One prompt. Every post described.

Provide a clear, vivid, accessibility focused visual description of the content I have attached.

Write for a person who cannot see the image, video, graphic, screenshot, or document. Describe what a sighted viewer would understand from seeing it, including the overall scene, people, actions, expressions, clothing, objects, setting, composition, important colors, and any meaningful visual relationships.

Prioritize information that helps the reader understand the meaning, mood, context, and purpose of the visual rather than cataloging every visible detail.

Transcribe all important visible text exactly when legible. For charts, diagrams, maps, infographics, or data visualizations, explain the main message, relevant labels and values, patterns, comparisons, and takeaways rather than merely describing their appearance.

When describing people, describe observable characteristics that are relevant to understanding the visual, but do not guess identity, disability, ethnicity, gender identity, emotions, relationships, or other personal characteristics that cannot be established visually. You may describe apparent facial expressions, gestures, positioning, and interaction without claiming to know what a person is thinking or feeling.

Organize the description naturally, beginning with the most important overall information and moving into useful detail. Avoid phrases such as “image of,” “picture of,” or “as you can see.” Do not add information that is not visually supported.

Output two versions:

Short description: 1 to 3 sentences suitable for alt text.

Detailed visual description: A concise but rich description for someone who wants substantially the same visual context available to a sighted viewer.

If this is video, describe important changes, actions, scene transitions, on screen text, and visually meaningful moments in chronological order.