Fill in missing image alt text

Missing alt text is the accessibility and SEO debt every media library carries, because nobody types it in at upload time. Fixing hundreds of images by hand never happens. This turns it into a review job instead of a typing job: the agent drafts the alt text, you approve it.

Help me fill in the missing image alt text in the media library. First, find
every image with an empty alt field and list them out: a thumbnail or filename
I can recognize, plus the alt text you'd suggest based on what the image
actually shows, kept short and descriptive. Change nothing yet. I'll go through
your suggestions and tell you which to keep, tweak, or skip. Then write only
the ones I approved, a few at a time, and give me back a list of what you set
on each image.

Why this is safe to run

Finding the empty-alt images is a read, so you get the full list and every proposed description before anything saves. The write reaches only the alt text field on the attachment itself: it does not alter the image file, and it does not touch any post that embeds the image. The agent is asked to describe what the image shows rather than to invent claims about it, and you keep, tweak, or skip each suggestion, so nothing goes in that you did not read first.

Already connected? Ask your agent for the “fill in missing image alt text” recipe and it runs the same find-then-approve-then-write flow on your own library.