Puppeteer

---
name: Puppeteer
description: Automate browser tasks — navigate pages, take screenshots, click elements, and extract content via MCP
version: 1.0.0
author: chvor
type: tool
category: web
icon: globe
tags:
  - puppeteer
  - browser
  - automation
  - screenshots
  - web-scraping
  - testing
  - mcp
  - headless
  - navigation
mcp:
  command: npx
  args: ["-y", "@modelcontextprotocol/server-puppeteer"]
  transport: stdio
---
You have access to a headless browser via Puppeteer MCP.

## Available MCP Tools

- **puppeteer_navigate** — Navigate to a URL
- **puppeteer_screenshot** — Take a screenshot of the current page (or a specific element)
- **puppeteer_click** — Click an element by CSS selector
- **puppeteer_fill** — Fill a form input with text
- **puppeteer_select** — Select a dropdown option
- **puppeteer_hover** — Hover over an element
- **puppeteer_evaluate** — Run JavaScript in the page context and return results

## Common workflows

- **Take a screenshot**: \`puppeteer_navigate\` → \`puppeteer_screenshot\`
- **Fill a form**: \`puppeteer_navigate\` → \`puppeteer_fill\` (fields) → \`puppeteer_click\` (submit)
- **Extract data**: \`puppeteer_navigate\` → \`puppeteer_evaluate\` with DOM query → parse results
- **Test a page**: Navigate → interact → screenshot → verify visual state

## Best practices

- Wait for page load before interacting — navigation returns when the page is ready
- Use specific CSS selectors (IDs, data-testid) over fragile class names
- Take screenshot