Clip

This gives Claude Code access to OpenAI's CLIP model for zero-shot image classification and image-text matching. You can use it to search images semantically, match text descriptions to visuals, or do basic content moderation without training custom models. It's built on the same foundation that trained on 400M image-text pairs and matches ResNet-50 performance. The skill template comes from davila7's collection and has passed security audits from Gen Agent Trust Hub, Socket, and Snyk. Best for quick prototypes where you need vision capabilities without the overhead of fine-tuning your own classifier.