Motion Reel logo

Motion Reel

Community
Daily-AC
motion-reel

Direct and render polished short motion-graphics reels, brand films, and kinetic showcases in code. Use for a punchy product or identity reel, or when existing animation feels static, sparse, incoherent, or unfinished. Includes deterministic rendering and synthesized sound. Not a live-action generator or long-form narrated-video workflow.

Overview

PublisherDaily-AC
Repositorymotion-reel
Skill namemotion-reel
Stars
5
Forks
1
Bundled files
21
LicenseMIT
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.

  • 21 bundled files

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

  • Open source

    Published by Daily-AC on GitHub. Read the source before you install it.

Installation

Install the Motion Reel 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/Daily-AC/motion-reel/tree/main/skills/motion-reel
  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/Daily-AC/motion-reel.git /tmp/motion-reel
mkdir -p .claude/skills
cp -r /tmp/motion-reel/skills/motion-reel .claude/skills/motion-reel
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Motion Reel 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 Motion Reel 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 Motion Reel 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.

Motion Reel

Deliver a finished film with a coherent identity, legible motion and a tested local rebuild. Use the included engine to remove rendering friction; direct the film for the user's subject rather than merely renaming its demo.

Start from the brief

For an existing film, read its sources and accepted decisions, then change the requested dimension. Preserve approved motion when feedback concerns visual density; preserve the accepted logo when changing atmosphere. Keep previous delivered renders.

For a new film, infer sensible defaults and state the decisions that matter. Ask only about missing information that changes the result and cannot be inferred. A request for the finished video authorizes local implementation, previews and rendering; it does not authorize paid generation or publication. Do not insert mandatory approval gates when the user delegated direction.

Choose the audience, one message, duration, aspect ratio, identity and audio role. Record a short treatment. At 15 seconds, let behavior demonstrate the capability instead of squeezing in a narration-sized explanation.

Direct before decorating

Read creative-direction.md for substantial creation or revision. Make four decisions:

  1. A visual premise. Name what transforms and why. “A thought becomes a constructed world” yields better decisions than “futuristic and epic.”
  2. A handoff. For each cut, name the shape, trajectory, color or sound carried forward. Match actual boundary positions and scale; shared geometry is more reliable than similar drawings.
  3. A density rhythm. Assign dense passages and quiet recognition moments. Sparse scenes may need their environment to respond to the subject, not particles everywhere.
  4. A return. Let an opening element earn its place in the close. The first frame should already communicate; the last hold should stay readable.

If ideation stalls, run scripts/creative_seed.py. Its random groups are associations, not instructions or quality scores. Select only associations that serve the film and record their visible consequences.

Build from the working starter

Resolve bundled paths from this skill directory, not a hardcoded agent home or the current project.

sh
python3 scripts/new_project.py /path/to/new-film --brand NOVA
cd /path/to/new-film
node render.mjs doctor
python3 build.py
node --check check.js
node render.mjs stills 0,0.8,4.6,8.6,10.6,14.3 --samples 8

The initializer copies a self-contained, editable 15-second example, fonts, renderer and verification tools. No API key or video-generation service is required. Read production.md before changing the project contract or diagnosing rendering.

  • Edit project.json for identity, palette and timing; author shots in scenes.js and sound in score.js. For a different shot count/order, also update SCENES, SHOT, LABELS, HITS, HUD_TONE and hud in engine.html; see the production guide.
  • Change the demo's visual argument and relevant shot behavior for a new subject. Eight shots, blue and 128 BPM are examples, not requirements.
  • Use actual logo assets for an existing brand. Preserve silhouette and internal spacing while adding depth. Never present the starter's original generic symbol as an official logo.
  • Keep scene state a pure function of time. Seed randomness at initialization. Do not integrate from prior frames or fetch assets during capture.
  • Select the shot from the frame-center time. All shutter samples of that frame belong to that shot. The compositor averages samples in linear light using RGBA16F.
  • Share impact timing between music and movement. Use a sonic motif sparingly. Loudness checks cannot judge musical taste.
  • Recompose for another aspect ratio; a landscape crop is not an authored portrait cut.

Iterate on the actual defect

FeedbackInspectUseful change
Weak motionFull action at 15–20 fps, not only its midpointPreparation, fast release, overshoot, scale change or spatial reconstruction
Empty frameHow the surroundings respond to the subjectMotivated light, shadow, far wire echo, construction plane or linked secondary pieces
Incoherent filmPalette, material, repeated motifs, cut endpointsCarry one family of shapes/light/sound and the real brand identity
ClutterCompeting focal points and illegible copyRemove the least meaningful layer; keep type and the close clear
Wrong-looking effectMoving strip and the effect's sourceRepair the relationship or delete the effect; a fake reflection can be worse than none

Read case-study.md for demonstrated revisions and rejected approaches. Do not turn one successful film into a fixed recipe for every subject.

Render, inspect, deliver

sh
node render.mjs render --out out/reel.mp4 --samples 16
python3 verify.py out/reel.mp4
uv run --with pillow python review.py out/reel.mp4 --times 0,0.8,4.6,8.6,10.6,14.98
uv run --with pillow python review.py out/reel.mp4 --strip 7.65:9.20:20

Read review.md for encoded-frame checks, motion strips, audio evidence and review handoffs. Use --scale 2 for a native 4K master when needed, after the direction survives review.

Separate technical validity, visual acceptance, and audience response. None proves the others. Do not claim aesthetic quality from frame counts or an agent's score.

For a substantial film, use an independent reviewing agent when available and allowed by the host. Give it the brief and fixed encoded artifact, not the desired verdict. Otherwise perform an explicit second review and disclose that it was not independent. Fix specific findings and recheck changed intervals rather than endlessly rescanning unchanged work.

Deliver the playable MP4, source/rebuild command and a useful poster. Report actual duration, material limits, and whether sound was only measured or also heard. Publish within the user's authorization. Read credits-and-assets.md before sharing third-party assets.

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 Motion Reel AI skill do?

Direct and render polished short motion-graphics reels, brand films, and kinetic showcases in code. Use for a punchy product or identity reel, or when existing animation feels static, sparse, incoherent, or unfinished. Includes deterministic rendering and synthesized sound. Not a live-action generator or long-form narrated-video workflow.

Why use Motion Reel on TypingMind?

Because you install it once and use it with any model. Motion Reel 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 Motion Reel in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Daily-AC/motion-reel/tree/main/skills/motion-reel. 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 Motion Reel?

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 Motion Reel?

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

Is the Motion Reel AI skill free?

Yes. It is published on GitHub by Daily-AC under the MIT license. 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 👇