<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: GatekeeperQA</title>
    <description>The latest articles on DEV Community by GatekeeperQA (gatekeeperqa-team).</description>
    <link>https://dev.to/gatekeeperqa-team</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Forganization%2Fprofile_image%2F15005%2F2a9e5a6a-6541-4bb8-9285-2f07862c07ba.png</url>
      <title>DEV Community: GatekeeperQA</title>
      <link>https://dev.to/gatekeeperqa-team</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/gatekeeperqa-team"/>
    <language>en</language>
    <item>
      <title>A disappeared accessibility finding is not a verified fix</title>
      <dc:creator>GatekeeperQA</dc:creator>
      <pubDate>Mon, 26 Oct 2026 20:28:41 +0000</pubDate>
      <link>https://dev.to/gatekeeperqa-team/a-disappeared-accessibility-finding-is-not-a-verified-fix-4571</link>
      <guid>https://dev.to/gatekeeperqa-team/a-disappeared-accessibility-finding-is-not-a-verified-fix-4571</guid>
      <description>&lt;p&gt;A finding disappears from the next scan. Can your report call it fixed?&lt;/p&gt;

&lt;p&gt;Not yet. The element could have moved, the page could be incomplete, or the rule could now require manual review.&lt;/p&gt;

&lt;p&gt;We build GatekeeperQA. While auditing its retest workflow after community feedback, we found a mismatch: individual-finding retests correctly reported uncertainty when a later scan had an unresolved check for the same rule, but the bulk comparison could count that finding as “no longer detected.” We corrected the bulk result and added regression coverage.&lt;/p&gt;

&lt;h2&gt;
  
  
  One example, three different conclusions
&lt;/h2&gt;

&lt;p&gt;Consider an earlier &lt;code&gt;label&lt;/code&gt; violation on a form field. These are illustrative scenarios, not customer results:&lt;/p&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Later scan&lt;/th&gt;
&lt;th&gt;Safe conclusion&lt;/th&gt;
&lt;th&gt;Next step&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;The same finding is present&lt;/td&gt;
&lt;td&gt;Still detected&lt;/td&gt;
&lt;td&gt;Inspect the fix and retest&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;The rule has an unresolved manual check&lt;/td&gt;
&lt;td&gt;Uncertain / not verified&lt;/td&gt;
&lt;td&gt;Review the affected control manually&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;No matching finding, with comparable completed coverage&lt;/td&gt;
&lt;td&gt;No longer detected&lt;/td&gt;
&lt;td&gt;Verify that the same control remains and works&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;An unreviewed check must stay unreviewed. A missing page result or execution error cannot establish a passing comparison. “Fixed” needs evidence appropriate to the issue, not just an empty list.&lt;/p&gt;

&lt;h2&gt;
  
  
  Match the finding, not its row number
&lt;/h2&gt;

&lt;p&gt;Rows can move between reports. One approach is an exact-location identifier built from the page URL, rule and selector structure. It stays consistent when rows are reordered.&lt;/p&gt;

&lt;p&gt;That is not a permanent DOM identity: a changed selector, different URL or replaced element still needs investigation. Preserve the original report references instead of assuming two similar-looking rows describe the same control.&lt;/p&gt;

&lt;h2&gt;
  
  
  Give the next reviewer enough context
&lt;/h2&gt;

&lt;p&gt;For a useful handoff, record:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;The earlier and later report IDs and scan dates.&lt;/li&gt;
&lt;li&gt;The tested page and finding location.&lt;/li&gt;
&lt;li&gt;The automated comparison result, separate from human review.&lt;/li&gt;
&lt;li&gt;What a person tested, which input method they used, and what remains untested.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;Store structured report references and dates with attached retest evidence, and carry them into exports. Do not infer dates or version information that older text-only notes never recorded.&lt;/p&gt;

&lt;p&gt;Regression tests should cover row reordering, unresolved checks and incomplete page coverage. Passing those tests is software regression evidence, not proof of accessibility conformance. Keyboard, screen-reader and other relevant manual checks remain necessary.&lt;/p&gt;

&lt;h2&gt;
  
  
  A regression case worth keeping
&lt;/h2&gt;

