agentleFS
VisualxIntelligenceVisualxIntelligence/motion-tools-and-skills/skills/claude-motion/claude-motion-review-loop/SKILL.md

claude-motion-review-loop

Skill9 starsChanged 22 days agoMITView on GitHub

Visual self-review loop for motion graphics — render a draft, build contact sheets at every timeline beat, look at them, critique against a checklist, fix, repeat. Use after any change to a video's visuals or timing, and before calling a video done.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Only palette colors from C.

    line 41
  2. Distinctive:

    When you finish, say what you checked, what you fixed, and which sheet shows it.

    line 50

≈ “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

Review loop

You can't watch the video, but you can look at frames. Never report a video as done without looking at contact sheets of the current render.

Loop

  1. npm run draft — half-resolution render to out/draft.mp4 (fast).
  2. npm run sheet — one frame per timeline beat, 0.3 s after it so motion has settled, tiled 4×2 into out/sheets/beats-N.png. Each tile is labelled with its timestamp.
  3. Open every sheet image and go through the checklist below.
  4. Zoom in on anything suspicious:
    • npm run sheet -- --from 3 --to 5 --every 0.2 to scrub a transition,
    • npm run sheet -- 3.7 3.75 3.8 for exact moments,
    • npm run still -- out/still.png --frame=222 for one full-resolution frame.
  5. Fix, then go back to step 1. Stop when a full pass finds nothing.
  6. Final: npm run build, then npm run sheet -- --video out/effort.mp4 and npm run wave.

Checklist

Composition

  • One focal point per frame. If two things compete, one of them waits.
  • Nothing collides or overlaps unintentionally, especially during transitions where an exiting and an entering element share space.
  • Safe area: nothing important within ~60 px of an edge.
  • Frame 0 is a real image, not black (it's the X thumbnail).

Type

  • Nothing clipped by a mask, container or the frame when it shouldn't be.
  • Key message readable when the sheet tile is shrunk to phone width.
  • Text isn't moving while it's meant to be read.

Timing

  • Each readable state holds long enough (see motion-design §4). If a tile mid-hold shows text still animating, the hold is too short.
  • Scenes overlap instead of cutting to empty frames.
  • Exits are faster than entrances.

Consistency

  • Only palette colors from C; the accent marks the point of the frame.
  • Same element looks the same in every scene it appears in.

Sound (after npm run wave)

  • Transients sit on beat lines in out/wave.png.
  • −14 LUFS ±1, true peak ≤ −1 dBFS.

Report

When you finish, say what you checked, what you fixed, and which sheet shows it. Don't claim a fix you haven't re-rendered and looked at.

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.