Story 3D

A tiny 3D film from your script: toy characters act, talk and emote while a virtual director cuts the shots.

Find the engine (the sibling rudywizz-motion skill folder) and a temp folder both bash and Node agree on:

RM="${CLAUDE_SKILL_DIR}/../rudywizz-motion"
[ -f "$RM/engine/cli.mjs" ] || RM=$(ls -dt "${CLAUDE_PLUGIN_ROOT}/skills/rudywizz-motion" ~/.claude/plugins/cache/*/rudywizz-motion/*/skills/rudywizz-motion ~/.claude/skills/rudywizz-motion .claude/skills/rudywizz-motion 2>/dev/null | head -1)
TMP=$(node -p "require('os').tmpdir()")
(cd "$RM" && node -e "require.resolve('playwright')" 2>/dev/null) || (cd "$RM" && npm install --no-audit --no-fund)   # first run only

Read $RM/SKILL.md once for setup, the workflow (check stills → render → verify) and the full catalogue. Put the user's text in single quotes (write '\'' for an apostrophe inside it) so the shell never interprets it.

Render (the user's text — $ARGUMENTS — goes in --text; default "Rudy [wave, happy]: Hey! I'm Rudy. / Ada [surprised]: Wait. You're 3D? / Rudy [present, excited]: Every frame of me is code. / Ada [think, thinking]: So I can write a whole film... / Ada [celebrate, excited]: ...just by typing it?! / Rudy [dance, happy, wide]: Yep!"):

node "$RM/engine/cli.mjs" still story-3d --text '<text>' --at 8 --size 540 --out "$TMP/story-3d-check.png"   # look at it first
node "$RM/engine/cli.mjs" render story-3d --text '<text>' --out story-3d.mp4

Options (--set key=value, JSON values allowed):

• text — Script: one beat per line — Name [pose, emotion, shot]: words. * text = narration card, [wide] = camera-only beat, a number 0–1 in the tags walks there
• set — studio | room | office | street | park | night | space
• finish — vinyl | clay | brick — vinyl toys, hand-made clay (stop motion on twos) or glossy toy bricks (studded set)
• cast — Characters: [{ name, skin, hair, hairStyle (short|long|bun|curly|spiky|bob|bald), shirt, pants, shoes, accessory (none|glasses|cap|headphones|beard), accent, x (0–1), height }]
• voices — Audio files, one per spoken line in script order (null skips a line). Each line lasts as long as its audio, the mouth follows its loudness, and the render mixes them in
• dialogue — subtitles | bubbles | none
• cuts — auto (director picks shots) | wide (one locked wide shot)
• bg — Background colour (studio set / sky)
• accent — Set accent colour (rug, picture, studio floor tint)

Combine freely with --camera, --grade, --bloom, --leak, --film, --format 1:1|4:5|16:9, --audio, --alpha --out x.mov.