&lt;p&gt;Given an earlier &lt;code&gt;label&lt;/code&gt; violation and a later unresolved &lt;code&gt;label&lt;/code&gt; check, assert that the bulk comparison reports uncertainty instead of counting the finding as no longer detected. Test the individual result and the bulk summary against the same fixture so they cannot quietly disagree.&lt;/p&gt;

&lt;p&gt;Disclosure: this article was generated with AI and checked against the project implementation and test evidence. The example scenarios are illustrative, not firsthand user-testing results.&lt;/p&gt;

</description>
      <category>a11y</category>
    </item>
    <item>
      <title>Your accessibility check failed in CI. Keep the evidence before fixing it.</title>
      <dc:creator>GatekeeperQA</dc:creator>
      <pubDate>Tue, 20 Oct 2026 17:00:58 +0000</pubDate>
      <link>https://dev.to/gatekeeperqa-team/your-accessibility-check-failed-in-ci-keep-the-evidence-before-fixing-it-3pf5</link>
      <guid>https://dev.to/gatekeeperqa-team/your-accessibility-check-failed-in-ci-keep-the-evidence-before-fixing-it-3pf5</guid>
      <description>&lt;p&gt;A red accessibility check tells you that a gate failed. The useful next questions are: which element triggered it, what page state was tested, and what evidence will you compare after the change?&lt;/p&gt;

&lt;p&gt;The workflow below keeps the original report, generates a separate retest report, and records manual observations alongside the automated result.&lt;/p&gt;

&lt;p&gt;Disclosure: I maintain GatekeeperQA, the tool used here. The local CLI runs without a hosted account. Hosted storage and review features are separate, with plan and retention limits.&lt;/p&gt;

&lt;h2&gt;
  
  
  Keep two reports, not one overwritten folder
&lt;/h2&gt;

&lt;p&gt;With Node.js 24 and an existing &lt;code&gt;package.json&lt;/code&gt;, install the pinned CLI:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight shell"&gt;&lt;code&gt;npm install --save-dev --save-exact @gatekeeperqa/cli@0.3.0
npx --no-install playwright install chromium
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Create &lt;code&gt;gatekeeperqa.json&lt;/code&gt;:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight json"&gt;&lt;code&gt;&lt;span class="p"&gt;{&lt;/span&gt;
  &lt;span class="nl"&gt;"urls"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"http://127.0.0.1:3000/"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
  &lt;span class="nl"&gt;"failOn"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"critical"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="s2"&gt;"serious"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
  &lt;span class="nl"&gt;"warnOn"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="p"&gt;[&lt;/span&gt;&lt;span class="s2"&gt;"moderate"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="s2"&gt;"minor"&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt; &lt;span class="s2"&gt;"unknown"&lt;/span&gt;&lt;span class="p"&gt;]&lt;/span&gt;&lt;span class="p"&gt;,&lt;/span&gt;
  &lt;span class="nl"&gt;"timeoutMs"&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="mi"&gt;30000&lt;/span&gt;
&lt;span class="p"&gt;}&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Start your app and wait until it is ready, then scan:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight shell"&gt;&lt;code&gt;npx --no-install gatekeeperqa --config gatekeeperqa.json --output reports/before
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;After implementing your change, run the same configuration against the updated app:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight shell"&gt;&lt;code&gt;npx --no-install gatekeeperqa --config gatekeeperqa.json --output reports/after
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;Use fresh directories for each attempt. A setup failure must not leave you interpreting an older report as the latest result.&lt;/p&gt;

&lt;p&gt;Open each directory's &lt;code&gt;report.html&lt;/code&gt;. Completed scans also produce JSON, Markdown and JUnit XML. The CLI does not directly export a PDF.&lt;/p&gt;

&lt;p&gt;This configuration checks an explicitly listed page in its initially loaded state. It does not crawl the site or exercise login, dialogs, checkout, or other interactions. Use only targets you own or have permission to test.&lt;/p&gt;

&lt;h2&gt;
  
  
  In GitHub Actions, save evidence before enforcing the gate
&lt;/h2&gt;

&lt;p&gt;If the scan exits nonzero, a normal subsequent step can be skipped. Separate evidence collection from the step that marks the job failed.&lt;/p&gt;

