SupaMaus provides spatial context for AI agents, translating gestures, speech, and workflows into actionable information for models like Claude and Codex. It operates on-device, eliminating the need for API keys and ensuring data privacy.

• SupaMaus App (macOS): A menu-bar application that captures screenshots, gestures, spoken words, and window text with a single keystroke, sending them as spatial context to your AI agent. It supports on-device transcription and spatial context across multiple displays.
• SupaMaus for Browser: An AI cursor embedded within web products that provides in-app onboarding and support. It answers user questions on the live screen, highlights relevant controls, and guides users through workflows, reducing support tickets and improving activation rates.
• Workflow Recording: Record your actions, clicks, scrolls, keystrokes, and spoken words to create structured timelines that can be replayed by AI agents to automate tasks.
• Key Features:
  • Pixel-perfect gesture capture
  • On-device speech-to-text mapping to gestures
  • Spatial context across applications and displays
  • One-command integration with Claude Code and Codex via MCP
  • Workflow recording and automation
  • On-device processing with no API keys required
• Value Proposition: SupaMaus enriches AI agent context with visual and interactive information, leading to faster software development, fewer errors, and more efficient automated workflows. For users, it provides seamless in-app onboarding and support.