compose-lookbook
Arrange several supplied outfit photographs into one intentional lookbook page with an editorial grid, hierarchy, spacing, and optional exact copy. Use after outfit images exist; do not invent outfits
SKILL.mdskills/compose-lookbook/SKILL.md
GitHubCompose Lookbook
Catalog
- Display name: Compose Lookbook
- Categories: Fashion, Social Media
- Status: Ready
- Route label: GPT Image 2 editorial layout
- Tone: plum
- Short description: Arrange several outfit photos into one editorial page.
- Similar skills: photo-collage, create-social-carousel, create-image-deck
What this skill does
- Places multiple supplied looks into one ordered editorial page.
- Preserves each outfit and person while controlling crop, whitespace, captions, and reading order.
How to use
Provide two to five outfit photos, ratio, visual direction, and exact copy.
Compose a lookbook
Arrange these four outfit photos as one portrait lookbook page titled CITY LAYERS.
Workflow
- Inventory each look; lock order, crop-safe regions, exact copy, and ratio.
- Choose hero-plus-support, equal grid, staggered editorial, or contact-sheet rhythm.
- Use
gpt-imagev1.0, GPT Image 2 Medium, all references, requested ratio,batchCount: 1. - Keep garments, faces, colors, and proportions source-faithful; render exact text once.
- Verify every photo is visible, order and margins are clear, copy is exact, and no extra look appears.
When this Skill selects GPT Image 2, use
quality: "medium"andimageSize: "2K"by default; change either only when the user explicitly requests another tier.
Every GPT Image 2 call in this Skill defaults to Medium at 2K; use another tier only when the user explicitly requests it.
User-facing output
- Media type: One lookbook page image
- Default quantity: 1
- Content per image: All supplied looks in distinct regions
- Default layout: Editorial multi-photo page
- Model policy: GPT Image 2 Medium
- Downstream use: Fashion decks, social posts, and line sheets