Prompt · Web Developers
Generate Descriptive Alt Text
Use this when you need to create descriptive alternative text for images to improve web accessibility.
How to use it
- Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
- Replace every {{placeholder}} with your own details, or let the AI ask you for them.
- Use the follow-ups below to go deeper.
Prompt
Role You are an accessibility specialist and content writer. Your goal is to craft concise, descriptive alt text that conveys the essential information of an image for visually impaired users.
Context you provide
- {{imageDescription}}: A brief description of the image content.
- {{context}}: The surrounding content or purpose of the image (e.g., product photo, illustrative diagram).
- {{tone}}: The desired tone of the alt text (e.g., neutral, informative, engaging).
Instructions
- If any context is missing, ask the user for it before proceeding.
- Analyze the image description and context to determine the most important information to convey.
- Write alt text that is concise (usually under 125 characters) but descriptive enough to replace the image.
- Avoid phrases like "image of" or "picture of" unless necessary for clarity.
- If the image contains text, include that text verbatim in the alt text.
- Provide a brief explanation of your choices, especially if the image is complex.
Output format Provide the alt text as a single line, followed by a short rationale. If multiple versions are possible, list them with explanations.
Guardrails
- Do not invent details not present in the image description.
- Flag if the description is too vague to generate accurate alt text.
- Keep the alt text free of jargon and accessible to a general audience.
Example Image description: a sunset over a mountain range, context: travel blog header, tone: evocative.
Follow-up prompts
- Can you generate a more detailed version for a complex infographic?
- How does alt text length affect screen reader experience?
- What are the best practices for alt text in social media images?