Frontend Design Taste Enforcer
A quality layer for eliminating generic AI aesthetics in frontend code, with style archetypes and anti-cliché rules.
- design
- frontend
- style
- quality
Video coming soon
This skill's short video goes here.What it does
Applies a set of style rules to prevent frontend work from looking generic or 'AI-flavored': visual archetypes (standard, soft-premium, editorial-minimal), configuration dials, and animation choreography standards. Raises the perceived quality of the final product.
When to use it
- Reviewing the visual result of an AI-generated page
- Defining the style archetype for a new project
- Polishing animations and micro-interactions before delivery
Steps
- Choose the style archetype that best fits the brand
- Review palette, typography, and spacing against anti-cliché rules
- Adjust design configuration dials according to the archetype
- Review animation and transition choreography
- Compare the result against high-quality reference examples
- Document remaining adjustments before final delivery
The skill
design-taste-enforcer/SKILL.mdDownload SKILL.md
--- name: design-taste-enforcer description: A quality layer for eliminating generic AI aesthetics in frontend code, with style archetypes and anti-cliché rules. --- # Frontend Design Taste Enforcer ## 1. What it does Applies a set of style rules to prevent frontend work from looking generic or 'AI-flavored': visual archetypes (standard, soft-premium, editorial-minimal), configuration dials, and animation choreography standards. Raises the perceived quality of the final product. ## 2. When to use it - Reviewing the visual result of an AI-generated page - Defining the style archetype for a new project - Polishing animations and micro-interactions before delivery ## 3. Steps 1. Choose the style archetype that best fits the brand 2. Review palette, typography, and spacing against anti-cliché rules 3. Adjust design configuration dials according to the archetype 4. Review animation and transition choreography 5. Compare the result against high-quality reference examples 6. Document remaining adjustments before final delivery ## 4. Safety rules Always work in draft mode: nothing is sent, published, or paid for without explicit client approval. Never invent data, metrics, or results. No passwords, tokens, or API keys are requested or shown in chat. ## 5. Deliverable A clear, reviewable draft of the output (text, code, report, or file), ready for the user to approve before it becomes final.
For Claude Code, save it as ~/.claude/skills/design-taste-enforcer/SKILL.md. In Cursor, ChatGPT or another assistant, paste it as project instructions.
Want to build it together?
Inside PubliExpert AI Guild you get snapshots, live sessions and builders to finish it with.