Zach D Films logo

Zach D Films

Community
artemnovitckii
zach-d-films

Makes a 3D cartoon "what if" explainer short (9:16) in the style of Zack D. Films-type videos, entirely in code with three.js. Procedural cartoon people, props and effects, a blue studio sky, slow-motion impacts, a red force-ripple overlay, quick whip cuts and 1 to 3 word captions timed to the voiceover. Use when the user runs /zach-d-films or asks for a 3D cartoon explainer, a "could X stop Y" or "what if" short with characters and props.

Overview

Publisherartemnovitckii
Repositoryclaude-animation-skills
Skill namezach-d-films
Stars
147
Forks
21
Bundled files
10
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 10 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by artemnovitckii on GitHub. Read the source before you install it.

Installation

Install the Zach D Films AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/artemnovitckii/claude-animation-skills/tree/main/skills/zach-d-films
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/artemnovitckii/claude-animation-skills.git /tmp/claude-animation-skills
mkdir -p .claude/skills
cp -r /tmp/claude-animation-skills/skills/zach-d-films .claude/skills/zach-d-films
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Zach D Films in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Zach D Films on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Zach D Films is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

/zach-d-films

┌──────────────────────┐
│      blue sky        │  procedural 3D people and props, soft shadows
│    ◕‿◕   ⛨           │  depth of field, bloom, light grain
│   /▓▓▓\  ⟶  •        │  slow-motion impacts, sparks, a red force ripple
│     ▓▓               │
│     arrows           │  1-3 word captions, white bold, lower third
└──────────────────────┘

A worked example comes with it: engine/ is a finished 29.5 s short, "Could ancient armour stop a bullet?" (9 shots: a gun shot, a slow-motion bullet flattening on a vest, a sword on chain mail, a mace on plate with a red ripple, an archer, an arrow and a scream, a bullet through two stands). It ships without its voiceover, so you bring a voice. Not affiliated with Zack D. Films; "the style" means the format: a question, a misconception, a "But" twist, building examples and a payoff.

First use

This skill's folder is the "Base directory" Claude Code shows when it loads it.

bash
SKILL_DIR="<base directory of this skill>"
KIT=$(python3 -c 'import os,sys;print(os.path.realpath(os.path.join(sys.argv[1],"..","..","kit")))' "$SKILL_DIR")
[ -d "$KIT/node_modules/playwright" ] || bash "$KIT/setup.sh"
[ -x "$KIT/.venv/bin/python" ] && [ -d "$KIT/models" ] || bash "$KIT/setup.sh" --voice   # word timings + free voice, ~1 GB once
PY="$KIT/.venv/bin/python"

Needs ffmpeg, Node and Python 3.9+. It renders with WebGL in headless Chrome: about 0.2 to 0.5 s a frame on a Mac, a few minutes for a short. On a machine without a usable GPU, set GL=swiftshader (much slower).

The engine

FileWhat
src/shots.jsThe edit. CAPTIONS ([start, text, end?]), FADES (dips to black), DURATION, one function per shot and the SHOTS list ({start, end, build}). A shot's build({makeStage}) returns {scene, camera, update(t)}; update gets time inside the shot and may return {dof, bloom, exposure}.
src/human.jsmakeHuman(outfit): a poseable cartoon person ('vest', 'chainmail', 'plate', 'knight', 'archer'). setPose({joint: [x, y, z]}) for pelvis spine chest neck head lSh lEl lWr rSh rEl rWr lHip lKn lAn rHip rKn rAn (+ drop), setMouth(0..1), lerpPose(a, b, t), standArmor(kind).
src/props.jspistol, gunArm, bullet (with a flatten morph), sword, shield, mace, arrow, bow, stand, glowSprite, sparks, streak, rippleOverlay (the red force ripple).
src/anim.jskf keyframes, easings, range, cam(fov), place(camera, pos, look), shake, ikArm (2-bone arm IK so hands can grab things).
src/textures.js, src/materials.jsCanvas-made steel scratches, chain-mail rings, wood, fabric, skin, and the materials built on them.
src/main.jsRenderer, sky, lighting, depth of field, bloom, grade, and window.renderAt(t).

