agentleFS
tamtree-aitamtree-ai/stick-stage/skills/story-episode/SKILL.md

story-episode

Skill2 starsChanged 15 days agoMITView on GitHub

Make a narrated 16:9 cartoon story with StickStage's flat shape cast (Tofu, Bobo, Pim, Mabel, Otto, Fennel, Biscuit, Wren, Blip) — an idea to a checked, voiced MP4. Covers the story format (style "story", narrator, scenes with described places), new characters written as shape specs, drawn props for missing objects, the missing-asset report, and the review loop. Use when asked for a story, an episode, a bedtime or kids' cartoon, a Hey Duggee–style video, a new shape character, or "make a cartoon about X".

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 2 are distinctive, and 5 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.

  • 5

    Very specific

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

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

    Write it: Run pnpm story prompt "<idea>" --seconds=<n> and follow that prompt yourself (you are the writer), or use --model with the user's …

    line 14
  2. Very specific:

    Land it: Run pnpm story draft <id> --reply=<file>.

    line 18
  3. Very specific:

    Voice and check: Run pnpm voice:tts <id>, then pnpm check <id>.

    line 22
  4. Very specific:

    Render and review as video: Run pnpm render <id>. Pull frames with ffmpeg -i out/<id>.mp4 -vf …

    line 25
  5. Distinctive:

    Don't name a prop, pose or set that isn't in the catalog without drawing or describing it.

    line 34
  6. Very specific:

    Don't hand-draw a new character in code when a spec will do.

    line 35
  7. Distinctive:

    Don't copy an existing show's characters.

    line 37

≈ “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.

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

Story episodes

Guide: docs/stories.md. Plan and history: ../planning/2026-10-24-shape-toon-stories/ and the "shape tier" entries in DECISIONS.md. Worked examples: public/skits/lost-balloon/, public/skits/biscuit-shell/.

Steps

  1. Write it. Run pnpm story prompt "<idea>" --seconds=<n> and follow that prompt yourself (you are the writer), or use --model with the user's own model. Rules that matter: style: "story", a narrator, 2–4 scenes each with its own place, at most 4 cast, lines under 15 words, and a mark in each scene for anyone who speaks there (the skit-level mark may be offstage).
  2. Land it. Run pnpm story draft <id> --reply=<file>. Read the missing-asset report and fix each row by its rung: near (use the suggestion), spec (add the character to characters), place (describe it), draw (pnpm story prop <id> "<noun>" --reply=<file>, drawn props only, ±0.6 bounds), cheat (narrate it, cut away, hold a reaction).
  3. Voice and check. Run pnpm voice:tts <id>, then pnpm check <id>. Fix every error: speaker-offstage (give a mark), emotion-closeups (move the expression early in the line), contrast (a shape figure the colour of the place: change the place or the colour).
  4. Render and review as video. Run pnpm render <id>. Pull frames with ffmpeg -i out/<id>.mp4 -vf "select='not(mod(n\,90))',scale=640:-1,tile=3x5" -frames:v 1 out/<id>-frames.png and look: everyone on stage who speaks, nobody floating, props in paws, captions clear of faces.
  5. Hand back. Give the MP4 path and the frame sheet. Nothing is approved or posted. New characters or props promoted to src/data/ change the catalog hash, so say t-shoot needs a re-vendor.

Don't

  • Don't name a prop, pose or set that isn't in the catalog without drawing or describing it.
  • Don't hand-draw a new character in code when a spec will do. New part kinds go in src/engine/rig/shape/ and need the owner's OK on ShapeLab.
  • Don't copy an existing show's characters. Original designs only (AGENTS.md rule 6).

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.