&lt;p&gt;GatekeeperQA's &lt;a href="https://gatekeeperqa.com/guides/developer-quick-start/" rel="noopener noreferrer"&gt;complete GitHub Actions setup&lt;/a&gt; installs the published package, captures the scan exit code, uploads available report artifacts, and then enforces that result.&lt;/p&gt;

&lt;p&gt;The important sequence is:&lt;/p&gt;

&lt;ol&gt;
&lt;li&gt;Run the scan and retain its exit code.&lt;/li&gt;
&lt;li&gt;Upload the evidence even when the quality gate fails.&lt;/li&gt;
&lt;li&gt;Fail the job when the captured result requires it.&lt;/li&gt;
&lt;/ol&gt;

&lt;p&gt;Do not turn a failed scan into a successful check just to keep the artifacts. Setup failures may have no report; inspect the logs in that case.&lt;/p&gt;

&lt;p&gt;In this CLI version, PASS and WARN exit with &lt;code&gt;0&lt;/code&gt;, FAIL with &lt;code&gt;1&lt;/code&gt;, and setup/configuration errors with &lt;code&gt;2&lt;/code&gt;. WARN still needs review. A gate result is not a compliance certificate.&lt;/p&gt;

&lt;p&gt;Also check what the workflow actually targets. Scanning production does not test code that exists only in a pull request. For PR validation, run the checked-out app or scan that PR's deployed preview after it is ready.&lt;/p&gt;

&lt;h2&gt;
  
  
  Update: use the GitHub Marketplace Action
&lt;/h2&gt;

&lt;p&gt;&lt;em&gt;October 22, 2026:&lt;/em&gt; &lt;a href="https://github.com/marketplace/actions/gatekeeperqa-accessibility-evidence" rel="noopener noreferrer"&gt;GatekeeperQA Accessibility Evidence v1.0.0 is now on GitHub Marketplace&lt;/a&gt;. The free, open-source Action uses the published CLI 0.3.0 with locked dependencies. No GatekeeperQA account or API key is required.&lt;/p&gt;

&lt;p&gt;Add this job to an existing workflow after choosing its trigger. Commit the configuration above, then install/start your app and wait for its health check where indicated. The Action reference below pins the tested release commit.&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight yaml"&gt;&lt;code&gt;&lt;span class="na"&gt;permissions&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;
  &lt;span class="na"&gt;contents&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s"&gt;read&lt;/span&gt;
&lt;span class="na"&gt;jobs&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;
  &lt;span class="na"&gt;accessibility&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;
    &lt;span class="na"&gt;runs-on&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s"&gt;ubuntu-24.04&lt;/span&gt;
    &lt;span class="na"&gt;timeout-minutes&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="mi"&gt;15&lt;/span&gt;
    &lt;span class="na"&gt;steps&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;
      &lt;span class="p"&gt;-&lt;/span&gt; &lt;span class="na"&gt;uses&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s"&gt;actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1&lt;/span&gt; &lt;span class="c1"&gt;# v7.0.1
&lt;/span&gt;        &lt;span class="na"&gt;with&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;
          &lt;span class="na"&gt;persist-credentials&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="kc"&gt;false&lt;/span&gt;
      &lt;span class="c1"&gt;# Install/start your app and wait for its health check here.
&lt;/span&gt;      &lt;span class="p"&gt;-&lt;/span&gt; &lt;span class="na"&gt;uses&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s"&gt;gatekeeperqa/accessibility-action@c7ba35dab80e9d6f65ecc36440a205df15c9235e&lt;/span&gt; &lt;span class="c1"&gt;# v1.0.0
&lt;/span&gt;        &lt;span class="na"&gt;with&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt;
          &lt;span class="na"&gt;config&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s"&gt;gatekeeperqa.json&lt;/span&gt;
          &lt;span class="na"&gt;artifact-name&lt;/span&gt;&lt;span class="p"&gt;:&lt;/span&gt; &lt;span class="s"&gt;accessibility-reports&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;After the run, open &lt;strong&gt;Actions → the workflow run → Artifacts&lt;/strong&gt;, download &lt;code&gt;accessibility-reports&lt;/code&gt;, unzip it, and open &lt;code&gt;report.html&lt;/code&gt;. Completed scans include JSON, Markdown, HTML and JUnit evidence. Artifacts are retained for seven days; use a unique artifact name for each invocation. Setup failures may produce no report—read the failed step logs.&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;PASS:&lt;/strong&gt; the configured automated gate passed; relevant manual checks remain.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;WARN:&lt;/strong&gt; the job succeeds, but warnings and unresolved checks still need review.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;FAIL:&lt;/strong&gt; the job fails; inspect both policy findings and page execution failures. An unreachable page fails the gate.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;ERROR:&lt;/strong&gt; the job fails; correct the setup or configuration and rerun.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;For untrusted PRs, use &lt;code&gt;pull_request&lt;/code&gt;, not &lt;code&gt;pull_request_target&lt;/code&gt;, and do not expose secrets. Review report contents and artifact access before scanning private pages. Requiring this check before merging needs separate branch-protection settings.&lt;/p&gt;

