Claude is multimodal — it can analyze and describe images in addition to text.
Reference:
Vision documentation