Skip to content

Screenshots + Page Context, one API

The web.
In your hands.

Turn any URL into screenshots and structured page data. Ready for your app, your workflow, your next AI agent.

Free demo No signup No card.

▣   https://vercel.com
▲VercelProducts⌄Solutions⌄Resources⌄PricingLog in

Build without
limits

The platform for developers to build, deploy
and scale the web.

Read the docs
▲Vercel☰

Build without
limits

The platform for developers to build, deploy and scale the web.

Page Context (JSON)
{
  "url": "https://vercel.com",
  "title": "Vercel – Build without limits",
  "description": "The platform for developers",
  "headings": [
    "Build without limits",
    "Deploy with confidence"
  ],
  "links": 28,
  "images": 12
}
THE
CAPTURE
ATELIER

Explore the output

One URL.
Every perspective.

Get screenshots in PNG, JPEG, WEBP or AVIF, an editable Basic Vector SVG, plus structured page context — from a single URL.

1440 × 1024
● ● ● 🔒 https://snapforge.web-tasarimci.comPNG
Recorded desktop screenshot of the ScreenshotMCP Spectral homepage
Your URL. Your next result.Open the free Studio

Less browser work. More possibility.

What will you make visible?

Give your agent a pair of eyes.

Bring visual evidence and readable page content into the same workflow. Connect through MCP, or call Page Context from your own agent backend.

Screenshots + structured page data
Explore MCP integration

Your stack. Your way.

From idea
to API call.

A familiar HTTP endpoint. Typed SDKs. A native tool for your AI agent. Start with the interface you already know.

Read the documentation

Use tokens on your server or in a secret manager. Keep credentials out of browser code.

Quick start
curl https://snapforge.web-tasarimci.com/api/v1/screenshot \
  -H 'Content-Type: application/json' \
  -H "Authorization: Bearer $SCREENSHOTMCP_TOKEN" \
  -d '{
    "url": "https://example.com",
    "device": "both",
    "fullPage": true
  }'
Real interfaces. Ready for your environment.

The big picture, in 24 seconds

A URL goes in.
Possibility comes out.

See how the capture workflow fits together, from a public web page to screenshots and page data.

Then try it yourself

Submit a URL · Render viewports · Extract context · Deliver artifacts

Try it first. Scale when you're ready.

Small start.
Big possibilities.

A free browser demo to see it work.
Production plans from $4/month.

Compare all plans

Starter

For solo builders and small product teams

$4/month
2,000 screenshots / monthChoose Starter
  • REST + MCP + CLI
  • Desktop & mobile
  • 7-day artifact retention
  • 2 concurrent captures

Pro

For production

For production apps and AI agents

$12/month
10,000 screenshots / monthChoose Pro
  • Everything in Starter
  • Page Context extraction
  • 10 concurrent captures
  • 30-day artifact retention

Scale

For agencies and high-volume automation

$39/month
50,000 screenshots / monthChoose Scale
  • Everything in Pro
  • 40 concurrent captures
  • 90-day artifact retention
  • Dedicated rate limits

Monthly billing shown. Annual plans include two months free. No seat fees.

Before your first capture

Good questions.
Clear answers.

Can I try it without signing up?

Yes. The free Studio captures public websites without an account or a credit card. Fair-use limits apply. Paid plans are designed for production workloads.

Are the previews on this page live?

The desktop and mobile previews below are recorded examples and the Page Context panel is illustrative. The URL field above opens a live capture workspace that runs the real private MediaHarvester-backed capture pipeline without leaving this page.

What is Page Context?

Page Context returns a structured representation of a rendered page, including visible text, headings, links and interactive elements. Hosted Page Context is included in Pro and Scale. It is useful when your agent needs readable page data alongside screenshots.

Can I capture private or authenticated pages?

The public demo is for public websites. Private-network targets and credential-bearing URLs are protected. Check the product documentation for the supported security model before designing an integration.

How do I connect an AI agent?

Add the hosted /mcp URL to a compatible MCP client and complete OAuth when prompted. You can also use the REST API, JavaScript/TypeScript SDK, Python SDK or CLI. The docs include setup and authentication examples.

Your next idea starts with a URL

The web is full of possibility.
Capture yours.

Make your first capture Explore the docs

Free demo · No account · No credit card