agentleFS
adamperlisadamperlis/video-skills/skills/narrated-product-film/SKILL.md

narrated-product-film

Skill45 starsChanged 4 days agoMITView on GitHub

Make a product film led by voiceover, with one real product surface or visual proof per spoken claim. Use for narrated launches, capability films, and scripted product stories.

The file

Files
SKILL.md

Narrated product film

Let the voice lead, but make every line earn a picture. Pair each spoken claim with one visual subject that proves or clarifies it.

Script first

Write short, speakable lines. One capability or claim per line usually makes one beat. Trace factual claims to the current product. When a feature depends on a mode or setting, name that condition in the line and show it on screen. Read the script aloud before building the timeline.

Mark the intended voice, pace, pronunciation, and pauses. Record or synthesize the voice only through a service the user has authorized and for which the voice rights are clear. When the user has Walkie connected to their assistant, the walkie-voiceover skill synthesizes the read on their own computer and returns the audio file. Keep line and word timings so a revised read can retime captions and picture without guessing.

Pair words and images

Make a beat table with the spoken line, the real UI or visual proof, the entry and exit, and any on-screen text. Use subtitles or highlighted words where they improve comprehension; do not stack the full script over a busy interface. A simple recurring layout can make different features feel like one film.

Timing should follow the voice. Let actions happen with the words that name them, and let the result settle before moving on. Keep transitions motivated by the product or story, not by the availability of an effect.

Sound and verification

Mix music below intelligible speech and use sound effects only where they support a visible action. Use owned or licensed audio. Check the export by listening and watching separately: spoken words match the script, captions match the audio, each visual matches its line, no UI control is hidden, and the closing is long enough to read. Reframe and inspect each requested ratio.

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.