What this skill does
Alt text is the kind of accessibility and SEO detail that everyone agrees matters and almost nobody fills in properly, because writing a good description for every single image is tedious at any real volume. This skill batch-generates accurate, specific alt text for every image you point it at — a folder of product photos, every image in a blog post — in one pass.
How it works
- Image analysis — reads each image and identifies the concrete, relevant content (not generic filler like "image of a person") — objects, text visible in the image, and the context implied by surrounding content if provided.
- Alt text drafting — writes alt text following accessibility best practice (concise, descriptive, no "image of" prefix, decorative images flagged as such rather than over-described) and, where relevant, naturally includes a target keyword if one is supplied.
- Batch output — returns a mapping of filename → alt text ready to paste into your CMS or HTML directly.
Requirements
- Works on Claude AI (any surface with image input support — claude.ai, Claude Code, or the API with vision).
Setup
- Unzip the package into your Skills directory, keeping the folder structure intact.
- Point it at a folder of images or attach them directly — see
SKILL.mdfor the exact invocation pattern and the optional target-keyword input.
What's included
SKILL.md— the skill definition and alt-text drafting logicexamples/sample-run.md— a worked example across a batch of product photos
