dev-challenge: Skill for Claude Code

.claude/skills/design-fidelity/SKILL.md

design-fidelity is a skill for Claude Code from micheltlutz/dev-challenge. It costs 75 tokens per session (1,124 once invoked), scanned A, original, MIT.

A guide for checking whether an implemented web or mobile screen matches its approved design, including colours, typography, spacing, layout, and required test identifiers.

In plain words
What is it for?
Use it to review a screen against Figma designs, design tokens, style guidance, or mockup images, with optional screenshot comparison at the specified desktop and mobile sizes.
Why use it?
It identifies specific visual and structural differences instead of relying on a general impression that the screen looks close.

Skill for Claude Code

Written for Claude Code: installed under .claude/.

This is micheltlutz/dev-challenge's own configuration. It tells Claude Code how to work on dev-challenge itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything dev-challenge configures →

Reuse

Borrowing it

Nothing to install: this file belongs to micheltlutz/dev-challenge. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/micheltlutz/dev-challenge/main/.claude/skills/design-fidelity/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/micheltlutz/dev-challenge

Made for: Claude Code.

Wrote 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.

agentmods badge for design-fidelity

README.md
[![agentmods](https://agentmods.dev/badge/skills/micheltlutz/dev-challenge/design-fidelity/github.svg)](https://agentmods.dev/skills/micheltlutz/dev-challenge/design-fidelity)
Your own site
<a href="https://agentmods.dev/skills/micheltlutz/dev-challenge/design-fidelity"><img src="https://agentmods.dev/badge/skills/micheltlutz/dev-challenge/design-fidelity/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.

agentmods 80×15 button for design-fidelity

Your own site · 80×15
<a href="https://agentmods.dev/skills/micheltlutz/dev-challenge/design-fidelity"><img src="https://agentmods.dev/badge/skills/micheltlutz/dev-challenge/design-fidelity.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 75 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,124 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00075 $0.01124
Opus 5 $0.00037 $0.00562
Sonnet 5 $0.00015 $0.00225
Haiku 4.5 $0.00007 $0.00112

Measured 9d ago against content hash dc75fd784953, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

design-fidelity 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 9d 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.

.claude/skills/design-fidelity/SKILL.md · 113 lines

How it starts

The opening of the file, as written. The whole thing — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Design fidelity

Check an implementation against the lab's design system, and say specifically what is off rather than "looks close".

Sources, in order of authority

  1. Figma — the canonical design. Node ids per screen are in specs/design/screens.md. Reachable via the optional Figma MCP server in .mcp.json.
  2. specs/design/tokens.json — sampled from the exported mockups. The machine-readable fallback, and good enough for every challenge.
  3. specs/design/style-guide.md — the same thing in prose, with the reasoning.
  4. The mockup PNGs in readme_files/images/.

If a token disagrees with Figma, Figma wins — and the token is a bug worth an issue.

How to check

If browser automation is available, screenshot the running UI at the design's viewport (1440×1024 desktop, 375×812 mobile) and compare against the mockup side by side. Otherwise read the stylesheet and component source.

Sample computed styles rather than trusting the source — a variable can be defined correctly and then overridden.

The checklist

Colour

Role Token Value
Primary action --color-action-default #19191C
Brand purple --color-brand-primary #8E48EC
Deep purple --color-brand-primary-deep #4E01B4
Canvas --color-surface-canvas #EFEFF0
Credit --color-semantic-credit #0CB97B
Debit --color-semantic-debit #E02244

The mistake to look for first: a purple primary button. It is the single most common fidelity failure. Purple is for brand surfaces — the auth split panel, the active nav item, the statement table header. Actions are near-black.

Second: hard-coded hex. If the stylesheet contains #8E48EC rather than var(--color-brand-primary), say so. The tokens exist to be imported.

Money semantics

Credit green, debit red — and never colour alone. The design pairs it with / and +/-, and WEB05-AC3 checks for it. A table that only uses colour is unreadable to a red-green colour-blind user, which is roughly one in twelve men.

Read the full file on GitHub · 113 lines

Changes

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.

  1. 9d ago First seen · 113 lines · 75 tokens per session scan A dc75fd784953

Subscribe to this mod's changes

design-fidelity is a skill published in the GitHub repository micheltlutz/dev-challenge (32 stars, last pushed 7d ago), licensed MIT. It adds 75 tokens to every session and 1,124 once invoked, about $0.0004 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.