agentleFS
adamperlisadamperlis/video-skills/skills/feature-video/SKILL.md

feature-video

Skill18 starsChanged 47 days agoMITView on GitHub

Make a short video that demonstrates a real software feature. Use for web or mobile app demos, UI clips, launch teasers, and social-ready feature videos.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    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
  1. Distinctive:

    Write one sentence for the feature and one for the proof: “The user does X.

    line 12
  2. Distinctive:

    Use the product's real colors, typography, and UI.

    line 14
  3. Distinctive:

    When animation timing matters, preserve the app's actual entrance, hover, and transition.

    line 25
  4. Distinctive:

    Start on the useful state or the action that leads to it.

    line 29
  5. Distinctive:

    Keep one visual focus at a time.

    line 31
  6. Distinctive:

    Use short on-screen words only when the UI alone cannot explain the result.

    line 32

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Feature video

Show the feature working in the actual product. Capture the interface first, then build the video around the clearest user action and resulting state.

Decide what the viewer must see

Write one sentence for the feature and one for the proof: “The user does X; the product visibly does Y.” Choose a short task that reaches that result without relying on narration. If the user supplied a particular feature or flow, keep it central.

Use the product's real colors, typography, and UI. A simple browser or device frame can add context. Avoid decorative captions over controls the viewer needs to read.

Capture the real interaction

  1. Inspect the working app and identify the starting state, user action, and final state.
  2. Set a suitable capture size and clear disposable test data. Never expose private accounts, keys, messages, or customer data in footage.
  3. Capture the real browser, simulator, or screen. Use a pointer or touch indicator only when it clarifies the action.
  4. Record a little time before and after the action so the cut has room to breathe.

For a web app, use the browser or recording tools available in the environment. For a mobile app, use a simulator or device capture. If only a supplied recording is available, edit that recording and state any limits in what it proves.

When animation timing matters, preserve the app's actual entrance, hover, and transition. Check the capture for dropped or repeated frames. Heavy scenes may need a deterministic frame-by-frame capture, if the project supports one.

Edit the story

  • Start on the useful state or the action that leads to it.
  • Cut waiting, loading, and setup only when removing them does not misrepresent the experience.
  • Keep one visual focus at a time. A slow camera move or crop can guide attention, but never cut off the control, label, or outcome.
  • Use short on-screen words only when the UI alone cannot explain the result.
  • End on the completed state long enough to read it.

Choose sound deliberately. A silent UI clip can work well in a feed; narration or music should add information or rhythm. Use licensed audio only.

Formats and verification

Render the formats requested by the user, commonly 16:9, 4:5, 9:16, or 1:1. Reframe each one against the real UI instead of relying on automatic center cropping. Make a poster for each format from a frame that shows the feature clearly.

Watch every final export. Check text legibility, pointer placement, transition timing, sound, and the final outcome. Confirm the files decode and report the format, duration, and any limitation in the capture. Publishing to an external service is a separate step that follows the user's instructions.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.