- Home
- Skills
- Testing & Debugging
- Visual Match Verify: Does the Build Match the Mockup?
Works with the AI Agents you already use
Visual Match Verify: Does the Build Match the Mockup?
UI implementations rarely match the original design tokens or spacing on the first pass.
Free
Visual Match Verify: Does the Build Match the Mockup?
Example session with this skill installed
Verify the new login page implementation against the design artifacts in /docs/designs. Run it at both desktop and mobile viewports.
- Read your context and instructions
- Compiled the visual match verify
- Generated the design tokens stylesheet
I have executed the visual match verification for the Login Page. I started the dev server at localhost:3000, disabled transitions, and captured screenshots for both desktop and mobile viewports.
Visual Verification Report
| Screen | Viewport | Diff % | Verdict | Issue | Fix (file:line) |
|---|---|---|---|---|---|
| Login Hero | 1440x900 | 1.2% | Minor | Button padding is 12px; spec calls for 16px. | LoginButton.tsx:42 |
| Login Form | 1440x900 | 0.05% | Match | - | - |
| Login Hero | 390x844 | 4.8% | Mismatch | Heading text wrapping causes 20px layout shift. | Hero.module.css:15 |
| Error State | 390x844 | 0.1% | Match | - | - |
Acceptance Criteria
- Desktop Layout Match (1440px)
- Mobile Layout Match (390px)
- Design Token Compliance (Colors/Fonts)
- A11y Pass (axe-core)
Observations:
The mobile mismatch is caused by the h1 font-size not scaling down, which pushes the login form below the fold. The desktop button padding is a simple CSS variable mismatch in the Tailwind config.
Next steps
- Apply the padding fix to
LoginButton.tsx. - Adjust the media query for the
h1tag inHero.module.css. - Re-run the verification script to confirm the mobile layout fix.
LoginButton.tsx
CSS · design tokens
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
UI implementations rarely match the original design tokens or spacing on the first pass. This skill provides objective evidence of design fidelity by running automated visual regression tests between your approved artifacts and your running dev server. It eliminates the guesswork of manual visual QA by generating pixel diffs and side-by-side comparisons.
What it does
- Automated Capture uses Playwright to snap screenshots of both the reference design and your local implementation at specific breakpoints.
- Pixel Comparison calculates the percentage of differing pixels using a 0.1 threshold to highlight layout shifts.
- State Verification forces specific UI states like mock data, open menus, or error messages to ensure edge cases are styled correctly.
- Style Auditing checks computed CSS values for colors, font sizes, and border radii against the design spec.
- Fix Generation maps visual discrepancies directly to the source code file and line number for rapid correction.
How it works
- Prepare Environment by starting the dev server and loading the reference artifacts or design exports.
- Execute Capture using Playwright scripts for desktop (1440x900) and mobile (390x844) viewports with animations disabled.
- Analyze Diffs to generate side-by-side comparison images and a pixel-match percentage report.
- Audit Tokens by checking computed styles and running a basic accessibility pass with axe-core.
Frameworks & tools
Built for projects using Playwright (Node or Python), pixelmatch, PIL, and axe-core. It integrates with any frontend framework that can run a local dev server.
Why this beats prompting it yourself
Writing reliable visual regression scripts involves complex logic for handling network idle states, font loading, and disabling transitions. This skill automates the configuration of viewports and state injection while providing a structured reporting format that maps visual errors to specific lines of code.
Use cases
- Verifying a new React component matches the Figma export exactly.
- Performing visual regression testing before a major release.
- Auditing mobile responsiveness to catch horizontal scroll or overflow bugs.
- Ensuring brand-critical design tokens are preserved in the final build.
Known limitations
Requires a running dev server and accessible design artifacts. Pixel comparisons may flag false positives due to font antialiasing differences across operating systems.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Testing & Debugging.
Requires local dev server access. Accuracy depends on the availability of high-resolution reference artifacts. Does not support automated animation testing.
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 buying a skill.
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
1 use
Times used to date
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 4 days ago
- Free with an Agensi account