- Home
- Skills
- Frontend & Web Apps
- ui critic
Works with the AI Agents you already use
ui critic
Independent visual design auditing and regression testing for web, mobile, and Storybook components.
$5
ui critic
Example session with this skill installed
Run a UI critique on the local dev server at http://localhost:3000 focusing on the homepage and the checkout flow. include an accessibility pass for dark mode and high contrast.
- Read your context and instructions
- Compiled the ui critic
- Generated the UI component
I have initialized the UI critic, captured the homepage and checkout routes, and performed a visual audit including accessibility variants.
UI Critique Summary: localhost
| Page | Finding | Type | Logic / Fact |
|---|---|---|---|
| Homepage | Primary CTA has low contrast (3.2:1) in Dark Mode | Accessibility | Measured: #707070 on #000000 |
| Homepage | "Latest Products" header lacks visual hierarchy | Judged | Brief: Brand is "Bold and Premium" |
| Checkout | Payment method icons are too small for touch (32px) | Measured | Violation: Minimum 44px required |
| Checkout | Missing focus ring on the 'Place Order' button | Interaction | Fact: Keyboard tab confirmed no visible state |
| General | Token drift: 3 different shades of "Success Green" used | Lint | #22C55E, #21BA58, #28A745 |
Critic's Requests:
- Provide the
design-tokens.jsonfile to reconcile the Success Green sprawl. - Clarify if the placeholder "lorem ipsum" in the checkout footer is intentional for this build.
Next steps
- Review the full visual report in
ui-critic/out/critique.htmlto see annotated screenshots. - Triage the contrast and touch-target findings in
ui-critic/decisions.md. - Provide the requested token file and rerun the critique to verify alignment.
critique.html
TSX · React component
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Visual design is often a blind spot for AI agents focused on code correctness. This skill integrates an independent visual critic into your development loop to audit web and mobile interfaces for taste, usability, and design system adherence.
What it does
- Visual Auditing captures screenshots of websites, Storybook components, or mobile apps (Android/iOS) for critique by Gemini or OpenAI.
- Interaction Sweeps automatically detects missing hover states, keyboard focus issues, and touch target violations using pixel-confirmed measurements.
- Design System Linting identifies token drift and spacing sprawl by comparing rendered styles against your defined design tokens.
- Fidelity Checking compares live implementations against Figma files to report layout deviations and missing elements.
- Regression Testing performs before/after comparisons to verify visual fixes and ensure no new UI regressions were introduced.
How it works
- Initialize the environment with
uicritic initto generate a project brief and configuration defining your brand, audience, and design principles. - Capture and Critique by running
uicritic runagainst a URL, emulator, or Storybook instance to generate a detailed finding report. - Triage the findings in a structured loop, accepting visual improvements and recording rejections in a decisions file to prevent future re-litigation of taste.
- Verify implementations with a comparison run that highlights improved versus regressed UI elements.
Frameworks & tools
Works with React, Vue, Storybook, Playwright, and Figma. Supports web applications, Android (via adb), and iOS (via Simulator). Requires a Gemini or OpenAI API key.
Why this beats prompting it yourself
Generic prompts lack the context of your specific brand and user goals. This skill uses a structured brief, measured facts, and a persistence layer for past decisions to ensure critiques are consistent and relevant to your product's unique design language.
Use cases
- Auditing a new landing page for visual polish and accessibility before shipping to production.
- Identifying styling inconsistencies in a component library via Storybook.
- Verifying that a mobile app refactor didn't break layouts across different font scales or dark mode.
- Generating App Store and Play Store marketing assets directly from reviewed screens.
Known limitations
Requires an active API key for Gemini or OpenAI. iOS support is currently experimental and requires a macOS host. Does not support capturing screens behind complex manual authentication unless the session is pre-established on the device.
Requires Gemini/OpenAI API key. iOS support is experimental. Authentication requires pre-active sessions or environment variables. Unscoped 'ui-critic' npm package is not supported.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 8 days ago
Requires