Draw one of your screens
Draw one of this person's own Trillion screens at a width, as they see it, to put beside their sites on a board or a canvas
With path (one of the workspace's screen addresses, such as /files), width (pixels, 1280 when not given) and origin (the service address the browser asked at), answers what the renderer drew: { status: ready, image_base64, width, height, routes } or { status: unsupported | unavailable, reason }. Drawn only for the person's own signed-in browser, which the renderer reads the screen as for that one drawing; an AI's key is refused.
What it asks for
- path (Needed, words):
- width (Optional, number):
- origin (Optional, words):
Use it from your AI
Connect the AI you already use to Trillion once. Then ask in your own words; your AI finds this tool by what it does and runs it.
Say to your AI: “Use Trillion to draw one of this person's own Trillion screens at a width, as they see it, to put beside their sites on a board or a canvas”
What your AI calls: run_tool {"tool_id":"tool:screen-picture"}