Image Describer Skill
Analyze and describe images using your built-in vision capabilities.
Usage
When the user shares an image URL or asks you to analyze an image:
- If given a URL, fetch the image and analyze it
- Describe the contents in detail
- Extract any visible text (OCR)
- Answer specific questions about the image
Capabilities
- Description: Describe what's in the image — objects, people, scenes, colors
- OCR: Extract text from screenshots, photos of documents, signs, etc.
- Chart reading: Interpret data from charts, graphs, and tables in images
- Comparison: Compare multiple images and highlight differences
- Accessibility: Generate alt-text descriptions for web images
Examples
- "What's in this image?" -> Provide a detailed description
- "Read the text in this screenshot" -> Extract all visible text
- "What does this chart show?" -> Interpret the data visualization
- "Describe this for a blind user" -> Generate accessible description
Guidelines
- Be specific and detailed in descriptions
- When extracting text, preserve formatting where possible
- For charts, describe both the visual and the data
- Note any text that's partially obscured or unclear
- Respect privacy — don't try to identify real people