2 hosted APIs tagged “screenshot”, each also an MCP server for AI agents. Free calls every day to try them; then pay per call with x402 (USDC) or prepaid credits from $5.
One-call URL-to-PDF and full-page screenshot API with Puppeteer's page.pdf()/page.screenshot() semantics — an alternative to running puppeteer yourself or the Browserless PDF API, without any headless-browser infrastructure.
POST /v1/pdf$0.005 · 3 free/IP/dayPOST /v1/screenshot$0.005 · 3 free/IP/dayPOST /v1/snapshot$0.005 · 3 free/IP/dayPOST /v1/markdown$0.005 · 3 free/IP/dayPOST /v1/links$0.005 · 3 free/IP/day
Hosted OpenCV for finding a pattern image inside a larger image: template matching, ORB/SIFT feature matching and RANSAC homography, over HTTP and MCP. Gives automation agents pixel boxes, corners and confidence for UI elements, logos and objects in screenshots. A hosted alternative to running OpenCV (opencv/opencv) yourself.
POST /v1/template-match$0.003 · 3 free/IP/dayPOST /v1/feature-match$0.004 · 3 free/IP/dayPOST /v1/homography$0.004 · 3 free/IP/day