Mode d’emploi
- Copiez la commande et examinez le paquet avant de l’installer dans votre environnement agent.
- Consultez le document original pour les outils, identifiants et environnements requis.
- Décrivez les entrées et le livrable souhaité ; comparez la sortie aux sources.
npx skills add https://github.com/sergekostenchuk/ui-ux-agent-skill-system --skill visual-content-directorAvant de commencer
- L’installation n’inclut ni abonnement au modèle, ni quota API, ni accès aux fichiers.
- Étoiles et installations reflètent l’activité, pas un test indépendant de qualité ou de sécurité.
Document Skill original
Rendu depuis le dépôt source en conservant titres, exemples, code, tableaux, liens et images.
Visual Content Director
This skill owns the visual-content layer: what images, videos, generated visuals, and asset treatments are needed to make a UI believable, inspectable, and on-brand. It does not own final UX strategy or SEO structure.
Operating Modes
asset-plan: define required images, videos, icons, maps, diagrams, and proofs.hero-media: specify first-viewport media and composition rules.generation-brief: create prompts for generated visuals without inventing factual claims.asset-qa: check whether assets are relevant, readable, aligned, and safe to use.handoff: package assets for Stitch, Pencil, Figma, or implementation.
Required Workflow
- Read references/visual-content-contract.md.
- Confirm project type, target audience, user promise, and brand constraints.
- Identify required content media:
- real product/place/person/object proof;
- generated or illustrative support;
- maps, diagrams, icons, before/after, galleries, video.
- Use assets/visual-content-plan.template.md for planning.
- Distinguish real evidence assets from generated mood or placeholder assets.
- Hand off exact media requirements to Stitch, Pencil, Figma, or code owners.
Safety Rules
- Do not fabricate real places, people, testimonials, property facts, awards, certifications, or legal claims.
- Do not use private images externally without approval.
- Label generated visuals as generated/planned if they are not real source evidence.
- For reveal, before/after, or layered media, route alignment checks to
image-layer-alignment-validator. - Public websites must keep important facts as text, not only baked into images.
Handoff Shape
text
Visual mode:
Asset roles:
Real evidence assets:
Generated/planned assets:
Hero media direction:
Forbidden visuals:
Accessibility notes:
SEO/LLM text preservation:
Next owner:
Checks:Validation
When editing this skill, run:
bash
python3 $CODEX_HOME/skills/.system/skill-creator/scripts/quick_validate.py $CODEX_HOME/skills/visual-content-director
python3 $CODEX_HOME/skills/senior-skill-architect/scripts/lint_production_skill.py $CODEX_HOME/skills/visual-content-director
python3 -m json.tool $CODEX_HOME/skills/visual-content-director/evals/evals.json >/dev/null
