playwright

Files (1)

• SKILL.md

SKILL.md (first lines)

# SKILL.md - Playwright Browser Automation

**Owner:** Appie-3 (CTO)
**Created:** 2026-04-07
**Purpose:** Browser automation for web scraping, form filling, testing, and repetitive web tasks

---

## What is Playwright?

Microsoft's browser automation framework. Gives your agent full control of a real browser — click buttons, fill forms, scrape content, take screenshots, test websites.

**Works with:** Chromium, Firefox, WebKit (all three browsers)

---

## Installation

```bash
# As npm package
npm install playwright

# Install browsers
npx playwright install chromium
npx playwright install firefox
npx playwright install webkit

# MCP server (for agent integration)
npx @playwright/mcp@latest
```

---

## Core Capabilities

| Capability | What It Does |
|------------|-------------|
| **Navigation** | Go to URLs, click links, navigate back/forward |
| **Interaction** | Click buttons, fill inputs, select dropdowns, checkboxes |
| **Extraction** | Scrape text, images, tables, JSON from any page |
| **Screenshots** | Full page or element screenshots |
| **PDF Generation** | Save pages as PDF |
| **Network Interception** | Block requests, mock responses |
| **Authentication** | Login to sites, handle cookies |
| **File Upload** | Upload files to web forms |

---

## Python SDK (Recommended for Agents)

```python
from playwright.sync_api import sync_playwright

with sync_playwright() as p:
    browser = p.chromium.launch(headless=True)
    page = browser.new_page()
    
    # Navigate
    page.goto("https://example.com")
    
    # Interact
    page.fill("#search-input", "my search query")
    page.click("#search-button")
    
    # Extract
    results = page.query_selector_all(".result-item")
    for result in results:
        title = result.query_selector("h3").inner_text()
        link = result.query_selector("a").get_attribute("href")
        print(f"{title}: {link}")
    
    # Screenshot
    page.screenshot(path="screenshot.png", full_page=True)
    
    browser.close()
```

---

## JavaScript/Node.js SDK

```javascript
const { chromium } = require('playwright');

(async () => {
  const browser = await chromium.launch();
  const page = await browser.newPage();
  
  await page.goto('https://example.com');
  await page.fill('#search', 'query');
  await page.click('#submit');
  
  const results = await page.$$eval('.result', els => 
    els.map(el => ({ title: el.textContent, url: el.href }))
  );
  
  console.log(results);
  await browser.close();
})();
```

---

## Common Use Cases for Appie

### 1. Web Scraping
```python
# Scrape job listings from a site
page.goto("https://jobs.company.com")
page.fill("#keyword", "engineer")
page.click(".search-btn")
page.wait_for_selector(".job-card")

jobs = []
for card in page.query_selector_all(".job-card"):
    jobs.append({
        "title": card.query_selector("h3").inner_text(),
        "company": card.query_selector(".company").inner_text(),
        "url": card.query_selector("a").get_attribute("href")

From S3YED/appie-kit, MIT.

More from S3YED/appie-kit