Write the listing from the photos
Rung 1 of 10 · RTO — Role · Task · Output
1 · Learn the move · RTO — Role · Task · Output
A multimodal model can draft the description from the photos themselves — but it will confidently call laminate 'granite' or add a fireplace it half-sees. The skill is forcing it to describe only what's unmistakably visible and mark everything else UNKNOWN. Attach the photos, name the rooms if you can, and make it earn every claim.
You are a listing writer. Draft MLS remarks from these attached photos. Describe only what's clearly visible; mark anything uncertain UNKNOWN. Nothing about who should live here. [attach the listing photos]
2 · Your turn — you write the prompt
You shot 12 photos of a 3-bed you're listing tomorrow and haven't written a word. You want a first-draft description built from the photos themselves — without the AI inventing finishes it can't actually see. Write a prompt that drafts the remarks from the images, describes only what's visible, and marks guesses UNKNOWN.
Remember: the AI sees only your prompt — not this page. If the situation isn't in your prompt, it doesn't exist.
Optional — these shape the output when you run your prompt below, not your score.