
Visual Verdict
- 551 installs
- 38.3k repo stars
- Updated August 4, 2026
- yeachan-heo/oh-my-claudecode
Visual Verdict is a Claude Code skill that compares generated UI screenshots to reference images and returns a strict JSON pass-or-fail verdict for layout, spacing, typography, and design fidelity.
About
Visual Verdict is a level-2 skill in yeachan-heo/oh-my-claudecode for structured visual QA. It accepts reference_images[], a generated_screenshot, and optional category_hint, then compares layout, spacing, typography, and component styling against intent. The skill returns deterministic pass/fail JSON guidance that can drive the next edit iteration before merge or release. Use it when visual fidelity requirements exist and you need automated judgment on polish and regressions instead of subjective chat feedback alone.
- Automated visual UI assessment
- Pre-ship design fidelity checks
- Layout and polish regression detection
- Structured verdict on frontend changes
- Claude Code integration for visual review
Visual Verdict by the numbers
- 551 all-time installs (skills.sh)
- +8 installs in the week ending Jul 27, 2026 (Skillselion tracking)
- Ranked #555 of 1,880 Design & UI/UX skills by installs in the Skillselion catalog
- Data as of Aug 5, 2026 (Skillselion catalog sync)
npx skills add https://github.com/yeachan-heo/oh-my-claudecode --skill visual-verdictAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 551 |
|---|---|
| repo stars | ★ 38.3k |
| Last updated | August 4, 2026 |
| Repository | yeachan-heo/oh-my-claudecode ↗ |
How do you automate visual UI review before merge?
Run a visual verdict on UI changes before merge or release when you need automated judgment on layout, polish, regressions, or design fidelity against intent.
Who is it for?
Frontend developers validating screenshot output against design references before merge or release.
Skip if: Backend-only changes or accessibility audits that require DOM inspection without reference screenshots.
When should I use this skill?
UI task includes reference images, a generated screenshot, and needs deterministic visual pass/fail before continuing edits.
What you get
Strict JSON visual verdict, pass-or-fail fidelity assessment, and iteration guidance for UI edits.
- json visual verdict
- pass-fail fidelity assessment
Files
<Purpose> Use this skill to compare generated UI screenshots against one or more reference images and return a strict JSON verdict that can drive the next edit iteration. </Purpose>
<Use_When>
- The task includes visual fidelity requirements (layout, spacing, typography, component styling)
- You have a generated screenshot and at least one reference image
- You need deterministic pass/fail guidance before continuing edits
</Use_When>
<Inputs>
reference_images[](one or more image paths)generated_screenshot(current output image)- Optional:
category_hint(e.g.,hackernews,sns-feed,dashboard)
</Inputs>
<Output_Contract> Return JSON only with this exact shape:
{
"score": 0,
"verdict": "revise",
"category_match": false,
"differences": ["..."],
"suggestions": ["..."],
"reasoning": "short explanation"
}Rules:
score: integer 0-100verdict: short status (pass,revise, orfail)category_match:truewhen the generated screenshot matches the intended UI category/styledifferences[]: concrete visual mismatches (layout, spacing, typography, colors, hierarchy)suggestions[]: actionable next edits tied to the differencesreasoning: 1-2 sentence summary
<Threshold_And_Loop>
- Target pass threshold is 90+.
- If
score < 90, continue editing and rerun/oh-my-claudecode:visual-verdictbefore any further visual review pass. - Do not treat the visual task as complete until the next screenshot clears the threshold.
</Threshold_And_Loop>
<Debug_Visualization> When mismatch diagnosis is hard: 1. Keep $visual-verdict as the authoritative decision. 2. Use pixel-level diff tooling (pixel diff / pixelmatch overlay) as a secondary debug aid to localize hotspots. 3. Convert pixel diff hotspots into concrete differences[] and suggestions[] updates. </Debug_Visualization>
<Example>
{
"score": 87,
"verdict": "revise",
"category_match": true,
"differences": [
"Top nav spacing is tighter than reference",
"Primary button uses smaller font weight"
],
"suggestions": [
"Increase nav item horizontal padding by 4px",
"Set primary button font-weight to 600"
],
"reasoning": "Core layout matches, but style details still diverge."
}</Example>
Task: {{ARGUMENTS}}
Related skills
How it compares
Use for screenshot-to-mockup fidelity gates; use DOM-based a11y or unit tests when behavior matters more than pixels.
FAQ
What inputs does Visual Verdict require?
Visual Verdict needs reference_images[] with one or more image paths, a generated_screenshot of the current UI, and optionally category_hint. The skill compares visual fidelity and returns strict JSON pass/fail guidance.
When should Visual Verdict run in a UI workflow?
Run Visual Verdict when a task has visual fidelity requirements and both reference and generated screenshots exist. Visual Verdict provides deterministic review before continuing edits, merging, or releasing.