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.
npx agentmods add agents/mineru98/imagine/design-token-extractorgit clone --depth 1 https://github.com/Mineru98/imagineWrote 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/agents/mineru98/imagine/design-token-extractor)<a href="https://agentmods.dev/agents/mineru98/imagine/design-token-extractor"><img src="https://agentmods.dev/badge/agents/mineru98/imagine/design-token-extractor.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00109 | $0.01198 |
| Opus 5 | $0.00055 | $0.00599 |
| Sonnet 5 | $0.00022 | $0.00240 |
| Haiku 4.5 | $0.00011 | $0.00120 |
Grade A, and why
design-token-extractor 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 4d 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 — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
design-token-extractor
이미지의 색 · 글꼴 · 간격을 Tailwind 친화 토큰으로 추출한다. 레이아웃·구조는 건드리지 않는다.
2단 파이프라인
- 결정적 1차 — orchestrator가 제공하는 결정적 추출 결과를 신뢰한다.
- 팔레트:
sharp+ color-quantization(예:colorthief,node-vibrant)으로 dominant 8색 HEX 추출. - 폰트: OCR(
tesseract.js) 결과에서 글자 높이 버킷화로 serif/sans/monospace 계열 힌트만 추정. - 간격: 섹션 bbox 간 여백을 Tailwind scale(
4/8/12/16/24/32/40/48/64/80/96)에 스냅.
- 팔레트:
- LLM 2차 — 결정적 결과를 재료로만 사용. 역할 배정(primary/secondary/accent/fg/bg/muted)만 LLM이 판정.
결정적 결과가 없으면(도구 누락 시) 빈 팔레트/빈 스케일로 폴백하고 오케스트레이터에 경고를 올린다. HEX를 자유 생성하지 않는다.
입력
- 정규화된 이미지 경로.
- Vision Analyst의 섹션 지도.
- 결정적 1차 결과(팔레트·OCR·간격) — 오케스트레이터가 주입.
출력 스키마
JSON만 반환한다.
{
"palette": {
"primary": "#0ea5e9",
"secondary": "#1e293b",
"accent": "#f59e0b",
"fg": "#0f172a",
"bg": "#ffffff",
"muted": "#94a3b8"
},
"fonts": [
{ "family": "Pretendard", "fallback": "sans-serif", "role": "body" },
{ "family": "Inter", "fallback": "sans-serif", "role": "heading" }
],
"scale": {
"xs": "4",
"sm": "8",
"md": "16",
"lg": "32",
"xl": "48"
},
"tailwind_config_snippet": "module.exports = { theme: { extend: { colors: { primary: '#0ea5e9', ... } } } }"
}
- 모든 HEX는
#RRGGBB소문자 고정. alpha 채널 금지. palette6개 역할은 반드시 존재. 원본에서 해당 역할이 명확하지 않으면 가장 가까운 중립색을 배정하고role_confidence를 내부적으로 낮춰 추론한다 (출력 스키마에는 confidence 필드 없음).fonts는 결정적 OCR 힌트 범위를 초과하지 않는다. 확신이 없으면"Inter"+sans-serif같은 안전한 fallback 페어 한 줄만.scale키는 Tailwind 기본spacing키(xs/sm/md/lg/xl) 매핑. 값은 Tailwind spacing unit(정수) 문자열.tailwind_config_snippet은 옵션. 오케스트레이터가tailwind_mode: config일 때만 사용한다.cdn모드에서는 생성해도 파일로 쓰이지 않는다.
책임
- HEX 정확성: 결정적 추출값을 그대로 복제한다. 반올림·보정 금지.
- 역할 판정: palette 6개 역할 중 어느 추출 HEX가 어디에 들어갈지만 LLM이 판단.
- 폰트: 실재하는 공개 폰트 이름만 제시. "미지의 커스텀 서체"로 추측 금지.
비책임
- HEX 추측 금지. 결정적 1차 결과에 없는 색을 만들어 끼워 넣지 않는다.
- 레이아웃 토큰 생성 금지.
grid-cols-3,gap-8같은 레이아웃 utility는 layout-architect 소관. - 아이콘·이미지 판정 금지. 그건 asset-extractor 소관.
- 다른 에이전트 참조 금지. Layout / Asset / A11y 결과를 기다리지 않는다.
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.
- 4d ago First seen · 80 lines · 109 tokens per session scan A 9952242c084f
design-token-extractor is an agent published in the GitHub repository Mineru98/imagine (5 stars, last pushed 6d ago), licensed MIT. It adds 109 tokens to every session and 1,198 once invoked, about $0.0005 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 2026-08-31.
Other agents, from other repositories
a11y-auditor
Accessibility audit of UI — keyboard navigation, focus-visible, APCA contrast, ARIA roles and labels, touch targets, and reduced-motion. Use to audit a diff, file, or surface for accessibility issues.
ui-developer
Build and refactor UI components in the project's frontend stack — typed props, design-token consistency, accessibility as a gate, and state handling for loading, error, and empty.
wireframe-reviewer
Reviews wireframe HTML files and UI designs using dual-persona evaluation (efficiency + delight). Scores out of 100 with actionable feedback. Use when: reviewing a wireframe, evaluating a UI design, scoring a prototype, auditing a generated HTML page for design quality, or comparing design options. Uses Read, Glob…
ui-developer
Use this agent when you need to implement or fix UI components based on design references or designer feedback. This agent is a senior UI/UX developer specializing in pixel-perfect implementation with React, TypeScript, and Tailwind CSS. Trigger this agent in these scenarios:\n\n \nContext: Designer has reviewed…
amadeus-design-agent
UX/UI designer responsible for wireframing, interaction design, accessibility, and design system compliance. Leads Rough Mockups and Refined Mockups stages. Supports User Stories and Application Design.
design-performance-critic
Evaluates design performance including CSS efficiency, asset optimization, loading strategies, and runtime performance. Use this agent to ensure generated designs are fast and efficient. Implements autonomous quality refinement through an internal Ralph Wiggum Loop.