altloom
An accessibility + alt-text drafting tool, built by yafira.

find the images your alt text forgot.

Paste in HTML. altloom flags every image with missing or weak alt text, then drafts starting-point candidates from whatever context it can find — filename, captions, existing alt — run through a word-level Markov chain.

What this doesn't do: altloom can't see your images — it only reads text. The drafts below are a starting point built from context clues, not a description of actual image content. Always look at the image and edit before you publish.