calm-ui-audit
Skill1 starsChanged 5 months agoMITView on GitHub
Audits UI for prototype tells and applies calm Design Covenant fixes (subtraction-first, token tweaks, ≤3 CSS wins). Use when the user says the app feels primitive, prototype, unpolished, or asks for design polish, empty states, or Cursor/Linear-level UI density.
In this repository33 other files it gives its agents.
- architecture-blueprintSkill
- burnSkill
- capitalizeSkill
- cardSkill
- commit-mineSkill
- continuity-contextSkill
- elonizeSkill
- frontierSkill
- full-throttleSkill
- grade-sessionSkill
- groundSkill
- harmonizeSkill
- harvestSkill
- honeSkill
- milkSkill
- mirrorSkill
- moonwalkSkill
- nightshiftSkill
- page-auditSkill
- pmfSkill
- postSkill
- promptSkill
- publishSkill
- razeSkill
- redlineSkill
- relaySkill
- snipSkill
- stashSkill
- tasteSkill
- tidySkill
- untilSkill
- vacationSkill
- vaultSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 33
Prefer: remove borders, unify radius, fix empty state copy, reduce font weights
- Distinctive: line 34
Use the project's existing spacing/type tokens — do not invent a parallel scale
- Distinctive: line 35
Avoid: new components, animations, glow, badge stacks, architecture refactors in the same pass
- Distinctive: line 40
Do not add: notification center, analytics widgets, prompt libraries, graph theater, engagement bait, red error theater in non-error feeds, …
- Distinctive: line 42
Do not violate: one primary ask per screen, progressive disclosure, calm copy.
- Very specific: line 46
If repo is Continuum, read docs/brief/continuum-design-brief.md and docs/brief/surface-design-prd.md §2 before editing.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Calm UI Audit
Goal
Move one screen from "internal tool" to "shippable product feel" in one session — without a redesign, new dashboard, or feature creep.
Reference bar
| Reference | Borrow |
| Cursor | Calm density, muted chrome, status-not-toast |
| Linear | Typography hierarchy, tight empty states |
| Notion | Plain-language empty copy, soft borders |
Workflow
- Scope one surface — user picks or infer from open files (one panel/page, not whole app).
- List prototype tells — aim for ≥5 real tells, up to 10. Specific elements only, e.g. "mixed padding on cards (p-2 vs p-4)", "headers same weight as body", "empty state is single word 'Empty'". No vague "looks bad". Stop at 5 if the surface is already tight; do not pad with filler.
- Optional
DESIGN-NOTES.md(≤1 page) — only when the user wants to keep an audit artifact on disk:- Spacing scale (4/8/12/16/24)
- Type scale (10/12/13/14px roles)
- Muted text rule (primary / secondary / tertiary)
- One accent usage rule
- Apply ≤3 CSS-only wins (total diff <50 lines):
- Prefer: remove borders, unify radius, fix empty state copy, reduce font weights
- Use the project's existing spacing/type tokens — do not invent a parallel scale
- Avoid: new components, animations, glow, badge stacks, architecture refactors in the same pass
- Verify — no new notification toasts, KPI tiles, or fifth navigation items.
Design Covenant (hard rejects)
Do not add: notification center, analytics widgets, prompt libraries, graph theater, engagement bait, red error theater in non-error feeds, simultaneous map + map (dual SVG).
Do not violate: one primary ask per screen, progressive disclosure, calm copy.
Continuum-specific
If repo is Continuum, read docs/brief/continuum-design-brief.md and docs/brief/surface-design-prd.md §2 before editing. Match existing tokens in src/index.css and src/lib/themes/.
Output template
# UI audit — [surface name]
## Prototype tells (≥5, up to 10)
1. ...
## Quick wins applied
- [file:line] — ...
## Deferred (needs product call)
- ...Adversarial self-check
- Did I subtract more than I added?
- Are all listed tells actionable (file/element named)?
- At least 5 real tells (not padded to hit 10)?
- Under 50 lines of CSS changed?
- Empty states use calm prose, not "No data"?
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.