WCAG 2.2 checker
Find missing names and image alternatives in pasted HTML, then record checks on your rendered page. Your entries stay in this browser.
Selected markup checks
2 markup findings
Context reviews
2
Manual checks left
5
A zero-finding report covers these checks only. It does not certify accessibility.
The example has 2 markup findings. 5 manual checks remain.
Markup findings
| Location | Criterion | Finding |
|---|---|---|
| input at line 5 | 4.1.2 | This native control has no name in the supported markup checks. Add an associated label or another valid accessible name. |
| a at line 6 | 4.1.2 | This link has no name in the supported markup checks. Give it descriptive text or an accessible name. An image with empty alt text cannot name the link. |
All findings are shown.
Context reviews
| Location | Criterion | Review |
|---|---|---|
| img at line 2 | 1.1.1 | Check the image in context. Confirm that its alternative conveys its purpose, or that an empty alternative or decorative role is appropriate. |
| img at line 6 | 1.1.1 | Check the image in context. Confirm that its alternative conveys its purpose, or that an empty alternative or decorative role is appropriate. |
All context reviews are shown.
Entered colour pair
Not checked. Enter both colours to calculate contrast.
Automated checks and manual judgement
The WCAG 2.2 checker finds selected naming and image-alternative issues in pasted HTML. It also records manual checks on the rendered page. Each finding identifies a markup location and a relevant criterion.
For native controls, the name checks recognise associated labels, aria-label, aria-labelledby and supported native names. An image’s alternative can name an image-only link. A placeholder alone is insufficient for these checks.
Empty headings receive a finding. Skipped heading levels and repeated IDs receive a context review. A skipped level alone does not establish failure. WCAG 2.2 removed the old Parsing criterion, so duplicate IDs alone are not reported as that failure.
CSS visibility, generated content and the complete accessible-name algorithm remain outside these checks. A name found in markup can still be misleading. A snippet can also omit a naming reference that exists elsewhere on the page.
Use the manual answers after testing the rendered page. They record your judgement without changing the markup findings. Keyboard operation, focus and reflow need a browser. Screen-reader testing also reveals how controls behave during a purchase.
- On the rendered page, complete a purchase task using only the keyboard. Check every control, modal and menu.
- Follow focus through the rendered page. Check that the focused control is visible and is not entirely covered by other content.
- Check that alternatives communicate each image’s purpose and headings describe their sections. Review decorative images in context.
- Check that shoppers can see each field’s purpose and any input instructions. An accessible name alone does not establish a visible label.
- Inspect the rendered page at 320 CSS pixels wide. Check that ordinary content works without horizontal scrolling or lost information.
Include label targets in the same snippet.
HTML is parsed as data. Links and image URLs are not opened. Styles and scripts are not run.
Check a product page example
The fictional example contains a product image, a size selector and a quantity input. Its basket link contains an image with empty alt text.
The checker reports two findings. The quantity input has no supported name. The basket link also lacks a name because its image has an empty alternative. Both images receive a separate review of their meaning.
Add a label whose for attribute matches quantity. Give the basket link text such as “View basket”, or a suitable accessible name. Keep the empty image alternative when that image is decorative beside the link’s name.
Run the corrected snippet again, then test the complete product page. A clean local result leaves the manual checks and every untested criterion unresolved.
Use the report with WCAG
The Web Content Accessibility Guidelines (WCAG) define requirements for web accessibility. The name checks relate to WCAG 4.1.2, Name, Role, Value. W3C’s label technique H44 explains explicit control labels. The checker tests selected naming cases, not every requirement of 4.1.2.
Image-alternative findings relate to 1.1.1 and image technique H37. Empty-heading findings and heading reviews relate to 2.4.6, Headings and Labels. The meaning of an image or heading requires judgement.
The optional pair uses the WCAG relative-luminance formula. Criterion 1.4.3 sets AA minimums of 4.5:1 for normal text and 3:1 for large text, with exceptions. The threshold comparison uses the unrounded ratio.
Large text is at least 18 pt regular or 14 pt bold. Enter opaque colours only. Alpha colours, gradients and images need their rendered backgrounds assessed separately. Entering a passing pair does not verify that the page uses it.
The manual list covers keyboard operation (2.1.1), visible and unobscured focus (2.4.7 and 2.4.11), meaningful images and headings (1.1.1 and 2.4.6), visible labels (3.3.2) and reflow (1.4.10).
Download the report to assign repairs and retain your manual answers. All other criteria remain untested. The WCAG conformance requirements cover complete pages and processes. This report is not an accessibility certification.
The text report includes all findings, manual answers and untested scope.