silent-product-demo
Skill45 starsChanged 3 days agoMITView on GitHub
Produce a designed, silent product demo from real UI with camera moves and concise on-screen words. Use for feature launches, announcements, and social clips that should work without narration.
In this repository11 other files it gives its agents.
- brand-filmSkill
- brand-grid-videoSkill
- creator-videoSkill
- feature-videoSkill
- narrated-product-filmSkill
- social-video-hooksSkill
- ui-component-clipSkill
- ui-tutorial-videoSkill
- video-brand-systemSkill
- video-thumbnailsSkill
- walkie-voiceoverSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 14
Use the real app as the authority.
- Distinctive: line 19
Place a cursor only for a decision.
- Distinctive: line 20
Keep each beat focused on one subject.
- Distinctive: line 21
Use the product's motion timing and visual tokens.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Silent product demo
Show one feature being used and the result it creates. The interface, cursor, and camera carry the story. On-screen text may name a beat or clarify an outcome, but it should not narrate every click.
Plan the sequence
Identify the real starting state, the user's decision, the app's response, and the final outcome. Make a short storyboard with time, screen state, pointer action, camera focus, and any text. Start with something legible and meaningful on frame one.
Use the real app as the authority. Capture it directly, or build a frame-accurate port from its UI source when a designed camera or extreme zoom requires one. Preserve actual labels, layout, and interaction states. Do not invent controls, results, or supported integrations.
Direct the motion
- Move the camera to make the current action easier to see, then let it settle before the next action.
- Place a cursor only for a decision. Confirm its tip touches the target when the pressed state appears.
- Keep each beat focused on one subject. Remove old props or text when the story moves on.
- Use the product's motion timing and visual tokens. In frame-based rendering, calculate animation from frame or time explicitly so every render is deterministic.
- Let transitions connect related states. A direct cut is fine when continuity adds no information.
Render stills at the first frame, each action, every transition midpoint, and the final frame before committing to full exports. Then watch the rendered files muted. Check that the feature is understandable, text remains readable at phone size, no interaction misses, and each aspect ratio preserves the outcome. Make a cover for each requested format.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

