mcp-ux-vision
Provides AI-powered visual analysis for UI/UX assessment using Google's Gemini Vision API, supporting screenshot capture, UI element detection, accessibility auditing, and comprehensive JSON reporting.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@mcp-ux-visionRun a full UX report on https://example.com"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP UX Vision
An MCP server that provides AI-powered visual analysis for UI/UX assessment using Google's Gemini Vision API.
Features
Screenshot capture from URLs
AI-powered UI element detection and analysis
WCAG contrast ratio analysis
Color palette and typography extraction
Visual accessibility auditing
Comprehensive JSON reporting
Related MCP server: websight
Installation
git clone <repository-url>
cd mcp-ux-vision
npm install
npm run buildConfiguration
Add to your MCP client configuration:
{
"servers": {
"ux-vision": {
"command": "node",
"args": [
"path/to/mcp-ux-vision/build/index.js"
]
}
}
}Tools
screenshot_url
Capture a screenshot of any webpage.
url(required): URL to screenshotfullPage(optional): Capture full page vs viewportwaitTime(optional): Delay before capture (ms)
analyze_screen
Analyze the most recent screenshot with AI, extracting:
UI elements with geometry and styling
Color palette and typography
Accessibility metrics
Visual hierarchy assessment
generate_report
Create a comprehensive JSON report from the last analysis.
testUrl(required): URL that was analyzedappName(optional): Application nameoutput_path(optional): Report output directory
analyze_url_full_report
One-step workflow combines all above tools with the same parameters: screenshot → analyze → report.
Usage
Once the server is running and configured in your MCP client (like Claude), you can use natural language prompts to call the tools:
screenshot_url
"Take a screenshot of https://example.com"
"Capture a full page screenshot of localhost:3000"
"Screenshot https://myapp.com and wait for the .loading element to disappear"
analyze_screen
"Analyze the current screenshot for UI elements"
"Examine the accessibility of this interface"
"What UI components do you see in the screenshot?"
generate_report
"Create a UX report for the analysis of https://example.com"
"Generate a comprehensive UI audit report"
analyze_url_full_report
"Take a screenshot of https://example.com and create a full UX analysis report"
"Analyze https://myapp.com and generate a complete accessibility audit"
License
MIT
This server cannot be deployed
Maintenance
Related MCP Connectors
Capture screenshots, detect visual regressions between page versions, and analyze with AI.
Hosted browser for AI agents: screenshots, post-JS DOM, console, WCAG. No install, no API key.
Read, screenshot, diff, verify and audit any URL: SEO, links, sitemaps, schema, a11y.
Desktop and mobile website screenshots plus page context for AI agents and automation workflows.
Related MCP Servers
- AlicenseAqualityDmaintenanceEnables AI assistants to capture screenshots of web pages using automated browser sessions. Supports full-page and element-specific screenshots, device simulation, and JavaScript execution for comprehensive web testing and monitoring.66 npmMIT
- AlicenseNot gradedqualityDmaintenanceEnables AI agents to see, analyze, and visually verify web page changes through pixel-perfect diffing, theme extraction, layout analysis, and interactive element detection.19 npmMIT
- FlicenseAqualityDmaintenanceEnables AI agents to perform comprehensive web application testing including visual, functional, performance, accessibility, and SEO analysis using browser automation without requiring API keys.7-
- AlicenseAqualityBmaintenanceScreenshot, visual-diff, and AI page-analysis API for AI agents. Capture any URL as PNG, JPEG, WebP, PDF, or HTML, diff two versions of a page to catch visual regressions, and get an AI summary of what a page contains.344 npm1MIT