Website screenshot or PDF of any URL
screenshotScreenshot API for AI agents: send {url}, get JPEG, PNG, WebP or PDF as base64 JSON. Options: viewport, full page, dark mode, selector. It does not get past bot protection (use unblock for those pages' text). Failed or blank captures aren't charged.
Tool call template
Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.
{
"id": 1,
"jsonrpc": "2.0",
"method": "tools/call",
"params": {
"arguments": {
"url": "<url>"
},
"name": "screenshot"
}
}Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| url | Yes | absolute http or https URL of a public page, e.g. "https://example.com/" | |
| width | No | custom viewport width in pixels (overrides the device) | |
| device | No | viewport: desktop 1920x1080, desktop_hd 1366x768, mobile (phone), tablet | desktop |
| format | No | image or document format of the capture; pdf is always the full page | jpeg |
| height | No | custom viewport height in pixels (ignored with fullPage) | |
| delayMs | No | extra wait after the page loaded, in milliseconds | |
| fullPage | No | capture the whole scrollable page (height is capped at 16,384 px), not just the viewport | |
| selector | No | CSS selector: capture only the first matching element | |
| colorScheme | No | emulated prefers-color-scheme | |
| waitForSelector | No | CSS selector to wait for before capturing |