Find the starting point
Separate a missing alt attribute from an explicitly empty one and an existing text value. None of these observations, on its own, proves that an image is accessible.
DIGITAL ACCESSIBILITY / CONTENT REVIEW
Turn a pasted HTML fragment into a working image inventory. Compare existing alternatives, draft changes and record the decisions that still need a person.
Separate a missing alt attribute from an explicitly empty one and an existing text value. None of these observations, on its own, proves that an image is accessible.
Use the source context to ask what a reader needs to understand or do. Verify the actual image and its page before accepting any alternative.
Export every image, draft, decision and note, including records hidden by a filter. Original observations remain separate from your proposed changes.
Scope: HTML img and input type="image" elements in the pasted fragment. CSS images, inline SVG, canvas, embedded documents, nested template contents and images created by scripts are outside this inventory. This tool does not visit a website, display images, calculate accessible names or certify conformance.
Paste a fragment from your own page source. It is parsed as inactive markup: scripts are not run and image addresses are not opened. Your source and notes stay in this tab. There is no automatic saving; export your work before leaving.
The source has changed. These are the previous observations and drafts. Analyze the new source, or restore the previous source, before editing or exporting.
Exports contain all records, image attributes and your notes, not the full pasted HTML. CSV cells that could be interpreted as spreadsheet formulas receive a protective apostrophe.
No images match this view. Choose another view to continue.
aria-label is a cue to investigate, not proof of a useful name.Need help choosing words? Use the alt-text lab. Record broader findings in the digital testing worksheet.
References checked 15 September 2026. A browser may repair malformed HTML while parsing; the inventory reflects that parsed fragment. Text excerpts are limited to 500 characters, are not rendered text or accessible names, and cannot establish whether an image alternative is equivalent. Review statuses are your decisions, not automated pass or fail results.