Documentary Skill

The Documentary Skill builds a scroll-driven visual narrative from a single prompt. As the reader scrolls, background media transitions while the text tells the story — the same format used by major news outlets for immersive longform pieces.

How to trigger it

Ask for an immersive story or documentary in natural language, in any language:

• “Create an immersive story about the Amazon rainforest”
• “Build a scroll documentary on the Apollo missions”
• “Make a longform piece on the history of jazz”
• “Racconta la storia della Via della Seta in formato documentario”

What you get

A self-contained HTML page where scroll drives the experience. The documentary includes:

• A strong visual opening — a full-bleed image or video with a short hook
• Narrative sections where text floats over or beside full-bleed media
• Editorial elements within sections — timelines, before/after comparisons, annotated images, photo galleries, pull quotes
• A closing that resolves the story — a final image, a reflection, or a key takeaway

Tips for best results

• Tell the story arc upfront. Opening scene, key chapters, and how it ends — the AI structures the pacing from there.
• Provide your own images and videos. This format lives or dies by its visuals. The AI will ask for them per section and use labelled placeholders until you supply them.
• Think in beats, not paragraphs. Each scroll step is one moment in the story — short, focused, visually anchored.
• Iterate freely. Ask to add a chapter, swap a photo, insert a timeline — the AI updates the page in place.