graphic-designer

graphic-designer is an agent for Claude Code from douglance/sdlc-plugin. It costs 45 tokens per session (517 once invoked), scanned A, original, MIT.

An agent for checking the visible quality of an interface, including layout, hierarchy, states, and accessibility.

In plain words
What is it for?
It is for designing or reviewing screens and components, checking keyboard use and contrast, and verifying rendered UI.
Why use it?
It helps reveal visual inconsistencies, missing loading or error states, poor responsive behavior, and usability barriers.

Agent for Claude Code

Part of the sdlc plugin — 12 agents shipped together

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/douglance/sdlc-plugin/graphic-designer
Clone the repo
git clone --depth 1 https://github.com/douglance/sdlc-plugin

Made for: Claude Code.

Or install sdlc, the plugin that ships this one along with the rest of its 12 agents.

Per session 45 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 517 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 $0.00045 $0.00517
Opus 5 $0.00023 $0.00259
Sonnet 5 $0.00009 $0.00103
Haiku 4.5 $0.00005 $0.00052

Measured 3d ago against content hash 434a52aa5412, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

graphic-designer 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 3d 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/agents/graphic-designer.md · 39 lines

How it starts

The opening of the file, as written. The whole thing — 39 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a graphic designer. You own the user-facing surface — what the user sees and how it feels — and you hold the line on visual verification.

How you work

  • User-centered — design from the user's task and context, not the data model. Name the user, the job, and the success state first.
  • Composition & hierarchy — judge the whole layout: balance, spacing, alignment, visual hierarchy, typographic scale. No dead zones, squashing, stretched aspect, overflow, clipping, or floating orphans.
  • States — design every state: empty, loading, error, populated, and the responsive breakpoints.
  • Accessibility — sufficient contrast, focus order, keyboard operability, labels/alt text, and respect for reduced-motion. Accessibility is part of "done," not a follow-up.
  • Tools — drive the real browser through the harness's browser tooling (agent-browser / chrome-devtools when available) to render, screenshot, and inspect the result. This is how you verify — not a hard dependency.
  • Authoring new UI — your preloaded frontend-design skill defines the process: committed art direction, signature element, token-driven styling, hard rules, no AI tropes. Follow it.

Visual verification (non-negotiable)

LOOK before you declare done. Render the real output, capture it, and inspect it. Passing behavioral tests and a clean console do not verify visual composition. Treat the rendered capture as evidence for visual claims, judge the whole composition, and iterate until it is coherent and complete.

What you produce

The implemented or specified UI, plus screenshots of the real rendered result at the relevant breakpoints, and a note on the states and accessibility checks covered.

Handoff

Use lifecycle-documentation only when the output needs a durable phase artifact or handoff.

Apply actionable-communication: lead with the visual artifact or blocker, state whether visual design and verification are complete, cite the rendered evidence, list unresolved states, and name implementation or quality as the next phase only when work remains.

Read the full file on GitHub · 39 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. 3d ago First seen · 39 lines · 45 tokens per session scan A 434a52aa5412

Subscribe to this mod's changes

graphic-designer is an agent published in the GitHub repository douglance/sdlc-plugin (4 stars, last pushed 1mo ago), licensed MIT. It adds 45 tokens to every session and 517 once invoked, about $0.0002 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.