SlimSnap

SlimSnap transforms screenshots into structured JSON, providing visual context for command-line tools. Key features include: • Converts annotated screenshots to structured JSON • Drastically reduces token count compared to raw images • Provides deterministic layout with element coordinates and IDs • Includes built-in OCR to read labels and text • Operates entirely on-device for privacy and security This Mac application enhances the capabilities of coding assistants like Claude Code, Aider, and Codex CLI by allowing them to 'see' and interact with user interfaces described in text. Instead of typing lengthy descriptions of UI elements, users can simply capture, annotate, and copy a screen as JSON, which can be pasted directly into any text-based agent or terminal environment. This significantly streamlines workflows for debugging, UI development, and interaction with visual applications where traditional screenshots are impractical. SlimSnap is ideal for engineering teams, developers, and power users who frequently interact with command-line interfaces and coding assistants. It helps bridge the gap between visual information and text-based tools, ensuring clarity and efficiency when discussing user interfaces or debugging visual components. The open MIT schema allows for flexibility and integration into various development pipelines.