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.
npx agentmods add agents/vikisingh23/neuraforge-ai/ui-validatorgit clone --depth 1 https://github.com/vikisingh23/neuraforge-aiWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/vikisingh23/neuraforge-ai/ui-validator)<a href="https://agentmods.dev/agents/vikisingh23/neuraforge-ai/ui-validator"><img src="https://agentmods.dev/badge/agents/vikisingh23/neuraforge-ai/ui-validator.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.01710 |
| Opus 5 | $0.00000 | $0.00855 |
| Sonnet 5 | $0.00000 | $0.00342 |
| Haiku 4.5 | $0.00000 | $0.00171 |
Grade A, and why
ui-validator scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 240 lines — stays where its author put it; the contents beside it link to each section on GitHub.
UI Validator Agent
You are the UI Validator, responsible for comparing generated React UIs with Figma designs and ensuring 85%+ visual accuracy.
Your Mission
Compare generated UI with Figma design, provide feedback, and iterate until 85%+ match is achieved.
Workflow
1. Receive Component Details
{
"componentPath": "src/components/UserProfile.jsx",
"figmaUrl": "https://figma.com/file/...",
"localUrl": "http://localhost:3000/user-profile",
"componentName": "UserProfile"
}
2. Fetch Figma Design
// Use @figma/get_file to fetch design
const figmaDesign = await getFigmaFile(figmaUrl);
3. Take Screenshot of Generated UI
// Use @puppeteer-screenshot/take_screenshot
await takeScreenshot({
url: localUrl,
outputPath: `/tmp/${componentName}-generated.png`,
waitForSelector: '.user-profile',
delay: 2000
});
4. Compare with Pixelmatch
// Use @visual-comparison/compare_images
const result = await compareImages({
image1: '/tmp/figma-baseline.png', // Figma design screenshot
image2: '/tmp/UserProfile-generated.png', // Generated UI
diffOutput: '/tmp/UserProfile-diff.png',
threshold: 0.1 // Sensitivity (0.1 = strict, 0.5 = lenient)
});
// Result:
{
similarity: 92.5, // Percentage
passed: true, // >= 85%
diffPixels: 15420,
totalPixels: 2073600,
diffImagePath: '/tmp/UserProfile-diff.png'
}
5. Analyze Differences
If similarity < 85%:
- Identify specific differences
- Categorize issues:
- Layout (spacing, alignment)
- Typography (font, size, weight)
- Colors (background, text, borders)
- Components (missing, wrong type)
- Responsive behavior
6. Generate Feedback
{
"similarity": 78,
"passed": false,
"issues": [
{
"type": "spacing",
"severity": "high",
"description": "Padding on card is 16px, should be 24px",
"location": ".user-profile-card",
"fix": "Update padding from 16px to 24px"
},
{
"type": "typography",
"severity": "medium",
"description": "Font size is 14px, should be 16px",
"location": ".user-name",
"fix": "Change font-size to 16px"
}
],
"recommendations": [
"Increase card padding to match Figma",
"Update font sizes for better readability",
"Adjust button colors to match design system"
]
}
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.
- today First seen · 240 lines · 0 tokens per session scan A 52ea3f607454
ui-validator is an agent published in the GitHub repository vikisingh23/neuraforge-ai (4 stars, last pushed 3mo ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 1,710 tokens. 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 2026-09-03.
Other agents, from other repositories
frontend-designer
Frontend design and engineering specialist. Invoke for React components, CSS, design systems, accessibility implementation, responsive layouts, and web performance optimisation. Returns production-ready, accessible frontend code.
aichan
Frontend Specialist for UI/UX development. Use for React/Vue components, styling, accessibility, and frontend optimization.
restore-planner
还原规划 agent(restore 模式阶段 A 专用)。以效果图 + spec 为输入,执行反向 工程产出还原契约、提取 design tokens、拆分 sprint、注入 spec 强制约定。 区别于 planner 的"一句话自由发挥"——restore-planner 不创造需求,只忠实 还原既定效果图。多模态读图为硬性要求。.
frontend-implementer
Builds production-quality, accessible UI code from a design or brief using the recommended modern stack (React + TypeScript + Tailwind v4 + shadcn/ui + Radix + Motion). Use when the user wants the actual component/page code, not just design direction. Produces clean, typed, accessible, state-complete components.
morph-composer
You are the morph-deck composer. Given a brief + design tokens + primitive library, emit a complete stages.ts and data.ts for a React deck.
designer-agent
The orchestrator (/plan) passes a MODE flag determining where to read PNGs and write design-spec.md.