agentleFS
soilmasssoilmass/motion-design-agent/skills/orchestrator/SKILL.md

orchestrator

Skill7 starsChanged 10 months agoMITView on GitHub

Autonomous decision engine for generating complete animations from high-level descriptions. Use when you need to parse natural language animation requests, apply narrative intentionality, resolve parameters, select Disney principles, and output production-ready code with story coherence.

80 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use worker, reviewer and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • stotihvSkill

    worker

    Execute beads autonomously within a track. Handles bead-to-bead context persistence via Agent Mail, uses preferred tools from AGENTS.md, and reports progress to orchestrator.

    Also in 3 projects that use this

  • MicrosoftSkill

    reviewer

    ASP.NET Core ハンドブックの記事をレビューする。Use when: 記事をレビューする、ドキュメントを校正する、コードレビューする、PRをレビューする。技術ドキュメントのレビュー、品質チェック、校正。

    Also in 4 projects that use this

  • am-willSkill

    planner

    Create comprehensive, phased implementation plans with sprints and atomic tasks. Use when user says: "make a plan", "create a plan", "plan this out", "plan the implementation", "help me plan", "design a plan", "draft a plan", "write a plan", "outline the steps", "break this down into tasks", "what's the plan for", or any similar planning request. Also triggers on explicit "/planner" or "/plan" commands.

    Also in 4 projects that use this

  • yakatotechSkill

    advisor

    Provide improvement suggestions and advice on work content. Use for document review, efficiency suggestions, and best practices

    Also in 3 projects that use this

  • AWSSkill

    planning

    Discovers user intent and generates a structured, step-by-step plan for model customization workflows. This skill must always be activated alongside any other skill when the user's request relates to model customization — including fine-tuning, training, building, customizing, reviewing data, or getting advice on approach, regardless of domain. Do not skip this skill even if the immediate ask is narrow (e.g., reviewing data format or a single workflow step), because planning discovers the full scope of work needed. Also activate when the user wants to resume, continue, or modify an existing plan.

    Also in 3 projects that use this

  • OpenAISkill

    observability

    Vercel Observability expert guidance — Drains (logs, traces, speed insights, web analytics), Web Analytics, Speed Insights, runtime logs, custom events, OpenTelemetry integration, and monitoring dashboards. Use when instrumenting, debugging, or optimizing application performance and user experience on Vercel.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 66 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When generating animations autonomously:

    line 289
  2. Distinctive:

    Apply emotion: if specified or inferred

    line 293

“Common” is measured: kept by at least 6 in 10 of the 66 edited versions of this file we could read. “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

Motion Orchestrator

Autonomous decision engine for generating complete animations from high-level descriptions. The orchestrator parses input, applies narrative intentionality, resolves parameters via unified pipeline, selects appropriate Disney principles, and outputs production-ready code.

Orchestration Pipeline (v3.0 — Narrative + Resolution)

Input Description
       │
       ▼
┌──────────────────┐
│   Input Parser   │ → Extract: element, action, emotion, brand, platform
└────────┬─────────┘
         │
         ▼
┌──────────────────┐
│  Intent Parser   │ → What PURPOSE? (clarify/guide/confirm/warn/celebrate/reveal)
└────────┬─────────┘
         │
         ▼
┌──────────────────┐
│  Journey State   │ → Where emotionally? (curious/engaged/accomplished)
└────────┬─────────┘
         │
         ▼
┌──────────────────┐
│  Narrative Arc   │ → What story phase? (exposition/rising/climax/resolution)
└────────┬─────────┘
         │
         ▼
┌──────────────────┐
│ Scene Composer   │ → Dramaturgical structure (beats, pauses, rhythm)
└────────┬─────────┘
         │
         ▼
┌──────────────────┐
│ Attention Path   │ → Reading order (F-pattern/Z-pattern/center-out)
└────────┬─────────┘
         │
         ▼
┌──────────────────┐
│  Consequence     │ → Cause-effect chains (direct/ripple/cascade/echo)
└────────┬─────────┘
         │
         ▼
┌──────────────────┐
│   Microstate     │ → Component state machines (idle→active→success)
└────────┬─────────┘
         │
         ▼
┌──────────────────┐
│ Recipe Matcher   │ → Find matching recipe or compose new
└────────┬─────────┘
         │
         ▼
┌──────────────────┐
│Parameter Resolver│ → Merge recipe + emotion + brand + platform
└────────┬─────────┘
         │
         ▼
┌──────────────────┐
│Principle Select  │ → Which Disney principles apply?
└────────┬─────────┘
         │
         ▼
┌──────────────────┐
│Conflict Resolver │ → Handle competing requirements
└────────┬─────────┘
         │
         ▼
┌──────────────────┐
│ Code Generator   │ → Output for target platform
└────────┬─────────┘
         │
         ▼
   Complete Animation with Story

