A read-only reviewer for Figma screens that checks design-system compliance, accessibility, design drift, visual quality, and impact. Accessibility means making an interface usable by people with disabilities; design drift means differences between the design and the implementation.

Running compliance, accessibility, drift, visual-quality, or impact reviews on screens and producing recommendations for fixes.

It identifies likely design and accessibility problems without changing the code or designs, and can reuse recent audit reports.

ds-auditor — Agent for Claude Code by atezer/FMCP

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.

Clone the repo

git clone --depth 1 https://github.com/atezer/FMCP

Made for: Claude Code.

Or install figma-mcp-bridge, the plugin that ships this one along with the rest of its 29 skills, 10 commands, 3 agents, 1 MCP server.

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.

README.md

[![agentmods](https://agentmods.dev/badge/agents/atezer/fmcp/ds-auditor/github.svg)](https://agentmods.dev/agents/atezer/fmcp/ds-auditor)

Your own site

<a href="https://agentmods.dev/agents/atezer/fmcp/ds-auditor"><img src="https://agentmods.dev/badge/agents/atezer/fmcp/ds-auditor/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.

Your own site · 80×15

<a href="https://agentmods.dev/agents/atezer/fmcp/ds-auditor"><img src="https://agentmods.dev/badge/agents/atezer/fmcp/ds-auditor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 84 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,008 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.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00084	$0.01008
Opus 5.5	$0.00034	$0.00403
Sonnet 5.5	$0.00017	$0.00202
Haiku 4.5	$0.00008	$0.00101

Measured 1mo ago against content hash 0fd79be68829, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

ds-auditor 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 1mo 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/ds-auditor.md · 87 lines

How it starts

The opening of the file, as written. The whole thing — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.

DS Auditor — Claude Code Thin Delegator

Bu agent sadece Claude Code'a özgü bir wrapper'dır. Tüm iş mantığı (audit type routing, skill registry, read-only discipline, cache-first, self-audit, rapor formatı) fmcp-ds-audit-orchestrator skill'inde yaşar. Diğer platformlar (Cursor / Desktop / Web) aynı orchestrator skill'i doğrudan kullanır.

Görev

Adım 0 — Orchestrator Skill'i Yükle

Read("skills/fmcp-ds-audit-orchestrator/SKILL.md")

Essentials bölümü common case için yeterlidir. Advanced bölümüne sadece şu koşullarda in:

Çoklu audit tipi (compliance + a11y + drift aynı anda)
Cache hit ama forced refresh istendi
Büyük ekran SEVERE drill-down gerektiriyor
Fix önerisi karmaşık (narrow vs full karar)
Skill evolution ihtiyacı (yeni audit tipi)

Adım 1 — Audit Type Tespit

Net değil → AskUserQuestion ile çoklu seçim sor.

Adım 2 — Cache Kontrol

.claude/audits/<YYYY-MM-DD>-<nodeId>.md

24h içinde rapor var → oku, cache hit (lazy-create dizini ilk kullanımda). Yoksa fresh audit.

30 günden eski dosyaları temizle (her çalıştırma başında housekeeping).

Adım 3 — Sub-Skill Read (Audit Type'a Göre)

Mod	Read
compliance	`skills/audit-figma-design-system/SKILL.md`
a11y	`skills/figma-a11y-audit/SKILL.md`
drift	`skills/design-drift-detector/SKILL.md`
visual_qa	`skills/visual-qa-compare/SKILL.md`
impact	`skills/ds-impact-analysis/SKILL.md`

Sadece seçilen modun sub-skill'ini Read et — diğerlerine DOKUNMA.

Adım 4 — Read-Only Discipline

Hiçbir mutation YAPMA. Bulgular + düzeltme önerisi. Kullanıcı "fix uygula" derse:

Onayı al
Öneri: screen-builder agent'ını tetikle (bu düzeltmeyi o yapar) veya fmcp-screen-orchestrator skill'ini çağır
Narrow (1-3 bulgu tek frame) → fix-figma-design-system-finding
Full (4+ bulgu veya dağınık) → apply-figma-design-system

Read the full file on GitHub · 87 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.

1mo ago First seen · 87 lines · 84 tokens per session scan A 0fd79be68829

Subscribe to this mod's changes

ds-auditor is an agent published in the GitHub repository atezer/FMCP (10 stars, last pushed 12d ago), licensed MIT. It adds 84 tokens to every session and 1,008 once invoked, about $0.0003 per session on Opus 5.5. 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.

Design and UX ai claude cursor design-system design-tokens figma

design-token-extractor

Use this agent to extract design tokens from CSS, SCSS, JavaScript, or Tailwind config files. Trigger when the user wants to document their token system, migrate tokens between formats, audit token coverage, or generate a design system reference from existing code. Context: User wants to document their CSS custom…

Adityaraj0421/naksha-studio · 280 tokens

preview-html-author

Use ONLY as part of the /design-md skill pipeline. Builds one self-contained preview HTML file (preview.html) carrying both themes, which visually demonstrates a design.md's tokens and components. Hero on top + component showcase grid below. Writes to staging only — never to public/preview/ directly.

CaesiumY/ko-design-md · 66 tokens

preview-html-reviewer

Use ONLY as part of the /design-md skill pipeline. Scores preview.html against the approved design.md and references/rubric-preview.md, producing a preview-review-{N}.json. Read-only on inputs; writes only the review JSON.

CaesiumY/ko-design-md · 57 tokens

muriel-forger

Forger juror seat for a muriel jury. Asserts "I can reproduce this from a generic template in four minutes," then actually builds the counterfeit from a default stack, diffs it against the original, and reports the residue — the decisions that survive. Judges one thing only, distinguishability from competent generic…

andyed/muriel · 103 tokens

muriel-pedant

The Pedant jury seat. Judges a muriel artifact's text alone — unlabeled numbers, missing units, unverified numeric claims, and terminology drift between legend, annotation, and caption — with the composition withheld so a handsome layout cannot charm it. Applies the numeric-claim probe list from muriel-critique.md…

andyed/muriel · 112 tokens

muriel-squinter

The Squinter jury seat. Blurs a muriel artifact with muriel.squint and judges only what survives — focal dominance, reading order, figure/ground, whether the composition has one center of mass or three. One lens, one loss function, no opinions on typography, contrast, copy, or brand. Emits the juror ballot defined in…

andyed/muriel · 89 tokens