboard guidance for performance-oriented browser analysis
- 10 documented Playwright error patterns with prevention guidance
- Skill metadata cites daily systems processing 10,000+ pages/day in production scraping
- Documents Playwright v1.57 Speedboard for performance analysis
Adoption & trust: 508 installs on skills.sh; 841 GitHub stars; 1/3 security scanners passed (skills.sh audits).
What problem does it solve?
You need reliable browser automation on your own hardware but keep hitting CI hangs, bot blocks, and undocumented Playwright edge cases.
Who is it for?
Solo builders who test or scrape in real Chromium/Firefox/WebKit locally and want agent guidance that matches daily automation practice.
Skip if: Teams that only need unit tests with no DOM, or strictly managed cloud-only browser grids with no local execution.
When should I use this skill?
You mention playwright, browser automation, web scraping, e2e testing, screenshots, PDF generation, stealth scraping, or bypassing bot detection on local browsers.
What do I get? / Deliverables
You get repeatable local browser scripts with documented fixes for common failures, plus paths for stealth scraping, auth sessions, and visual exports.
- Runnable local Playwright scripts for tests or scraping
- Screenshots, PDFs, or extracted page data from automated sessions
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
End-to-end browser verification is the primary solo-builder moment before release, so Ship testing is the canonical shelf even though scraping also touches Build. Playwright-local focuses on automated browser runs, flaky CI patterns, and real-environment checks—core testing concerns under Ship.
Where it fits
Run headed or headless E2E suites against staging before tagging a release.
Scrape partner portals or authenticated dashboards into your pipeline with persisted cookies.
Capture rendered HTML and screenshots of marketing pages to verify meta and layout after deploy.
Automate periodic extraction of competitor or directory listings for content research.
Quickly script a landing-page flow demo without wiring a full test framework first.
How it compares
Opinionated local Playwright playbook with error catalog—not a generic Selenium tutorial or hosted browser SaaS.
Common Questions / FAQ
Who is playwright-local for?
Indie developers automating browsers for E2E tests, data extraction, or PDF and screenshot generation on their own machines with Claude or similar agents.
When should I use playwright-local?
Use it in Ship when validating flows before release, in Build when wiring scraping or login automation, and when launch or growth tasks need repeatable checks on live pages with real browser behavior.
Is playwright-local safe to install?
It implies browser, filesystem, and network access for automation—review the Security Audits panel on this page and avoid pointing agents at credentials or production data you have not scoped.
SKILL.md
READMESKILL.md - Playwright Local
{ "name": "playwright-local", "description": "Build browser automation and web scraping with Playwright on your local machine. Prevents 10 documented errors including CI timeout hangs, extension testing failures, and Ubuntu compatibility issues. Includes stealth mode for anti-bot bypass, authenticated sessions, infinite scroll handling, screenshot/PDF generation, and v1.57 Speedboard performance analysis. Use when: automating browsers, scraping protected sites, testing with real IPs, bypassing bot detection, generating screenshots/PDFs, or ", "version": "1.0.0", "author": { "name": "Jeremy Dawes", "email": "jeremy@jezweb.net" }, "license": "MIT", "repository": "https://github.com/jezweb/claude-skills", "keywords": [] } # Playwright Local Browser Automation **Status**: Production Ready ✅ **Last Updated**: 2026-01-10 **Production Tested**: Daily scraping systems processing 10,000+ pages/day --- ## Auto-Trigger Keywords Claude Code automatically discovers this skill when you mention: ### Primary Keywords - playwright - puppeteer - browser automation - headless browser - chromium - firefox - webkit - selenium alternative - end-to-end testing - e2e testing ### Secondary Keywords - web scraping - data extraction - content scraping - automated testing - browser testing - screenshot generation - pdf generation - form automation - form filling - login automation - infinite scroll - dynamic content loading - session persistence - cookie management - authenticated scraping - logged-in scraping ### Stealth & Anti-Detection - stealth mode - anti-bot - bypass bot detection - avoid captcha - bypass cloudflare - bypass detection - residential ip - datacenter detection - navigator.webdriver - fingerprinting - browser fingerprinting - user agent rotation - proxy rotation - playwright-stealth - puppeteer-extra-plugin-stealth ### Error-Based Keywords - "target closed" - "browser crashed" - "element not found" - "selector failed" - "timeout exceeded" - "navigation timeout" - "page timeout" - "detached frame" - "execution context destroyed" - "websocket connection failed" - "protocol error" - "cdp error" - "download not completing" - "file download failed" - "infinite scroll not loading" - "scroll detection" - "waiting for selector failed" ### Comparison Keywords - playwright vs puppeteer - playwright vs selenium - local vs cloudflare browser rendering - residential ip vs datacenter ip - stealth plugins - anti-detection tools ### New Features (v1.56-1.57) - playwright mcp server - playwright init-agents - consoleMessages - pageErrors - page.requests - chrome for testing - mouse steps - drag steps - playwright docker - playwright container - nodriver - undetected-chromedriver --- ## What This Skill Does Provides comprehensive knowledge for building browser automation and web scraping with Playwright on your local machine. Includes complete stealth mode setup for anti-bot bypass, authenticated session patterns, infinite scroll handling, and residential IP advantages. ### Core Capabilities ✅ 7-step stealth configuration for anti-bot bypass ✅ Authenticated session scraping with cookie persistence ✅ Infinite scroll handling with deduplication ✅ Full page screenshot and PDF generation ✅ Multi-tab orchestration for concurrent scraping ✅ Human-like mouse movement simulation ✅ User agent rotation and fingerprint masking ✅ Retry patterns with exponential backoff ✅ Integration with Claude Code via Bash tool ✅ **NEW**: Debug methods (consoleMessages, pageErrors, requests) ✅ **NEW**: Advanced mouse control with `steps` option ✅ **NEW**: Playwright MCP Server for AI agent integration ✅ **NEW**: Docker deployment with official images --- ## Known Issues This Skill Prevents | Issue | Why It Happens | Source | How Skill Fixes It | |-------|---------------|---------|-------------------| | Target Closed Error | Page closed before action or browser crash | [GitHub #2938](https://github.com/microsoft/playwright/issues/2938) |