Narrative Layer (v3.0)

Intent Detection

Parse semantic purpose from the request:

KeywordsIntentSpeedIntensity
explain, show, revealclarifymeasured0.4
lead, direct, pointguidedeliberate0.5
done, saved, submittedconfirmresponsive0.6
error, danger, alertwarnurgent0.8
success, complete, wincelebrateexpressive1.0
unveil, discover, unlockrevealdramatic0.7

Journey State Detection

Determine user's emotional position:

ContextJourney StateNext States
First interactioncuriousengaged, confused
Active engagementengagedaccomplished, frustrated
Task completedaccomplisheddelighted, engaged
Multiple successesdelightedneutral

Arc Phase Selection

Map to Freytag's pyramid for UI:

DurationArc PresetPhases
<400msmicroincident → climax → resolution
400-800msshortexposition → rising → climax → resolution
800-2000msmediumall five phases
>2000msextendedfull with sub-beats

Consequence Chain Selection

PatternTimingUse Case
direct0msImmediate feedback
ripple30ms/stepRadiating effects
cascade50ms/stepSequential flow
echo200msDelayed reflection

Phase 1: Input Parser

Element Detection

KeywordsElement Type
button, btn, ctabutton
card, tile, panelcard
modal, dialog, overlaymodal
list, items, gridlist
toast, notificationtoast
menu, dropdowndropdown

Action Detection

KeywordsAction Type
click, press, tappress
hover, mouse overhover
enter, appear, showentrance
exit, leave, closeexit
error, failerror
success, completesuccess

Emotion Detection

KeywordsEmotion
bouncy, fun, playfulplayful
elegant, refined, subtleprofessional
celebratory, excitingcelebratory
calming, gentlecalming
urgent, importanturgent

Phase 2: Parameter Resolution

Resolution Order

Recipe (base) → Emotion (modifier) → Brand (constraint) → Platform (adaptation) → Bounds → Accessibility (override)

Resolution Algorithm

function resolveParameters(
  recipe: Recipe,
  emotions: WeightedEmotion[],
  brand: Brand,
  platform: Platform,
  context: Context
): ResolvedParams {
  // 1. Start with recipe base
  let duration = recipe.duration;
  let easing = recipe.easing;
  let exaggeration = recipe.exaggeration ?? 1.0;

  // 2. Blend emotions (weighted average for numeric, dominant for discrete)
  const blendedEmotion = blendEmotions(emotions);

  // 3. Apply blended emotion as modifier
  if (blendedEmotion) {
    duration *= blendedEmotion.durationMultiplier;
    easing = blendedEmotion.easing ?? easing;
    exaggeration *= blendedEmotion.exaggerationBoost ?? 1.0;
  }

  // 4. Apply brand constraints (caps, not multipliers)
  if (brand) {
    exaggeration = Math.min(exaggeration, brand.maxExaggeration);
    if (!brand.allowsSquashStretch) squashStretch = "none";
  }

  // 5. Apply platform adaptation
  duration *= platform.durationMultiplier;

  // 6. Clamp to bounds
  duration = clamp(duration, platform.minDuration, platform.maxDuration);

