Alt text is what screen-reader users hear in place of an image, and it is one of the most common accessibility failures on the web. This checker scans your HTML for every image and flags the mechanical problems — the ones you can catch without a manual review.
How it works
The pasted HTML is parsed in your browser using the built-in DOMParser, the
same engine the browser itself uses, so the img elements are found exactly as a
real page would see them. For each image the checker inspects the alt
attribute and applies five rules:
- Missing — no
altattribute at all. Assistive technology falls back to the filename or silence. This is the most serious issue. - Empty —
alt="". Correct for decorative images, flagged as info so you can confirm the image really is decorative. - Too long — over 125 characters. A WCAG-aligned guideline; long descriptions belong in a caption.
- Redundant prefix — starts with “image of”, “photo of”, and similar. Screen readers already announce the element as an image.
- Filename — the alt text equals the
srcfilename, a tell-tale sign of auto-generated placeholder text.
Worked example
<img src="harbour.jpg" alt="Sunset over the harbour at low tide"> <!-- ok -->
<img src="logo_v2.png"> <!-- missing -->
<img src="spacer.gif" alt=""> <!-- decorative -->
<img src="hero.jpg" alt="Photo of hero.jpg"> <!-- redundant + filename -->
The first image is ideal: concise and descriptive. The second has no alt at
all and is flagged red. The third is a deliberate decorative empty alt. The
fourth trips two rules at once.
Tips
- Describe the image’s purpose in context, not every pixel. A product photo’s alt might be the product name; a chart’s alt should state its key takeaway.
- Use empty
alt=""deliberately for decoration — but never omit the attribute. - This tool finds the mechanical problems; a human still needs to confirm the remaining alt text actually describes each image. Combine it with a real screen-reader pass for full coverage.