Steps

  1. Script. Use the user's script, or draft one in the format: "If you…" hook, "some people think…", a "But" twist, examples that build, the mechanism, a payoff that answers the hook. About 2.5 words a second (70 to 90 words for 30 s). Check the facts.
  2. Project. Copy the engine to a new folder (<project> must not exist yet) and link the shared packages:
    bash
    cp -R "$SKILL_DIR/engine" <project> && ln -s "$KIT/node_modules" <project>/node_modules
    Write the script into <project>/script.json as a list of sentences.
  3. Voice. Best is the user's own recording; convert it with ffmpeg -i <take> -ac 1 -ar 44100 <project>/audio/vo.wav. Otherwise, from the project, the free Kokoro voice: "$PY" "$SKILL_DIR/../stick-figure/engine/tools/tts_kokoro.py" [voice] [speed] (defaults am_michael 1.12; also af_heart, am_adam, bm_george) writes audio/vo_kokoro.wav; copy it to audio/vo.wav. For the browser preview only, also save it as audio.m4a (main.js plays that file).
  4. Captions. "$PY" "$KIT/words.py" audio/vo.wav script.json writes audio/vo.words.json (the voiceover's name with .words.json). Then python3 captions.py audio/vo.words.json prints CAPTIONS and DURATION to paste over the ones in src/shots.js. Set FADES around your "But" cut (the example dips at 6.2 s), or [].
  5. Shots. Plan a shot every 2 to 4 s (8 to 10 for 30 s, 3 for 10 s), cutting on caption beats, the mechanism shot longest. Write one function per shot by copying the closest example, change the props, poses and camera, and list them in SHOTS with start and end times (delete the example functions you don't use).
    • update(t) must depend only on t: a render can start in the middle of a shot. Example shot 7 keeps state between frames and needs a workaround, so don't copy that pattern.
    • Useful helpers inside shots.js: standPose, STAND, handsOnHips, surfaceHit (where a projectile meets a mesh). In props.js: box, shadow. rippleOverlay needs one mesh to sit on; for a target made of parts, give it an invisible disc (with shadows off).
    • New props go in props.js as small functions that return a THREE.Group. sparks are glowing points, so splinters or dust need their own meshes.
    • At 9:16 the frame is narrow (about 19° wide at fov 34). To keep a projectile in frame, ride behind it.
    • Keep faces and key props out of the caption band (y about 950 to 1030 of 1280) and away from the frame edges.
  6. Stills. From the project: node render.mjs stills 1.5,6.8,14 stills. Look at every shot before a full render.
  7. Render. node render.mjs video silent.mp4 30 (about 0.2 to 0.5 s a frame), then add the voice, level it and scale to 1080x1920:
    bash
    ffmpeg -i silent.mp4 -i audio/vo.wav -map 0:v -map 1:a -vf scale=1080:1920:flags=lanczos -c:v libx264 -crf 17 -pix_fmt yuv420p -af "apad,loudnorm=I=-14:TP=-1.5" -c:a aac -b:a 192k -shortest <name>.mp4
  8. Hand over the mp4 path.

For a live preview, serve the project (python3 -m http.server 8000) and open index.html (click to play with audio.m4a), or index.html?t=14.5 for one frame.

Style

  • Bright blue studio sky, soft daylight, metal that reads as metal. Cartoon people are simple shapes with big expressive faces; keep them in proportion with the props.
  • Slow motion on the impact, a short camera shake, sparks, bloom on hot things. The red ripple shows force spreading.
  • Quick cuts on caption beats; one dip to black on the "But" turn. Captions are white, bold, 1 to 3 words.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Zach D Films AI skill do?

Makes a 3D cartoon "what if" explainer short (9:16) in the style of Zack D. Films-type videos, entirely in code with three.js. Procedural cartoon people, props and effects, a blue studio sky, slow-motion impacts, a red force-ripple overlay, quick whip cuts and 1 to 3 word captions timed to the voiceover. Use when the user runs /zach-d-films or asks for a 3D cartoon explainer, a "could X stop Y" or "what if" short with characters and props.

Why use Zach D Films on TypingMind?

Because you install it once and use it with any model. Zach D Films is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Zach D Films in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/artemnovitckii/claude-animation-skills/tree/main/skills/zach-d-films. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Zach D Films?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Zach D Films?

As many as you like. As long as a model supports skills, you can use Zach D Films with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Zach D Films AI skill free?

It is published on GitHub by artemnovitckii. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