메뉴
Browser automation via Playwright CLI — screenshots, scraping JS-rendered pages, PDF generation, link extraction. Runs directly via Bash (no MCP server, no token overhead).
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Ask which skill or flow fits your situation. A router over the user-invoked skills in this repo.
Stages all changes and generates a lean Conventional Commits message via caveman-commit — user reviews and commits manually. Triggers on any commit intent: "commit", "stage my changes", "ready to push", "let's commit", "make a commit", "commit this", "push changes", "squash and commit", or any variation. Use whenever the user signals they're done with changes and ready to commit, even if they don't mention "message" or "caveman". Does NOT run git commit — outputs a ready-to-run command for the user to execute after review.
Autonomous ML research - agent modifies GPT training code, runs 5-min experiments, keeps improvements
Always use browser-harness for any web interaction: automation, scraping, testing, or site/app work.
Shared vocabulary for designing deep modules. Use when the user wants to design or improve a module's interface, find deepening opportunities, decide where a seam goes, make code more testable or AI-navigable, or when another skill needs the deep-module vocabulary.
AI spend analytics — see where your token budget goes by task, model, tool, and project across 31 AI coding tools including Claude Code. Use when the user asks about AI costs, token usage, spend breakdown, or wants to optimize their AI budget.
| name | playwright |
| description | Browser automation via Playwright CLI — screenshots, scraping JS-rendered pages, PDF generation, link extraction. Runs directly via Bash (no MCP server, no token overhead). |
Run node tools/playwright.js directly via Bash. All commands output JSON. No MCP server required.
npm install
Uses your system Chrome or Edge — no separate browser download needed. Falls back to Playwright's bundled Chromium automatically if neither is found (e.g. CI).
| Need | Tool |
|---|---|
| Screenshot any URL | node tools/playwright.js screenshot |
| Scrape JS-rendered page | node tools/playwright.js scrape |
| Generate PDF from URL | node tools/playwright.js pdf |
| Extract all page links | node tools/playwright.js links |
| Quick one-off screenshot (no install) | npx playwright screenshot <url> <file> |
| Large-scale scraping (1000+ pages) | Apify MCP |
| Static HTML / REST API content | Tavily MCP or curl |
| localhost screenshots (frontend dev) | node screenshot.mjs (see frontend-instructions.md) |
node tools/playwright.js screenshot <url> [output-path] [--fullpage] [--headed] [--wait-for <selector>] [--timeout <ms>]
.tmp/screenshot.png--fullpage — capture full scrollable page, not just viewport--wait-for — wait for a CSS selector to appear before capturingnode tools/playwright.js screenshot https://example.com .tmp/example.png
node tools/playwright.js screenshot https://example.com .tmp/full.png --fullpage
node tools/playwright.js screenshot https://app.example.com .tmp/app.png --wait-for "#dashboard"
node tools/playwright.js scrape <url> [--selector <css>] [--headed] [--wait-for <selector>] [--timeout <ms>]
--selector: returns { title, url, text, metaDescription } for the whole page--selector: returns array of { text, html, href } per matched element# Full page text
node tools/playwright.js scrape https://example.com
# Specific elements
node tools/playwright.js scrape https://example.com --selector "h2"
node tools/playwright.js scrape https://store.example.com --selector ".product-name" --wait-for ".product-name"
node tools/playwright.js pdf <url> [output-path] [--format <A4|Letter>] [--headed] [--timeout <ms>]
.tmp/output.pdfnode tools/playwright.js pdf https://example.com .tmp/report.pdf
node tools/playwright.js pdf https://example.com .tmp/report.pdf --format Letter
node tools/playwright.js links <url> [--headed] [--timeout <ms>]
Returns array of { text, href } for all external links on the page.
node tools/playwright.js links https://example.com
All commands write JSON to stdout. Save or inspect:
# Save to file
node tools/playwright.js scrape https://example.com > .tmp/data.json
# Pretty-print
node tools/playwright.js scrape https://example.com | python -m json.tool
Screenshots: after saving, use the Read tool to visually inspect the PNG.
npx playwright screenshot https://example.com .tmp/screenshot.png
npx playwright screenshot --full-page https://example.com .tmp/full.png
npx playwright pdf https://example.com .tmp/output.pdf
| Error | Fix |
|---|---|
Cannot find module 'playwright' | Run npm install |
Executable doesn't exist (after fallback) | Run npx playwright install chromium (CI / no system Chrome) |
net::ERR_NAME_NOT_RESOLVED | URL typo or site is down |
Timeout | Increase --timeout 60000 or add --wait-for |
For multi-step automation tasks, follow workflows/browser-automation.md.