46+ AI-Powered Tools · MCP Server + CLI + REST API · Free forever

Give your AI coding
assistant eyes.

Claude, Cursor, and Windsurf can read your code — but they can't see your UI. screenshotsmcp gives them the ability to screenshot any URL so they can actually help you build.

Works withCursorWindsurfVS CodeClaude DesktopClaude Coden8n

Quick Start

# Install & configure everything in one command
$npx screenshotsmcp setup
# Or use the install script
$curl -fsSL https://screenshotsmcp.com/install.sh | bash

Works on macOS, Linux, and Windows. Requires Node.js 18+.

How it works

1

Install in your IDE

One-click setup for Cursor, Windsurf, Claude Desktop, and VS Code. Your API key is pre-filled.

2

Ask your AI assistant

"Take a screenshot of my site" or "What does example.com look like?" — your AI handles the rest.

3

AI sees your UI

Screenshots are returned inline. Your AI can now debug layouts, compare designs, and verify deploys.

MCP Tool

Just ask your AI.

Once installed, your AI assistant can take screenshots on command. No code, no setup — just natural language.

You:Screenshot my site at https://myapp.com and tell me if the hero looks right
AI:Taking a screenshot… Here's what I see: the hero text is cut off on mobile. The CTA button is below the fold at 390px. Here's how to fix it…

What developers use it for

🐛

UI Debugging

AI sees your layout and finds the bug

🚀

Deploy Verification

Confirm your site looks right after pushing

🔍

Competitor Analysis

AI researches competitor sites visually

📱

Responsive Testing

Check mobile and desktop in seconds

📸

Auto Documentation

Keep screenshots in docs always up to date

🧪

Visual QA

Catch regressions before they reach production

Accessibility Audit

Inspect ARIA tree and detect a11y issues

Performance Testing

Measure Core Web Vitals and find bottlenecks

🔐

Authenticated Testing

Log in and test protected pages automatically

Everything you need

46+ tools across screenshots, browser automation, SEO, performance, accessibility, and AI analysis.

Visual AI Browser
AI agents can navigate, click, fill forms, scroll, and screenshot any website — just like a human.
Native MCP Server
Works with Claude, Cursor, Windsurf, and any MCP-compatible client out of the box.
Lightning Fast
Results in seconds via async queue. Sync mode available for simple use cases.
Any Viewport
Desktop, mobile, tablet, dark mode, custom width/height or built-in presets.
Full Page & PDF
Capture entire scrollable pages or export as PDF. Perfect for docs and reports.
SEO Audit
Check meta tags, Open Graph, Twitter cards, headings, structured data, and canonical URLs.
Performance Metrics
Capture Core Web Vitals (TTFB, FCP, LCP, CLS), network waterfall, and resource breakdown.
Smart Login
Auto-detect login forms, fill credentials, submit, and continue testing as an authenticated user.
Accessibility Tree
Inspect ARIA roles, headings, links, and interactive elements — no session required.
Visual Diff
Compare two screenshots pixel-by-pixel and get a diff overlay showing exactly what changed.
Batch & Cross-Browser
Screenshot multiple URLs in one call, or capture in Chromium, Firefox, and WebKit simultaneously.
API Keys
Scoped API keys with usage tracking and one-click revocation.

Simple pricing

Start free. Paid plans coming soon.

Free
$0forever
Unlimited screenshots
Most popular
Starter
$9/ month
2,000 screenshots / mo
Pro
$29/ month
10,000 screenshots / mo