ACCESSIBILITY / A WORKED STARTING POINT
Review accessibility evidence without inventing a pass
Identify what supplied markup shows and what still needs keyboard, visual or assistive-technology checks.
Template reviewed
WHEN TO USE IT
You have markup or a small interaction record and need an honest starting point for testing.
Bring these inputs
- {{evidence}}
- The markup and any actual keyboard, visual or assistive-technology observations.
01 / THE PROMPT
A template you can inspect.
Review the supplied interface evidence for accessibility concerns.
Separate findings visible in the evidence from checks that still need to be performed. For each finding, identify the element, the supporting evidence and a small remedy. Do not claim a conformance level, keyboard result, contrast ratio or screen-reader behaviour that has not been tested. Do not treat an automated check as a complete audit. If you cite a standard, verify the exact criterion from its primary source.
Supplied evidence:
{{evidence}}
Copying does not run the prompt. Workbench opens an editable draft with the example inputs. It does not save or send it.
02 / THE EXAMPLE TARGET
What a useful result should contain.
This is a target for the worked example, not a recorded model response. Equivalent wording can be valid where the task allows it.
The icon-only button has no name in the supplied markup. Add an appropriate visible label or accessible name. The input lacks a persistent visible label in this excerpt, so add or verify its associated label. Keyboard focus, contrast, reflow and screen-reader behaviour remain untested. Do not award a WCAG conformance level.
03 / JUDGE THE RESULT
Check the answer, not the confidence.
- Findings point to the actual button and input in the excerpt.
- No fabricated contrast measurement or keyboard test appears.
- Missing surrounding context and the need for real interaction checks are explicit.
The worked example is a target to inspect, not a saved response from a model. Use the checks to judge an actual result.
This template was revised during the September prompt review. The earlier text remains in Git history.
Read the prompt collection review ↗