Peekaboo: AI Screenshot Capture for macOS Agents

macOS CLI and optional MCP server enabling AI agents to capture screenshots with optional visual question answering.

Peekaboo: AI Screenshot Capture for macOS Agents

Peekaboo is a command-line interface (CLI) tool designed for macOS that empowers AI agents to programmatically capture screenshots. It offers an optional MCP server component, facilitating seamless integration with AI agent frameworks. This tool is specifically built for developers and AI builders looking to equip their macOS-based agents with visual context gathering capabilities.

What it Does

Peekaboo allows AI agents to take screenshots of the user's macOS desktop. Beyond simple capture, it can optionally perform visual question answering (VQA) on the captured image. This means an agent can not only see what's on the screen but also ask specific questions about its content, such as identifying elements, reading text, or determining states. The tool is designed to be run from the command line, making it scriptable and easily integrated into automated workflows.

Key Features

macOS Screenshot Capture: Programmatically capture screenshots of the entire desktop or specific windows.
Optional MCP Server: Integrates with MCP (Multi-Agent Communication Protocol) for agent-to-agent communication and control.
Visual Question Answering (VQA): Enables agents to query the content of screenshots, extracting information and understanding visual context.
CLI-Based: Designed for command-line execution, allowing for easy scripting and automation.
Developer-Focused: Built with AI builders and developers in mind, prioritizing functionality and integration.

Who it's For

Peekaboo is intended for AI developers and researchers working on macOS-based AI agents. This includes individuals building agents that require visual understanding of the user interface, such as agents for automation, accessibility, or complex task execution. If your AI agent needs to interact with graphical applications or understand visual information on a macOS system, Peekaboo provides the foundational capability to capture and analyze screen content.