Borrowing it
Nothing to install: this file belongs to rome-os/rome. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/rome-os/rome/main/.claude/skills/color-audit/SKILL.mdgit clone --depth 1 https://github.com/rome-os/romeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/rome-os/rome/color-audit)<a href="https://agentmods.dev/skills/rome-os/rome/color-audit"><img src="https://agentmods.dev/badge/skills/rome-os/rome/color-audit/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/rome-os/rome/color-audit"><img src="https://agentmods.dev/badge/skills/rome-os/rome/color-audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00199 | $0.02546 |
| Opus 5 | $0.00100 | $0.01273 |
| Sonnet 5 | $0.00040 | $0.00509 |
| Haiku 4.5 | $0.00020 | $0.00255 |
Grade A, and why
color-audit scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Color Audit
Evaluate a design system's color selection against the codified disciplines from color science and design-system practice (WCAG/APCA, Material HCT, Radix scale semantics, Adobe Leonardo, Stripe's accessible-palette work) and emit findings a coding agent can directly act on. This is a linter for the palette, not an aesthetic critique: most rules are mechanically checkable, and the bundled analyzer computes the numbers so no check is eyeballed.
Design philosophy (shared with ux-semantics-audit — it shapes every judgment call)
- Precision over recall. A false positive costs more than a miss: the consuming agent will "fix" a non-problem and degrade a palette someone chose deliberately. When in doubt, don't flag.
- No evidence, no finding. Every finding cites the token source (file, line, value) and — for Tier 1 rules — the measured number from the analyzer. If you cannot point at the value and the measurement, the finding does not exist.
- Exceptions are where precision lives. Each rule in
references/rules.mdhas an enumerated exceptions list; check it before flagging. Deliberate design (documented OLED black, brand-pinned steps, compressed dark ramps) is a pass, not a borderline flag. - Severity is trust calibration.
block= mechanical and near-certain (a declared pair failing WCAG is failing).warn= judgment involved. Never promote a judgment call toblock. - Measure, don't estimate. Never state a contrast ratio, lightness value, or CVD verdict from intuition — run
scripts/analyze-palette.mjs. A report whose numbers are wrong is worse than no report.
Workflow
Step 1 — Locate the palette
Find where color truth lives. Look for, in rough order: CSS custom properties (:root/theme blocks), a theme/token definition module, Tailwind config, and the framework vocabulary layer that exposes tokens to components.
In this repo the layers are:
- Primitives (oklch ramps):
packages/web/src/styles/primitives.css - Semantic values per theme × mode:
packages/web/src/lib/themes.ts(light/dark halves per theme) - Component-facing vocabulary: the
@theme inlineblock inpackages/ui/src/styles.css(its token comments are also the step-semantics documentation)
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 128 lines · 199 tokens per session scan A 2270c7db5cdd
color-audit is a skill published in the GitHub repository rome-os/rome (487 stars, last pushed today), licensed MIT. It adds 199 tokens to every session and 2,546 once invoked, about $0.0010 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
html-artifacts
Author the HTML for a plan artifact, dashboard iframe, or Slack attachment — structure, design plan, available runtime, theming, and craft. Read this before writing HTML for saveplan, outputiframe, or slackattachhtml.
browserwing-executor
Control browser automation through HTTP API. Supports page navigation, element interaction (click, type, select), data extraction, accessibility snapshot analysis, screenshot, JavaScript execution, and batch operations.
web-design
Penguin visual language for generated web pages and app UIs — GitHub-style simplicity with a single blue accent, light and pure-black dark themes, design tokens, component and chat-interface recipes, plus an opt-in warm paper editorial theme.
bento-slides
Create and edit Bento presentations — self-contained .bento.html decks whose document is JSON. Use whenever the user wants a slide deck or presentation: from scratch, from source material, or by improving an existing file.
ui-design
Design-quality reference for financial-research visual output: typography, color, composition, and avoiding generic AI aesthetics.
dashboard
Update the agent's Control Center. Use when the user says "update my Control Center", "update my dashboard", "redesign my home", or "/dashboard".