MCP Screenshot

📸 MCP ACS Screenshot Server

Give AI agents visual superpowers to see, analyze, and document your applications like senior UX designers.

This enterprise-grade MCP server transforms AI from code-only assistants into visual experts capable of UI analysis, accessibility auditing, documentation generation, and responsive design testing.

🔗 Repository

This package is now maintained in its own repository: [https://github.com/Digital-Defiance/mcp-screenshot](https://github.com/Digital-Defiance/mcp-screenshot)

This repository is part of the AI Capabilitites Suite on GitHub.

🤔 Why Do AI Agents Need Visual Capabilities?

AI agents today are powerful but visually blind:

• ❌ Can read HTML/CSS but can't see actual layouts
• ❌ Can suggest UI improvements without seeing the real user experience
• ❌ Can't detect accessibility issues like poor contrast or spacing
• ❌ Can't create visual documentation or bug reports
• ❌ Can't analyze responsive design across different screen sizes

Result: You're stuck manually creating screenshots, documentation, and visual analysis that AI should handle.

---

🎯 Revolutionary Use Cases

📚 "AI, create professional documentation"

You: "Document the checkout process with screenshots"

AI: *Captures each step, analyzes UI flow* AI: "Generated complete user guide with 6 annotated screenshots, highlighting key interactions and error states..."

🔍 "AI, audit this page for accessibility"

You: "Check if this form meets WCAG guidelines"

AI: *Captures form, analyzes visual elements* AI: "Found 4 accessibility issues: contrast ratio 2.1:1 (needs 4.5:1), missing focus indicators, touch targets too small..."

🐛 "AI, create a detailed bug report"

You: "The mobile navigation is broken"

AI: *Captures different viewport sizes* AI: "Created bug report with before/after screenshots, exact breakpoint where it fails, and CSS fix suggestions..."

🎨 "AI, compare these design variations"

You: "Which hero section design works better?"

AI: *Captures both versions, analyzes visual hierarchy* AI: "Version B has 28% better visual flow—CTA more prominent, text hierarchy clearer, better use of whitespace..."

📱 "AI, test responsive design"

You: "How does this look on different screen sizes?"

AI: *Captures multiple viewport sizes* AI: "Layout breaks at 768px—sidebar overlaps content. Here's the media query fix with visual proof..."

---

✨ What This Changes

Before: AI worked blind, relying on code descriptions

• ❌ "The button looks wrong" → AI guesses the issue
• ❌ "Create documentation" → AI writes generic text
• ❌ "Check accessibility" → AI only reviews code
• ❌ "Test responsive design" → AI can't see actual breakpoints

After: AI sees and analyzes your actual user interface

• ✅ Visual debugging - AI identifies exact pixel-level issues
• ✅ Smart documentation - AI creates guides with real screenshots and annotations
• ✅ Accessibility audits - AI measures actual contrast ratios and spacing
• ✅ Responsive testing - AI captures and compares different screen sizes
• ✅ Design analysis - AI evaluates visual hierarchy and user experience
• ✅ Professional reports - AI creates detailed visual evidence for bugs and improvements

---

🚀 Features

• Multi-format Support: PNG, JPEG, WebP, BMP with configurable quality
• Flexible Capture: Full screen, specific windows, or custom regions
• Privacy Protection: PII masking with OCR-based detection for emails, phone numbers, and credit cards
• Security Controls: Path validation, rate limiting, audit logging, and configurable policies
• Cross-platform: Linux (X11/Wayland), macOS, Windows with native APIs
• Multi-monitor Support: Capture from specific displays in multi-monitor setups
• Enterprise Security: Window exclusion, audit logging, rate limiting
• AI-Optimized: Structured responses perfect for AI agent workflows

Installation

NPM Installation

npm install @ai-capabilities-suite/mcp-screenshot

System Requirements

Linux:

• X11: imagemagick package (provides import command)
• Wayland: grim package

# Ubuntu/Debian

sudo apt-get install imagemagick grim

Fedora

sudo dnf install ImageMagick grim

Arch

sudo pacman -S imagemagick grim

macOS:

• Built-in screencapture command (no additional dependencies)
• Screen Recording permission required (System Preferences > Security & Privacy > Privacy > Screen Recording)

Windows:

• No additional dependencies required

MCP Configuration

Add to your MCP settings file (e.g., ~/.kiro/settings/mcp.json or .kiro/settings/mcp.json):

{

"mcpServers": { "screenshot": { "command": "node", "args": ["/path/to/mcp-screenshot/dist/cli.js"], "env": { "SCREENSHOT_ALLOWED_DIRS": "/home/user/screenshots,/tmp", "SCREENSHOT_MAX_CAPTURES_PER_MIN": "60", "SCREENSHOT_ENABLE_AUDIT_LOG": "true" } } } }

🛠️ 5 Professional MCP Tools

Purpose-built for AI agents to capture, analyze, and work with visual information:

The server exposes 5 comprehensive MCP tools that enable AI agents to see and understand your applications:

1. screenshot_capture_full

Capture full screen or specific display.

