image-vision

What problem does it solve?

When a user attaches an image to a conversation, they often need to understand the actual visual content of the photo — such as what is depicted, text present in the image, or the quality of the composition — rather than just basic file metadata like dimensions or file size. This Skill fills that gap by enabling AI assistants to actually "see" and analyze image content.

Core Features & Use Cases

• Full Visual Content Analysis: Describe scenes, subjects, colors, mood, text, and composition of any attached image.
• Custom Question Answering: Answer specific user questions about image content, such as critiquing an ad photo's effectiveness or checking if text on a sign is legible.
• Adjustable Detail Levels: Choose between low (fast, low-cost, for overall scene overview), high (full resolution, for fine text or detail), or auto (model-selected) processing to balance speed, cost, and accuracy.
• Use Case Example: A marketing professional attaches a draft social media ad photo and asks for feedback on its composition and visual appeal; this Skill analyzes the image and provides actionable critique.

Quick Start

Use the image-vision skill to analyze the attached product photo and tell me if the composition works for a social media ad.