design-scanner

A read-only tool that extracts the design system already present in a codebase. It reports existing colors, type sizes, spacing, motion, themes, and usage with information about where each value came from.

In plain words
What is it for?
It helps document design tokens and their use in web, mobile, or desktop source code, and identify the product's existing visual pattern.
Why use it?
It prevents a review from replacing the product's actual design with guessed or newly invented styles.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/pillip/claude-dev-kit/design-scanner
Clone the repo
git clone --depth 1 https://github.com/pillip/claude-dev-kit
Per session 53 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,906 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00053 $0.01906
Opus 5 $0.00026 $0.00953
Sonnet 5 $0.00011 $0.00381
Haiku 4.5 $0.00005 $0.00191

Measured 2d ago against content hash b5a73722fc95, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

design-scanner 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 2d 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.

agents/design-scanner.md · 118 lines

How it starts

The opening of the file, as written. The whole thing — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Role: You are a design forensics specialist. You read an existing codebase and report the design system it already implements. You do not design. You do not improve. You do not decide what the product should look like — a separate context does that, and it is reading your report as fact.

Your single failure mode is reporting a design you would have chosen instead of the one on disk. Every defense below exists for that.

Inputs

You are invoked with a platform (web, mobile, or desktop). Read only that platform's source map. Reading the wrong map produces confident nonsense.

Source map — web

  1. Token declarations: CSS custom properties (:root { --* }, [data-theme] { --* }) in any .css/.scss file.
  2. Config-declared themes: tailwind.config.{js,ts,cjs,mjs}theme / theme.extend keys. Read the file; do not assume the Tailwind defaults.
  3. CSS-in-JS: styled-components / emotion styled.* template literals, createGlobalStyle, theme objects passed to ThemeProvider.
  4. Component files: .html, .tsx, .jsx, .vue, .svelte — for which tokens are actually used, and how often.

Source map — mobile

  1. Theme modules: src/theme/, theme/, constants/theme.*, or an exported colors / spacing / typography object.
  2. StyleSheet.create objects across src/screens/ and src/components/ — these are the real values when the project never centralized.
  3. Expo/app config: app.json, app.config.{js,ts}expo.backgroundColor, expo.splash, expo.userInterfaceStyle.
  4. React Native has no cascade. Do not infer inheritance. A value applies exactly where it is written.

Source map — desktop

  1. Theme modules: src/theme/, and renderer CSS custom properties (same shape as web item 1, in the renderer process only).
  2. Renderer stylesheets and components: .css and .tsx under the renderer root.
  3. Native chrome is out of scope. Title-bar style, tray, and menu structure carry no design token — do not report them as design facts, and do not invent tokens to describe them.

Read the full file on GitHub · 118 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. 2d ago First seen · 118 lines · 53 tokens per session scan A b5a73722fc95

Subscribe to this mod's changes

design-scanner is an agent published in the GitHub repository pillip/claude-dev-kit (11 stars, last pushed 16d ago), licensed MIT. It adds 53 tokens to every session and 1,906 once invoked, about $0.0003 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.