Headless Browser Screenshot API

Running a headless browser yourself means managing Chromium installations, handling memory leaks, dealing with zombie processes, and scaling across multiple instances. SnapAPI handles all of this for you.

What you get with SnapAPI's headless browser:

• Real Chromium — Not a simplified renderer. Full JavaScript execution, CSS animations, web fonts, WebGL.
• Managed infrastructure — No servers to maintain, no Chrome versions to update, no memory leaks to debug.
• Instant scaling — Send 1 or 1,000 requests. Infrastructure scales automatically.
• SSRF protection — Private IPs blocked. No risk of internal network exposure.

Puppeteer gives you full control but requires significant infrastructure work:

• Puppeteer: Install Chrome, manage browser lifecycle, handle crashes, implement queuing, build health checks, manage memory, deploy to servers, set up monitoring.
• SnapAPI: One HTTP request. Done.

For most use cases, the trade-off is clear. SnapAPI handles the infrastructure while you focus on your product. Self-hosted Puppeteer makes sense only at very high volumes (10,000+ screenshots/month) or when you need custom browser behavior (clicking, typing, navigating).

Headless browser screenshots are the foundation of visual regression testing. Compare screenshots before and after code changes to catch unintended UI modifications:

// Visual regression test
async function visualTest(pageUrl, baselinePath) {
  const current = await fetch(
    `https://apisnap.dev/api/screenshot?url=${encodeURIComponent(pageUrl)}&width=1440&height=900`,
    { headers: { Authorization: `Bearer ${API_KEY}` } }
  );

  const currentBuffer = Buffer.from(await current.arrayBuffer());

  // Compare with baseline using pixelmatch or similar
  const baseline = fs.readFileSync(baselinePath);
  const diff = pixelmatch(baseline, currentBuffer, null, 1440, 900);

  if (diff > 100) { // More than 100 pixels changed
    console.error(`Visual regression detected: ${diff} pixels differ`);
    fs.writeFileSync('diff-current.png', currentBuffer);
    return false;
  }
  return true;
}

Capture entire pages including content below the fold with full_page=true:

curl -H "Authorization: Bearer snap_your_key" \
  "https://apisnap.dev/api/screenshot?url=https://en.wikipedia.org/wiki/Web_scraping&full_page=true&format=png" \
  -o full-page.png

Full-page capture automatically scrolls to the bottom of the page and captures everything. The output height adapts to the content length. Some pages can produce screenshots thousands of pixels tall.