Parameters:

• display (string, optional): Display ID to capture (defaults to primary display)
• format (string, optional): Image format - png, jpeg, webp, or bmp (default: png)
• quality (number, optional): Compression quality 1-100 for lossy formats (default: 90)
• savePath (string, optional): File path to save screenshot (returns base64 if not provided)
• enablePIIMasking (boolean, optional): Enable PII detection and masking (default: false)

Example:

{

"name": "screenshot_capture_full", "arguments": { "format": "png", "savePath": "/home/user/screenshots/desktop.png", "enablePIIMasking": true } }

Response:

{

"status": "success", "filePath": "/home/user/screenshots/desktop.png", "metadata": { "width": 1920, "height": 1080, "format": "png", "fileSize": 245678, "timestamp": "2024-12-01T10:30:00.000Z", "display": { "id": "0", "name": "Primary Display", "resolution": { "width": 1920, "height": 1080 }, "position": { "x": 0, "y": 0 }, "isPrimary": true }, "piiMasking": { "emailsRedacted": 2, "phonesRedacted": 1, "creditCardsRedacted": 0, "customPatternsRedacted": 0 } } }

2. screenshot_capture_window

Capture specific application window by ID or title pattern.

Parameters:

• windowId (string, optional): Window identifier (use windowId or windowTitle)
• windowTitle (string, optional): Window title pattern to match (use windowId or windowTitle)
• includeFrame (boolean, optional): Include window frame and title bar (default: false)
• format (string, optional): Image format (default: png)
• quality (number, optional): Compression quality 1-100 (default: 90)
• savePath (string, optional): File path to save screenshot

Example:

{

"name": "screenshot_capture_window", "arguments": { "windowTitle": "Chrome", "includeFrame": false, "format": "jpeg", "quality": 85 } }

Response:

{

"status": "success", "data": "iVBORw0KGgoAAAANSUhEUgAA...", "mimeType": "image/jpeg", "metadata": { "width": 1280, "height": 720, "format": "jpeg", "fileSize": 89234, "timestamp": "2024-12-01T10:31:00.000Z", "window": { "id": "12345", "title": "Google Chrome", "processName": "chrome", "pid": 5678, "bounds": { "x": 100, "y": 100, "width": 1280, "height": 720 } } } }

3. screenshot_capture_region

Capture specific rectangular region of the screen.

Parameters:

• x (number, required): X coordinate of top-left corner
• y (number, required): Y coordinate of top-left corner
• width (number, required): Width of region in pixels
• height (number, required): Height of region in pixels
• format (string, optional): Image format (default: png)
• quality (number, optional): Compression quality 1-100 (default: 90)
• savePath (string, optional): File path to save screenshot

Example:

{

"name": "screenshot_capture_region", "arguments": { "x": 100, "y": 100, "width": 800, "height": 600, "format": "png" } }

Response:

{

"status": "success", "data": "iVBORw0KGgoAAAANSUhEUgAA...", "mimeType": "image/png", "metadata": { "width": 800, "height": 600, "format": "png", "fileSize": 123456, "timestamp": "2024-12-01T10:32:00.000Z", "region": { "x": 100, "y": 100, "width": 800, "height": 600 } } }

4. screenshot_list_displays

List all connected displays with resolution and position information.

Parameters: None

Example:

{

"name": "screenshot_list_displays", "arguments": {} }

Response:

{

"status": "success", "displays": [ { "id": "0", "name": "Primary Display", "resolution": { "width": 1920, "height": 1080 }, "position": { "x": 0, "y": 0 }, "isPrimary": true }, { "id": "1", "name": "Secondary Display", "resolution": { "width": 1920, "height": 1080 }, "position": { "x": 1920, "y": 0 }, "isPrimary": false } ] }

5. screenshot_list_windows

List all visible windows with title, process, and position information.

Parameters: None

Example:

{

"name": "screenshot_list_windows", "arguments": {} }

Response:

{

"status": "success", "windows": [ { "id": "12345", "title": "Google Chrome", "processName": "chrome", "pid": 5678, "bounds": { "x": 100, "y": 100, "width": 1280, "height": 720 }, "isMinimized": false }, { "id": "67890", "title": "Terminal", "processName": "gnome-terminal", "pid": 9012, "bounds": { "x": 200, "y": 200, "width": 800, "height": 600 }, "isMinimized": false } ] }

Security Configuration

The server enforces security policies to control screenshot operations. Configure via environment variables or security policy file.

Environment Variables

• SCREENSHOT_ALLOWED_DIRS: Comma-separated list of allowed directories for saving screenshots
• SCREENSHOT_MAX_CAPTURES_PER_MIN: Maximum captures per minute (default: 60)
• SCREENSHOT_ENABLE_AUDIT_LOG: Enable audit logging (default: true)
• SCREENSHOT_BLOCKED_WINDOWS: Comma-separated list of window title patterns to exclude

Security Policy File

Create a security-policy.json file:

{

"allowedDirectories": ["/home/user/screenshots", "/tmp/screenshots"], "blockedWindowPatterns": [ ".*Password.*", ".*1Password.*", ".*LastPass.*", ".*Bitwarden.*", ".*Authentication.*" ], "maxCapturesPerMinute": 60, "enableAuditLog": true }

Load the policy when starting the server:

import { MCPSc