URL screenshots for AI agents

Let AI agents see rendered webpages

Capture a live webpage, a specific element, or a controlled browser state and pass the image to a vision model, task record, or review queue.

Start free Read the API docs
  • JavaScript-rendered pages
  • CSS selector capture
  • Repeatable viewport settings
Real URL screenshot API requests followed by their rendered webpage images.
Request
POST https://hcti.io/v1/image
{
"url": "https://google.com",
"viewport_height": 480,
"viewport_width": 640
}
https://hcti.io/v1/image/019fb43a-a8b4-7017-a337-6bac3e7305a0
https://google.com
Screenshot of Google in a small viewport

Live demo

Try the URL screenshot API

Enter a public URL, add optional CSS, and generate a real website screenshot with the same controls available through the API.

Loading playground…

Output formats

One API, the format your workflow needs

Keep the same URL and capture settings. Choose lossless detail, compact compatibility, modern web delivery, or a paginated document.

Lossless + transparent
.png

URL to PNG

Crisp text, sharp interface details, and transparent page backgrounds.

Explore PNG
Compact + universal
.jpeg

URL to JPEG

Smaller files and broad compatibility for photographic and content-heavy webpages.

Explore JPEG
Modern + efficient
.webp

URL to WebP

Strong visual quality with modern compression for web delivery.

Explore WebP
Document + printable
.pdf

URL to PDF

Paginated webpage captures with page sizing, margins, backgrounds, and print styles.

Explore PDF

Use cases

Support agent browser tasks with screenshots

Browse image examples

Page inspection

Capture the rendered page before an agent analyzes it or chooses its next action.

UI verification

Take before-and-after screenshots to check interface changes and completed browser tasks.

Research evidence

Attach current product, pricing, or company screenshots to research reports and summaries.

Failure snapshots

Preserve the page state when an automated browser run encounters an error or unexpected result.

Example conversation using the HCTI MCP integration.
Example conversation
MCP connected with OAuth
  1. You: Use the HCTI MCP to take a full-page screenshot of https://example.com in dark mode at 1440px wide.
  2. HTML/CSS to Image: Done. I captured the full page with a 1440px-wide viewport in dark mode: https://hcti.io/v1/image/…
  3. You: Now capture only #pricing at a 390 × 844 mobile viewport.
  4. HTML/CSS to Image: Done. I enabled mobile and touch emulation, rendered the page at 390 × 844, and cropped the image to #pricing.
  5. You: Compare the desktop and mobile versions.
  6. HTML/CSS to Image: I generated both viewport sizes and returned two hosted image URLs so you can review them side by side.

MCP integration

Use the HCTI API from your AI tools

Connect an MCP-compatible AI client to HCTI through OAuth.

Generate images and PDFs, capture webpages, and work with templates without putting an API key in the prompt.

Precise capture

Capture a full page, viewport, or element

Choose the exact area of the webpage that should appear in the screenshot.

Full-page screenshots

Capture the complete page from top to bottom, including content below the fold.

"full_screen": true

Element screenshots

Target a CSS selector and crop the final image to one component, card, chart, or section.

"selector": "#pricing"
#pricing

Viewport size

Set the browser viewport width and height for consistent desktop, tablet, or mobile captures.

1440 x 900
viewport_width + viewport_height

Browser controls

Capture JavaScript-rendered pages in the right state

Set viewport, color scheme, timezone, mobile behavior, consent-banner blocking, and capture delay before the screenshot is taken.

  • Timing controls

    Add a delay, set a timeout, or trigger capture with ScreenshotReady().

  • Device behavior

    Set mobile emulation, orientation, and touch input.

  • Page environment

    Choose light or dark mode, timezone, and screen or print media.

  • Clean output

    Inject custom CSS or attempt to hide cookie and consent banners.

POST https://hcti.io/v1/image request example.
Request
POST https://hcti.io/v1/image
{
"url": "https://example.com",
"viewport_width": 390,
"viewport_height": 844,
"viewport_mobile": true,
"viewport_touch": true,
"color_scheme": "dark",
"ms_delay": 500,
"block_consent_banners": true
}

URL to image

Create a website screenshot in three steps

  1. Send a URL and options

    Authenticate from your server and send the public URL with any viewport, selector, or browser settings.

  2. Get an image ID and URL

    The API response includes an image ID and a hosted URL for the screenshot.

  3. Use the screenshot

    Embed the hosted URL, download the image, or save the file in your own system.

Ready to capture?

Start taking website screenshots

Send one API request and receive a hosted image URL. Add viewport, full-page, selector, and browser controls when you need them.

Start free Read the API docs

Explore more workflows

See how URL screenshots fit your workflow

Explore practical ways different teams use webpage captures.

Marketing

Capture campaigns, product pages, customer sites, and published content.

View workflow

Developers

Automate webpage screenshots with explicit browser, viewport, and device settings.

View workflow

Sales

Add company, product, and contact-page screenshots to prospecting and CRM systems.

View workflow