agentleFS
Sign inStart free
mehbylabsmehbylabs/creator-skills/skills/creator-video/SKILL.md

creator-video

Skill7 starsChanged 16 days agoMITView on GitHub

Generate short video clips with whatever the user has — Higgsfield CLI (Seedance, Kling, Veo), Google Gemini API (Gemini Omni Flash), Google Flow or the Gemini web app via computer use — and fall back to animated stills (Ken Burns) when no video provider exists. Covers text-to-video, image-to-video, first/last-frame transitions, reference-guided clips, reframing to 9:16/16:9, and burned captions. Use when the user asks to make a video or clip, animate an image, image-to-video, B-roll, a Reel/Short/TikTok clip, a transition, or to reframe or caption a clip. Not for narrated multi-scene explainers (creator-explainer) or ads (creator-ad).

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. adamperlis/video-skills★ 18same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository10 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    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:

    Verify on disk (duration > 0, expected ratio) before reporting success.

    line 67
  2. Distinctive:

    Keep one "look sentence" across clips of the same project.

    line 69

≈ “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

Creator Video

Step 1 — Plan the clip

  • What moves — one subject, one action, one camera move per clip (references/video-prompting.md).
  • Ratio — Shorts/Reels/TikTok 9:16, everything else 16:9.
  • Duration — default 8 s; longer stories become several clips joined later.
  • Start frame? — if the look matters (brand, character, product), make the first frame as an image first (creator-image or creator.py image), approve it, then animate it with --start. This is the most reliable way to control the result on every provider.
  • End frame / references — --end for transitions and reveals, --ref for identity.
  • Resolution — 720p default; 1080p when asked.

Step 2 — Generate

python3 scripts/creator.py video --prompt "<motion prompt>" --out ./out/<slug>.mp4 --ratio 16:9 --duration 8 [--start first.png] [--end last.png] [--ref ref.png] [--resolution 1080p]

Video takes 1–10 minutes. Run it in the background when possible and don't start a second generation meanwhile.

Step 3 — Handle the result

statusDo
okpython3 scripts/media.py probe <path>; extract a frame and look at it; report path, length, provider
needs_browserFollow references/browser-flow-browser.md (or gemini) with the request fields; then creator.py adopt --src <download> --out <path>
blockedContent refused; revise with the user, don't route around it
failed and no video provider at allUse the stills fallback below and tell the user which provider would unlock real motion

Stills fallback (always available)

  1. Make 1–3 stills with creator.py image in the target ratio.
  2. python3 scripts/media.py still --image s1.png --out clip.mp4 --duration 6 --ratio 16:9 --motion in (vary --motion per still; join several with media.py assemble).

Step 4 — Finish (optional)

  • Reframe: media.py reframe --video clip.mp4 --out clip-9x16.mp4 --ratio 9:16 --mode blur (crop when the subject is centered).
  • Captions: write an .srt, then media.py captions --video clip.mp4 --srt subs.srt --out final.mp4.
  • Sound: creator.py sfx / music and mix with media.py assemble (single scene).

Rules

  • Verify on disk (duration > 0, expected ratio) before reporting success.
  • Mention when a route spends credits per clip (Flow, Higgsfield) if the user asks for many.
  • Keep one "look sentence" across clips of the same project.

References

  • references/video-prompting.md — shot formula, frames, references
  • references/providers.md — video order and fallback
  • references/media-ops.md — still, reframe, captions, assemble
  • references/browser-flow-browser.md, references/browser-gemini-browser.md

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.

Check your setup, free