What problem does it solve? Classifying images or searching image collections normally requires labeled training data and model fine-tuning. CLIP removes that requirement by matching images to natural-language descriptions, enabling zero-shot classification, semantic image search, and content moderation out of the box. ## Core Features & Use Cases - Zero-Shot Image Classification: Classify images into arbitrary text categories without any training data, using models like ViT-B/32. - Semantic Image Search & Retrieval: Encode images and text into a shared embedding space for text-to-image and image-to-text search, with optional vector database integration (Chroma/FAISS). - Content Moderation: Score images against safety categories such as NSFW or violent content using softmax probabilities. - Use Case: Index a folder of product photos with CLIP embeddings, then let users search the catalog with queries like "a red sneaker on white background" and return the top matching images. ## Quick Start Use the clip skill to classify the attached image against the labels "a dog", "a cat", and "a car" and show the probability for each.
