On-demand URL screenshot capture

CaptureKit renders any URL into an image or PDF on request, letting an agent capture what a page actually looks like at a point in time. A trigger such as a new deployment, a QA run, or a content update fires the capture, the agent reads the target URL and the rendering options it needs, such as viewport size, full page, or device type, and stores the resulting file where the team expects it. When a page fails to render, times out, or returns an unexpected layout, the agent routes the capture to a person to check rather than filing a broken image.

Scheduled and monitored captures

CaptureKit can be called on a schedule, which lets an agent watch a page for visual drift instead of capturing it once. On each run the agent requests a fresh screenshot of the same URL, compares it against the last stored capture or a rule the team set, such as a pricing block or hero banner that should not change, and logs whether the page matches expectations. When the comparison flags a difference, a missing element, or a page that stopped loading, the agent hands the case to the page owner instead of deciding what changed on its own.

Full-page and PDF rendering options

CaptureKit accepts rendering parameters such as full-page capture, PDF output, custom viewport and device emulation, and wait conditions for content to finish loading before the shot is taken. An agent sets these options based on the request type, for example a full-page PNG for a marketing review or a PDF for an archived record, then submits the capture and retrieves the file once ready. This lets the same workflow produce different output formats without a person choosing settings each time. Renders that come back blank, cut off, or clearly wrong are sent to a person to re-run or investigate rather than stored as-is.