Axiom Vision

If you're working with Apple's Vision framework, this pulls together the patterns you actually need: subject segmentation that excludes hands, pose detection with all 21 landmarks, text recognition modes, barcode scanning, and document processing including the new iOS 26 structured extraction. The diagnostic guide is genuinely useful because Vision has real gotchas around coordinate conversion (lower-left vs top-left origin), confidence thresholds, and main thread blocking on older devices. It also covers Visual Intelligence integration for camera features, which involves IntentValueQuery and SemanticContentDescriptor patterns that aren't obvious from the docs. Saves you from rediscovering why your OCR fails on certain languages or why subjects at frame edges don't detect properly.

Install to Claude Code

npx -y skills add charleswiltgen/axiom --skill axiom-vision --agent claude-code