AGENT SKILL

design-direction

Show a numbered prototype and hold it until the supervisor picks. Use when starting or redesigning a screen, and when a first pass would look the same on any similar brief.

适用职业设计师
使用场景视觉设计
License 许可
MIT
所属仓库
muratovv/ai-hats
最近更新
2026-11-05
Stars
★ 3
Forks
1

01 前言 Front Matter

name
design-direction
license
MIT
description
Show a numbered prototype and hold it until the supervisor picks. Use when starting or redesigning a screen, and when a first pass would look the same on any similar brief.

03 SKILL.md 正文

Design Direction

Turn a description into a numbered prototype, and wait until the supervisor picks.

When to Use

  • Filing the picked prototype is design-system. The quality pass is ui-ux-review.
  • A named taste (high-end-visual-design, brutalist, minimalist) is a separate call, and only when the brief asks for that style by name.
  • Skip when the supervisor already picked a prototype and the task is to file it.

Procedure

  1. Write the description: the subject, who it is for, and the job of this piece. One sentence each. When the brief does not say, ask before filling the gap.
  2. Make a prototype of the element you reworked. Name the additions it uses. Palette, type, and each color scheme the piece uses are the usual ones. One scheme is a complete answer. Name any other addition the piece actually rests on.
  3. Put a number on the prototype. When you show more than one, number each. The supervisor refers to a prototype by that number.
  4. Show the prototype and stop. Integration into the product waits until the supervisor picks a number or names the change. Silence is not a pick.

The prototype stays the proposal until that pick. design-system files the pick.

Open these when a line needs the source, and leave them as links:

  • anthropics/skills, skills/frontend-design/SKILL.md — the brief-wins rule and examples of an unchosen default.
  • Leonxlnx/taste-skill, skills/taste-skill/references/ — type, color, grid, motion, critique, when the line is about that.
  • nexu-io/open-design, craft/typography.md, craft/color.md — when the line is choosing type or color.

The anti-list this role applies is ui-ux-review. craft/anti-ai-slop.md is the same concern written elsewhere; read it for examples, and do not add its bullets as a second list.

Completion

  • A numbered prototype is shown, with the reworked element and the additions it uses.
  • The supervisor has picked a number or named a change. Silence is not a pick.
  • The product was not changed in this pass.

Validation scenario (RED). Asked for a settings screen on an accounting board, the agent produces a cream editorial hero, a small-caps pill over the title, and a reveal on every section, and integrates that into the product. The brief never asked for an editorial landing. GREEN: prototype 1, the settings screen with the element reworked for the board, the palette and type named, then a stop. No integration yet.

Anti-Patterns

  • Integrating into the product and calling that the prototype.
  • Showing several prototypes and not numbering them.
  • Treating the pause as optional because the prototype looks finished.
  • Shipping the cream editorial, the pill over the headline, or a reveal on every section as the look of a finished piece. Those are the common unchosen defaults, not the whole set.

Attribution

Procedure written for this catalog. The move comes from anthropics/skills skills/frontend-design (Apache-2.0) and the direction-then-show step of Leonxlnx/taste-skill (MIT). Their text is not copied.

04 同仓库更多 Skills

muratovv
muratovv/ai-hats

tool-evaluation-protocol

Time-bounded protocol for evaluating new tools, libraries, or methodologies before adoption. Use when evaluating a candidate dependency for the project, researching a methodology (e.g. ML/LLM frameworks, testing approaches), or following up on a hype-driven recommendation.

★ 3
muratovv
muratovv/ai-hats

systematic-debugging

4-phase bug-fix protocol (evidence, pattern, hypothesis, verify). Use when investigating any error, failure, or unexpected behavior, handling bug reports from users or monitoring, or diagnosing test failures with non-obvious causes.

★ 3
muratovv
muratovv/ai-hats

skill-template

Canonical template and validation guide for ai-hats skills. Use when creating a new skill, reviewing an existing one for structural compliance, or deciding which pattern fits a new behavior.

Skill与智能体开发
★ 3
muratovv
muratovv/ai-hats

skill-optimization

Audit and refactor library components to eliminate redundancy and staleness. Use during periodic library maintenance, after adding multiple new skills or rules, when a role's token budget is too high, or after retrospectives identify component issues.

Skill与智能体开发
★ 3
muratovv
muratovv/ai-hats

silent-fallback-gate

PreToolUse hook — refuses an Edit or Write that adds an inert broad except. Use when a Python edit was blocked for a silent fallback.

★ 3
muratovv
muratovv/ai-hats

retro-to-framework

Convert an observation about agent behavior — from a retrospective or from a single incident — into a framework-level change (hook, rule, skill, skill update). Use after a retrospective identifies problems that are not project-specific, after an incident where an agent misapplied a component it had read, when CLAUDE.md band-aids accumulate (more than 3 per-project rules that could be generic), or when the same problem recurs across multiple projects.

Skill与智能体开发
★ 3