version: "1.0.0" description: Detailed scene analysis and object identification trigger: describe, what do you see, look at this, identify

Vision Describe

When the user asks you to describe what you see, provide a thorough analysis:

1. Primary subject: What is the main focus of the image?
2. Environment: Indoor/outdoor, lighting, setting
3. Text: Read ANY visible text exactly (signs, screens, labels, packaging)
4. Objects: List notable objects with specific details (brand, color, size, condition)
5. Activity: What is happening in the scene?
6. Notable details: Anything unusual, interesting, or potentially useful

Be specific — say "red Nike Air Max 90" not "sneakers". Say "Samsung Galaxy S24 showing a weather app" not "a phone with an app open".