AI Aesthetic Judgment and Style Control
Paste the following prompt into your AI chat to install this skill:
Please follow https://skillhub.cn/install/skillhub.md and install @user_3c6cb52e/taste-skill-png.
About this skill
Problem
AI-generated output often works functionally but lacks design judgment: code snippets, image prompts, and prose may be correct while feeling generic or stylistically inconsistent. taste-skill-png breaks this judgment into loadable subskills, focusing on AI content aesthetic judgment and style control, and includes 13 subskills such as code-to-image and output optimization.
How It Works
The skill pack uses index-style loading: the main file does not expand every detail at once, but selects the relevant subskill file based on the task. It fits into an existing AI Agent workflow as an aesthetic check layer, rather than replacing the model or UI. Typical steps include:
- Clarify the task: determine whether code, image, text, or another output needs aesthetic constraints.
- Load the subskill: choose the capability that matches the current need from the 13 subskills.
- Apply control: perform style consistency, expression quality, code-to-image, or other concrete output refinements.
Boundaries
The materials do not list every subskill in detail, so the target subskill should be confirmed before use. It is suitable for Agent workflows on platforms such as Claude Code, Codex, and Hermes Agent, but it should not be treated as a complete design system; fixed brand rules still require your own design language and acceptance criteria.
Use Cases
- After generating frontend UI code in Claude Code, check visual hierarchy, spacing, and component expression for consistency.
- Before converting a code explanation into an image-style output in Codex, decide whether `code-to-image` fits the task.
- When producing multi-turn text output in a Hermes Agent workflow, align tone, terminology, and aesthetic judgment criteria.
- Before organizing AI-generated content in WorkBuddy, load the relevant indexed subskill to apply style control.
Best For
- Engineers using Claude Code: want style judgment for generated UI code, not only runnable output.
- Developers using Codex: need to optimize code output for easier review and reuse.
- Agent engineers using Hermes Agent: prefer task-based subskill loading instead of one large expanded file.
- Content engineers using WorkBuddy: need consistent aesthetic standards and style constraints for AI content.
Related Skills
Turn vague requests into structured prompts with role, task, context, and constraints, outputting only usable prompts without executing the task.
A skill for game NPC and enemy AI that separates decision, steering, and pathfinding, covering FSMs, behavior trees, A*, path caching, and pitfalls.
Provides multi-turn emotional companion chat through a third-party Agent API and supports continuing sessions via session_id.
An evaluation-driven workflow for diagnosing, optimizing, validating, and logging Agent Skills based on design patterns and anti-patterns.