MCP Screenshot Server

name: mcp-screenshot displayName: MCP Screenshot Server description: Cross-platform screenshot capture server for AI agents with multi-format support, PII masking, and security controls version: 1.0.2 author: name: Digital Defiance email: info@digitaldefiance.org url: https://digitaldefiance.org license: MIT homepage: https://github.com/digital-defiance/ai-capabilities-suite/tree/main/packages/mcp-screenshot repository: type: git url: https://github.com/digital-defiance/ai-capabilities-suite.git directory: packages/mcp-screenshot # Docker configuration docker: image: digidefiance/mcp-screenshot tag: latest alternativeImages: - ghcr.io/digital-defiance/mcp-screenshot:latest platforms: - linux/amd64 - linux/arm64 # Categories and tags categories: - productivity - utilities - development-tools tags: - screenshot - screen-capture - image-processing - pii-masking - privacy - security - cross-platform - multi-monitor - window-capture - region-capture # Supported platforms platforms: - linux - macos - windows # MCP tools provided tools: - name: screenshot_capture_full description: Capture full screen or specific display category: capture parameters: - name: display type: string required: false description: Display ID for multi-monitor setups - name: format type: string required: true enum: [png, jpeg, webp, bmp] description: Image format - name: quality type: number required: false min: 1 max: 100 description: Compression quality for lossy formats - name: savePath type: string required: false description: File path to save screenshot - name: enablePIIMasking type: boolean required: false default: false description: Enable PII detection and masking - name: screenshot_capture_window description: Capture specific application window category: capture parameters: - name: windowId type: string required: false description: Window identifier - name: windowTitle type: string required: false description: Window title pattern - name: includeFrame type: boolean required: false default: false description: Include window frame and title bar - name: format type: string required: true enum: [png, jpeg, webp, bmp] description: Image format - name: screenshot_capture_region description: Capture rectangular screen region category: capture parameters: - name: x type: number required: true min: 0 description: X coordinate in pixels - name: y type: number required: true min: 0 description: Y coordinate in pixels - name: width type: number required: true min: 1 description: Region width in pixels - name: height type: number required: true min: 1 description: Region height in pixels - name: format type: string required: true enum: [png, jpeg, webp, bmp] description: Image format - name: screenshot_list_displays description: List all connected displays with metadata category: information parameters: [] - name: screenshot_list_windows description: List all visible windows with metadata category: information parameters: [] # Features features: - Cross-platform screenshot capture (Linux, macOS, Windows) - Multiple capture modes (full screen, window, region) - Multi-format support (PNG, JPEG, WebP, BMP) - Configurable compression and quality settings - PII detection and masking with OCR - Multi-monitor support - Window enumeration and filtering - Security policies and path validation - Rate limiting and audit logging - Base64 encoding for in-memory transfer # Use cases useCases: - Visual debugging and UI analysis - Documentation and screenshot generation - Automated testing and visual regression - Screen monitoring and recording - Privacy-aware screenshot capture # Requirements requirements: node: ">=18.0.0" docker: ">=20.10.0" # Environment variables environment: - name: NODE_ENV description: Node environment (production/development) default: production required: false - name: TESSDATA_PREFIX description: Path to Tesseract training data default: /usr/share/tessdata required: false # Volumes volumes: - path: /app/screenshots description: Directory for saved screenshots required: false - path: /app/config description: Configuration directory required: false # Configuration example configuration: command: docker args: - run - -i - --rm - -v - ${PWD}/screenshots:/app/screenshots - -v - ${PWD}/config.json:/app/config/config.json:ro - digidefiance/mcp-screenshot:latest # Security security: runAsNonRoot: true readOnlyRootFilesystem: true capabilities: drop: - ALL securityOpt: - no-new-privileges:true # Resource limits resources: limits: cpu: "2" memory: 2Gi requests: cpu: "0.5" memory: 512Mi # Health check healthCheck: enabled: true interval: 30s timeout: 10s retries: 3 # Documentation documentation: readme: https://github.com/digital-defiance/ai-capabilities-suite/blob/main/packages/mcp-screenshot/README.md api: https://github.com/digital-defiance/ai-capabilities-suite/blob/main/packages/mcp-screenshot/API.md configuration: https://github.com/digital-defiance/ai-capabilities-suite/blob/main/packages/mcp-screenshot/CONFIGURATION.md examples: https://github.com/digital-defiance/ai-capabilities-suite/blob/main/packages/mcp-screenshot/EXAMPLES.md # Support support: issues: https://github.com/digital-defiance/ai-capabilities-suite/issues email: info@digitaldefiance.org documentation: https://github.com/digital-defiance/ai-capabilities-suite/tree/main/packages/mcp-screenshot