# AI output judgement

> Use when generated UI, copy or code needs a quality critique.

Critique generated UI, copy, or code when the user requests a quality review of AI output.

## Facts

- Type: Agent skill
- Discipline: Design
- Author: Karl Koch (https://github.com/kemiljk)
- Upstream: https://github.com/kemiljk/skills/blob/main/skills/ai-output-judgement/SKILL.md
- Source file: https://raw.githubusercontent.com/kemiljk/skills/main/skills/ai-output-judgement/SKILL.md (2154 bytes, sha256:3fda54971b1c8a1d95bfecfde686eacb98b8b106943b9e0d64b8de13f96cd8ae)
- Licence: MIT
- Upstream copy last changed: 2026-09-17
- Status: listed, not yet ruled on by the curator

## Use it

I want to use the AI output judgement skill by Karl Koch.
This is what it does, in the author's own words: Critique generated UI, copy, or code when the user requests a quality review of AI output.
Download the whole folder at https://github.com/kemiljk/skills/tree/main/skills/ai-output-judgement, SKILL.md and every file beside it, into your skills directory: ~/.claude/skills/ai-output-judgement/ for Claude Code, ~/.agents/skills/ai-output-judgement/ for Codex, ~/.cursor/skills/ai-output-judgement/ for Cursor. Get it from the author's repo, not from startupskills.dev.
Then tell me the path you installed it to and how to invoke it.

Nothing here is vendored: install from the author's repo, not from this site. A verbatim copy of the author's file follows for reading, under the licence reproduced after it; the current file is in the author's repo.

## Links

- Page: https://startupskills.dev/kemiljk/ai-output-judgement
- Author: https://startupskills.dev/kemiljk
- Index: https://startupskills.dev/llms.txt

## The file itself

---
name: ai-output-judgement
description: "Critique generated UI, copy, or code when the user requests a quality review of AI output."
license: MIT
---

# AI Output Judgement

Models output the median of their training data. Slop appears when that median is accepted uncritically. The job is to read the draft specifically and fix the last mile.

## Defaults

1. Treat generated output as scaffolding, never as the finished answer.
2. Review materials, not only screenshots.
3. Name concrete failures (padding, tokens, semantics, motion, identity), not vibes.
4. Prefer system references over one-off values the model invented.
5. Keep human taste on the exit path to production.

## Common median failures

- Hardcoded hex/rgb instead of semantic tokens
- `div` controls instead of native elements
- `transition: all` / linear timing for interactive motion
- Generic shadows, radii, and 24px padding
- Missing empty, error, disabled, and reduced-motion states
- React keys/identity that remount and reset state
- Accessibility bolted on after the visual pass

## Review rubric

1. **Semantics** — correct element and contract?
2. **Tokens** — colour, space, type, elevation from the system?
3. **Layout** — intentional hierarchy, not template residue?
4. **Motion** — physics/interruptibility where needed?
5. **States** — hover/focus/disabled/loading/empty/error?
6. **Fit** — matches neighbouring product surfaces?

## Prefer / Reject

| Prefer | Reject |
| --- | --- |
| Draft → critique → revise | Prompt → ship |
| Specific notes ("12px on spacing scale") | "Make it nicer" |
| System-aware replacements | Fresh one-off styling |
| Knowing when the model is wrong | Blind trust because the preview looks polished |

## Review checklist

- [ ] Did anyone read the code/CSS, not just the preview?
- [ ] Are tokens used everywhere values repeat?
- [ ] Are native semantics intact?
- [ ] Is motion intentional and interruptible?
- [ ] Do edge states exist?
- [ ] Would you defend every leftover default as deliberate?

## Source essays

- The slop isn't the models
- Force-multiplying design
- AI as pair design
- On giving AI taste

## Licence of the file above

The file above is https://raw.githubusercontent.com/kemiljk/skills/main/skills/ai-output-judgement/SKILL.md, by Karl Koch, reproduced without changes.
- Source: https://github.com/kemiljk/skills/blob/main/skills/ai-output-judgement/SKILL.md
- Copyright: Copyright (c) 2026 Karl Koch
- Licence: MIT, https://raw.githubusercontent.com/kemiljk/skills/main/LICENSE

The repo's licence file, verbatim from https://raw.githubusercontent.com/kemiljk/skills/main/LICENSE:

MIT License

Copyright (c) 2026 Karl Koch

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.