agentleFS
sanky369sanky369/vibe-building-skills/skills/creative/orchestrator/SKILL.md

orchestrator

Skill30 starsChanged 5 months agoMITView on GitHub

Dispatcher for the creative suite: diagnoses a multi-asset creative request with a few questions, then routes it through the right sequence of creative skills, naming the artifact each step passes to the next, and wires in the FAL.ai automation (single model fal-ai/nano-banana-pro via docs/creative_cli.py) when available. Use when the user asks for several creative deliverables at once or an outcome that implies them — 'assets for my product launch', 'set up my brand visuals', 'I need a logo, product shots, and social posts', 'a week of content', 'rebrand everything' — or when it's unclear which creative skill applies. Produces a Creative Production Plan: chosen route, ordered skill sequence with hand-off artifacts, generation checklist, and asset output map. For a single obvious asset, skip this and use the specific skill directly.

  • Reads credentials
  • Installs packages

85 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

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

  • 0

    Common

    Most of the 71 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use when the request spans multiple asset types, names an outcome rather than an asset ("launch assets", "brand refresh", "content for the …

    line 16
  2. Very specific:

    No style anchor exists and >1 asset is needed → every route below starts with skills/creative/creative-strategist.

    line 46
  3. Distinctive:

    One route: Commit with a one-line reason.

    line 112
  4. Very specific:

    No fake capability: If FAL_API_KEY is missing, say so.

    line 118

“Common” is measured: kept by at least 6 in 10 of the 71 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

Creative Orchestrator

Route multi-asset creative work through the right skills in the right order, so style gets decided once and every downstream asset inherits it. Prime rule: strategy before pixels. Never let generation start before a style anchor exists — assets generated pre-brief get regenerated. You are a dispatcher, not a lecturer: diagnose fast, pick one route, name what each step hands to the next.

When to use / when not to

Use when the request spans multiple asset types, names an outcome rather than an asset ("launch assets", "brand refresh", "content for the month"), or the right skill is ambiguous. Do NOT use for a single clear asset — route directly: one logo → skills/creative/brand-asset; one product shot → skills/creative/product-photography; one Instagram post → skills/creative/social-graphics; one generic image → skills/creative/image-generation.

Intake (diagnostic — one batch)

  1. Outcome — what event/goal are these assets for (launch, rebrand, ongoing content, storefront, video)?
  2. What exists — style guide or Creative Direction Brief? Logo/brand assets? Product photos? (Existing artifacts skip their producing step.)
  3. Deliverables & destinations — which assets, where they'll live (platforms, store, site)?
  4. Automation — is FAL_API_KEY (or FAL_KEY) set? If unknown, check: python docs/creative_cli.py test

Don't stall: with the outcome known, infer the rest from context, state assumptions, and present the plan. If automation is unavailable, the plan still runs — every step's deliverable becomes prompt specs plus runnable commands instead of generated files.

Routing table

Pick ONE route by the first rule that matches. Sequence is by dependency; each arrow names the artifact passed forward.

Decision rules

  • No style anchor exists and >1 asset is needed → every route below starts with skills/creative/creative-strategist.
  • Brand marks are among the deliverables → brand-asset runs before any skill that reuses the marks.
  • Video deliverables → stills first (they are the video's source material).
RouteTriggerSequence (artifact passed forward)
A — Brand identity from zero"new brand", "set up my brand visuals", nothing existscreative-strategist (Creative Direction Brief + style block) → brand-asset (approved logo/icons/pattern + guide entries) → social-graphics and/or product-photography (launch-ready assets using marks + style block)
B — Product launch, identity exists"launch assets", "images for my store/landing page"product-photography (winning shot set) → social-graphics (platform posts reusing the shots) → optional product-video (script/plan consuming hero stills)
C — Social campaign / content series"week of content", "posts for X", ongoing presencecreative-strategist (skip if brief exists) → social-graphics (per-platform graphics) → optional image-generation (thumbnails, misc visuals via custom CLI)
D — Rebrand / refresh"rebrand", "new look", "our visuals feel stale"creative-strategist (revised brief; diff vs old style) → brand-asset (regenerated marks) → social-graphics (announcement assets) → re-run B for product imagery if it exists
E — Video-led"product video", "presenter video", "animated demo"prerequisite stills via B or C → product-video or talking-head (plan/script) → remotion-script-writer if it's a coded Remotion video
F — Web/app design"landing page", "app design", responsive UIskills/creative/original-design (this is design, not asset generation); pull imagery needs back into B/C

Cross-cutting rule: skills/creative/image-generation is the utility skill for any asset the specialists don't cover (thumbnails, infographics, textures) — slot it wherever needed; it consumes the style block like everything else.

Automation wiring (facts, verified against the repo)

  • One model only: fal-ai/nano-banana-pro on FAL.ai. No fast/quality model variants exist.
  • Scripts at repo root: docs/fal_api.py (client), docs/creative_cli.py (CLI: product / social / brand / custom / test), docs/claude_integration.py (Python helpers incl. batch_generate_assets).
  • Setup: pip install requests; export FAL_API_KEY="..."; verify with python docs/creative_cli.py test.
  • Real knobs: --num-images 1–4, --aspect-ratio (21:9…9:16), --resolution 1K/2K/4K, custom also --format png|jpeg|webp and --web-search.
  • Outputs land in assets/product-photography/…, assets/social-graphics/…, assets/brand-assets/…, assets/<category>/… automatically.

Full reference: skills/creative/image-generation/references/automation.md — read it before writing any generation command into a plan.

Required output format

Deliver the plan, then execute it step by step (invoking each skill in order, carrying its artifact forward):

# Creative Production Plan — [outcome]

**Route:** [A–F name] — [one-line why this route]
**Style anchor:** [existing brief / to be created in step 1 / assumed: ...]
**Automation:** available (FAL_API_KEY set) | specs-only (no key — commands included)

## Sequence
| # | Skill (path) | Produces | Feeds forward |
| 1 | skills/creative/... | [artifact] | [what step 2 consumes] |
| 2 | ... | ... | ... |

## Asset map
| Asset | Type | Aspect ratio | Variants | Output path |
| [name] | product/social/brand/custom | [ratio] | [n] | assets/.../ |

## Checkpoints
- After step [n]: user picks [what] before [next step] starts.

## First action
[The single next thing you will do — usually "run step 1 now".]

Guardrails

  • One route. Commit with a one-line reason; offer at most one alternative if it's genuinely close. No route menus.
  • Never generate before the style anchor exists (or is explicitly waived for a throwaway).
  • User checkpoints at selection points — after logo concepts, after hero-shot variants — not after every image.
  • No fake capability. If FAL_API_KEY is missing, say so; deliverables are prompt specs + commands. Never report images that weren't generated.
  • Don't re-produce what exists. An existing brief/logo/shot set skips its step; consume it instead.
  • Facts stay verified. Model ID, script paths, and CLI flags come from the automation reference — never from memory.

Integration

Feeds this skill: skills/product-strategy/foundation-sprint (positioning and differentiators sharpen the brief in step 1 of routes A/D). Consumes its output: every skills/creative/* skill listed in the routing table — the plan tells each one what artifact it receives and what it must hand back.

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.