AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Design Critique

skill-getcrew44-crew44-design-critique · by getcrew44

Use to give structured feedback on a screen, flow, or mockup — first impression, hierarchy, usability, consistency, accessibility, and what to fix first.

No reviews yet
0 installs
33 views
0.0% view→install

Install

$ agentstack add skill-getcrew44-crew44-design-critique

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-getcrew44-crew44-design-critique)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
3mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Design Critique? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Design Critique

Give the kind of feedback a designer actually wants: specific, prioritized, and tied to the user's goal — not adjectives.

Frame the critique

Ask, then critique. The same screen gets different feedback at different stages.

  • Stage. Exploration (push for range), refinement (tighten), or polish (catch what shipped).
  • Context. Who is it for? What is the user trying to do? What is this screen in service of?
  • Focus. All of it, or one slice (navigation, the empty state, the hero, the form)?

Pass 1: First impression (two seconds)

  • What draws the eye first? Is that what should?
  • Is the purpose of the screen clear before you read any copy?
  • What's the emotional read — calm, urgent, dense, sparse — and does that match the user's state?

Pass 2: Hierarchy and reading flow

  • Does the reading order match what the user is trying to do?
  • Are the right elements emphasized? Anything competing with the primary action?
  • Is whitespace doing work, or just sitting there?

Pass 3: Usability

  • Can the user accomplish the goal without backtracking?
  • Are interactive elements obviously interactive? Are read-only elements clearly not?
  • Are there unnecessary steps, fields, or confirmations?

Pass 4: Consistency

  • Does this match the existing design system, or invent locally? Name the specific deviations.
  • Spacing, typography, color tokens — pulled from the system, or magic numbers?
  • Do similar elements behave similarly across the screen?

Pass 5: Accessibility floor

  • Color contrast: body text ≥ 4.5:1, large text and UI ≥ 3:1.
  • Touch targets ≥ 44×44.
  • Focus order is logical; focus indicators are visible.
  • Form inputs have labels, not just placeholders.

How to deliver feedback

  • Be specific. "The CTA competes with the navigation" beats "the layout is confusing."
  • Explain the why. Tie each note to a user need or principle, not just taste.
  • Suggest, don't just diagnose. A direction is more useful than a problem.
  • Call out what works. Reinforces the moves to keep.
  • Rank by impact. Lead with the change that buys the most.

Output

  • Overall impression — one or two sentences. What works, what's the biggest opportunity.
  • Findings — grouped by pass (hierarchy, usability, consistency, accessibility). Each finding names the element, the issue, the severity (critical / moderate / minor), and the suggested fix.
  • What works well — short list. Preserve these.
  • Top three changes — prioritized. For each: what to change, why, expected effect.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.