  // 7. Accessibility override (bypasses all above)
  if (context.reducedMotion) {
    return recipe.reducedMotion ?? DEFAULT_REDUCED_MOTION;
  }

  return { duration, easing, exaggeration, ... };
}

Platform Multipliers

PlatformDurationMaxMin
web1.0×600ms50ms
video0.6×400ms50ms
mobile1.1×500ms50ms

Resolution Checklist

  • Emotions blended before applying (prevents stacking)
  • Brand constraints enforced (caps exaggeration, disables features)
  • Platform multiplier applied
  • Duration clamped to bounds
  • Reduced motion checked last (overrides all)

Phase 3: Principle Selection

Decision Tree

Does it bounce or impact?
  Yes -> SQUASH-STRETCH
  No  -> Skip

Does it move through space?
  Yes -> ARC-MOTION
  No  -> Linear path OK

Is it a significant action?
  Yes -> ANTICIPATION (15-25% of total)
  No  -> Skip

Are there secondary elements?
  Yes -> FOLLOW-THROUGH (10-30% stagger)
  No  -> Single element

Phase 4: Conflict Resolution

Priority Rules

  1. Accessibility > Everything - Reduced motion always wins
  2. Brand > Emotion - Brand consistency trumps momentary emotion
  3. Context > Defaults - Device capability limits animations
  4. Duration Conflicts - Average within brand bounds
function resolveConflicts(params: Parameters, context: Context): Parameters {
  if (context.reducedMotion) {
    params = applyReducedMotionLimits(params);
  }
  params = clampToBrandLimits(params, context.brand);
  params = clampToDeviceLimits(params, context.device);
  return params;
}

Phase 5: Output Generation

Platform Selection

PlatformOutput Format
CSSKeyframes + transitions + media queries
GSAPTimeline with tweens
Framer MotionVariants + transitions
RemotionInterpolate functions + spring

Autonomous Generation Workflow

When generating animations autonomously:

  1. Parse the request using keyword detection
  2. Match to recipe or compose from principles
  3. Apply emotion if specified or inferred
  4. Apply brand constraints
  5. Check accessibility requirements
  6. Generate code for all platforms
  7. Include reduced motion alternative
  8. Document token references

Decision Checklist

  • Element type identified
  • Action type identified
  • Platform identified (web/video/mobile)
  • Emotion determined (explicit or inferred)
  • Emotions blended if multiple
  • Parameters resolved via resolution pipeline
  • Brand constraints applied (caps, not stacking)
  • Duration within platform bounds (50-600ms web, 50-400ms video)
  • Intent parsed (clarify/guide/confirm/warn/celebrate/reveal)
  • Journey state identified (curious/engaged/accomplished)
  • Narrative arc selected (micro/short/medium/extended)
  • Attention path planned (F-pattern/Z-pattern/center-out)
  • Consequence chain defined (direct/ripple/cascade/echo)
  • Microstate machine configured (component states + transitions)
  • Disney principles selected
  • Conflicts resolved
  • Code generated for target platform
  • Reduced motion included
  • Performance considerations noted
  • Determinism verified (same inputs = same outputs)

Related Skills

Core

  • choice-resolution/ - Unified parameter resolution pipeline
  • motion-principles/ - Disney 12 principles
  • easing/ - Easing curve selection
  • duration/ - Duration token selection
  • accessibility/ - Reduced motion requirements
  • animation-recipes/ - Recipe definitions and selection
  • emotion-mapping/ - Emotion-to-parameter mapping
  • platform-adaptation/ - Platform-specific adjustments
  • tools/ - Platform-specific code generation

Narrative (v3.0)

  • intent/ - Semantic motion purpose
  • emotional-journey/ - User emotional state machine
  • narrative-arc/ - Story structure for motion
  • dramaturgy/ - Scene composition and pacing
  • attention-choreography/ - Reading order and gaze paths
  • consequence/ - Cause-effect animation chains
  • microstate/ - Component state orchestration

See references/ for complete recipe matching tables and emotion mapping.

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.