skills/web-design/brand-visuals-ai/SKILL.md
Translates a brand identity into an AI visual production kit — a Visual DNA spec, prompt directions, and asset workflow for creating brand visuals with generative AI (text-to-image, image-to-image, image-to-video). Use this skill whenever the user is designing a brand and wants to create visuals with AI tools. Triggers include: "design visuals for my brand", "create AI prompts for my brand", "I want to design [product/brand] with AI", "help me make brand assets", "generate mockups for my product", "design a brand for X and create visuals", or any request combining brand work with generative AI image or video creation. Always use this skill BEFORE writing any image or video prompt for a brand project — it keeps everything consistent and on-brand. Works from a full brand brief OR a rough idea.
npx skillsauth add michailbul/laniameda-skills brand-visuals-aiInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Turns a brand into an AI-ready visual production system. Output: a Visual DNA spec and prompt directions the creator can take into any generative AI tool.
Goal: lead the creator, propose directions, maintain consistency.
Before anything else, ask these three things if you don't already know them:
These three answers drive everything. Don't skip them.
Check if a saved [brand-name]-brand skill exists. If yes, load it and skip to Step 3.
If no brand exists, collect what you need:
| Field | Example | |---|---| | Brand name + what it is | PawHaus — premium streetwear for French Bulldogs | | Target customer | French Bulldog owners, 25–40, urban, fashion-conscious | | Brand feel (3–5 words) | Bold, urban, playful, luxury | | Reference example to steal from | Supreme meets luxury pet | | Assets needed | See checklist below |
If you have a URL or uploaded images — analyze them first, only ask for gaps.
If no brand system exists yet, offer to run
brand-guidelines-builderfirst. For quick projects, proceed with what you have and fill gaps with judgment.
Use this as the canonical list. Go through it with the user and mark what's needed. Not everything is required for every project — prioritize what matters for the stage.
Distill the brand into a Visual DNA block before writing any prompt. This is the consistency anchor — every prompt inherits from it.
VISUAL DNA: [Brand Name]
────────────────────────────────────────────────
MOOD: [2–3 words — e.g. "cinematic, raw, urban"]
PALETTE: Primary #XXXXXX | Secondary #XXXXXX | Accent #XXXXXX
SURFACES: [e.g. "worn leather, raw concrete, brushed metal"]
LIGHT: [what the light does — e.g. "warm side light raking across
the surface, deep shadows on the opposite side, no fill"]
COMPOSITION: [e.g. "tight crops, negative space, centered subjects"]
REFERENCES: [2–3 visual touchstones — films, brands, photographers]
AVOID: [e.g. "stock photo feel, white backgrounds, cartoon style"]
────────────────────────────────────────────────
Lighting rule: describe what light does, never what it is. "A shaft of warm amber light falls across the left side, everything else drops into near-black" — not "there is a softbox on the left."
For each asset the user needs, write a prompt direction — a starting point to iterate from.
[SUBJECT — what / who, doing what]
[ENVIRONMENT — where, what surfaces, background]
[LIGHT — what it does, direction, quality, temperature]
[MOOD / TEXTURE — material feel, atmosphere]
[COMPOSITION — framing, crop, angle]
[OUTPUT STYLE — photorealistic / editorial / cinematic / etc.]
ASSET: [Name]
APPROACH: [text-to-image / image-to-image / image-to-video]
PROMPT DIRECTION:
[Plain descriptive language. Visual DNA elements baked in.
Describe the scene as if painting a picture.]
VARIATIONS TO TRY:
• [Different angle, lighting, or crop]
• [Push a different personality aspect]
WATCH FOR:
• [1 common failure mode + fix]
Logo / Logomark Generate in Ideogram or Recraft.ai — these handle typography and graphic marks better than general image models. Try multiple directions (wordmark only, icon only, combined) and iterate on the strongest.
Colour palette Use coolors.co to explore and lock the palette before generating anything. Pull hex values from there into the Visual DNA block. Don't guess colors — lock them first, then paste into prompts.
Product hero shot Always generate the product isolated first (neutral background, clean light, straight-on angle). That image becomes the reference input for all lifestyle scenes.
Lifestyle / scene shots Image-to-image from the isolated product shot. Describe the new environment and light — the product stays consistent because it's the input.
Promo video Generate a strong still keyframe first. That keyframe is your image-to-video input. Never go to video from a text prompt alone — always start with a locked image.
Text on images (social, web) Generate the background / scene first. Add text in a design tool afterward. Don't fight text rendering in image models.
Two things keep all assets feeling like one brand:
1. Visual DNA as prompt anchor Every prompt pulls from the same mood, palette, light behavior, surfaces, and avoids. Paste the relevant lines directly — don't rewrite them per asset.
2. Hero reference images After generating the first 2–3 hero images, pick the one that feels most on-brand. Use it as the style reference / img2img seed for everything that follows. Everything else should feel like it lives in the same world as that image.
# AI Visual Production Kit: [Brand Name]
## Visual DNA
[Block from Step 4]
## Asset Checklist
[Marked-up list from Step 3]
## Prompt Directions
[All asset blocks from Step 5]
## Consistency Strategy
[2-point summary from Step 6]
After delivering:
.md file[brand]-visuals skill for the projectdesign-inspiration skill before finalizing visual direction.development
Seedance 2.0 video prompt director. Converts plain-text scene descriptions into production-ready bilingual EN+ZH video prompts optimized for the Seedance 2.0 video generator. Handles all Seedance work — action (combat, pursuit, stunts), general (landscapes, journeys, atmosphere), dialogue (confrontations, negotiations, interrogations), and non-narrative commercial work (ad spots, music videos, fashion films, automotive inserts, product shots, pet/character demos, cutaway montages, social reels for TikTok / Reels / YouTube Shorts). Use whenever the user wants to create a Seedance video prompt, mentions Seedance, or describes a cinematic scene for video generation. For NARRATIVE screenplay-integrated work, use seedance-screenwriter instead.
development
Write Seedance 2.0 prompts in screenplay format for narrative storytelling — when the prompts will be cut into a film, short, or scene. Use whenever you're generating shots that will be edited into a continuous story with dialogue, character beats, scene continuity, or coverage. Pairs with the screenwriter skill — read the scene's screenplay first (or the project's `scene.md` if it exists), then translate each shot into a Seedance prompt that reads as a screenplay page, not as an engineering spec.
documentation
Скилл-инструмент для сценариста полнометражного фильма или сериала. Используй всегда, когда пользователь хочет писать сценарий, поэпизодник, разрабатывать сцены, бит-шит, диалоги, делать ревизии, считать экранное время, резать длину, работать с персонажами или мифологией истории. Скилл работает на основе методологий Макки, Кэмпбелла и Аристотеля, выдаёт Hollywood-формат .docx, поддерживает билингвальные сценарии (диалог на одном языке + перевод в скобках под ним), и помогает аудитировать структуру по причинности и движению ценности. Скилл не привязан к конкретной истории — пользователь приносит свою.
development
Extract shot composition DNA from any car photograph into structured JSON — camera angle, lens, framing, lighting — stripped of car-specific details. Then reuse extracted angles with any car identity to generate new images at scale. Use when: extracting angles from reference photos, building a shot library, batch-analyzing car photography, replicating a great angle with a different car, running extraction pipelines in Freepik or Flora. Triggers: "extract this angle", "steal this composition", "shot DNA", "analyze this car photo", "replicate this shot with my car", "batch extract angles", "car photography analysis", "angle extraction", "build a shot library".