&lt;p&gt;&lt;a href="https://gatekeeperqa.com/guides/developer-quick-start/#marketplace-action" rel="noopener noreferrer"&gt;Full setup guide&lt;/a&gt; · &lt;a href="https://github.com/gatekeeperqa/accessibility-action/actions/runs/39889074807" rel="noopener noreferrer"&gt;Successful verification run&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Try it on a project you are authorized to test: &lt;strong&gt;Was installation clear, and did the report help you decide what to fix next?&lt;/strong&gt;&lt;/p&gt;

&lt;h2&gt;
  
  
  Retest the same element and state
&lt;/h2&gt;

&lt;p&gt;Consider an icon-only button:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight html"&gt;&lt;code&gt;&lt;span class="c"&gt;&amp;lt;!-- Before: no accessible name --&amp;gt;&lt;/span&gt;
&lt;span class="nt"&gt;&amp;lt;button&lt;/span&gt; &lt;span class="na"&gt;type&lt;/span&gt;&lt;span class="o"&gt;=&lt;/span&gt;&lt;span class="s"&gt;"button"&lt;/span&gt; &lt;span class="na"&gt;id&lt;/span&gt;&lt;span class="o"&gt;=&lt;/span&gt;&lt;span class="s"&gt;"checkout"&lt;/span&gt;&lt;span class="nt"&gt;&amp;gt;&lt;/span&gt;
  &lt;span class="nt"&gt;&amp;lt;svg&lt;/span&gt; &lt;span class="na"&gt;aria-hidden&lt;/span&gt;&lt;span class="o"&gt;=&lt;/span&gt;&lt;span class="s"&gt;"true"&lt;/span&gt; &lt;span class="na"&gt;focusable&lt;/span&gt;&lt;span class="o"&gt;=&lt;/span&gt;&lt;span class="s"&gt;"false"&lt;/span&gt;&lt;span class="nt"&gt;&amp;gt;&lt;/span&gt;&lt;span class="c"&gt;&amp;lt;!-- icon --&amp;gt;&lt;/span&gt;&lt;span class="nt"&gt;&amp;lt;/svg&amp;gt;&lt;/span&gt;
&lt;span class="nt"&gt;&amp;lt;/button&amp;gt;&lt;/span&gt;

&lt;span class="c"&gt;&amp;lt;!-- After: visible text names the action --&amp;gt;&lt;/span&gt;
&lt;span class="nt"&gt;&amp;lt;button&lt;/span&gt; &lt;span class="na"&gt;type&lt;/span&gt;&lt;span class="o"&gt;=&lt;/span&gt;&lt;span class="s"&gt;"button"&lt;/span&gt; &lt;span class="na"&gt;id&lt;/span&gt;&lt;span class="o"&gt;=&lt;/span&gt;&lt;span class="s"&gt;"checkout"&lt;/span&gt;&lt;span class="nt"&gt;&amp;gt;&lt;/span&gt;
  &lt;span class="nt"&gt;&amp;lt;svg&lt;/span&gt; &lt;span class="na"&gt;aria-hidden&lt;/span&gt;&lt;span class="o"&gt;=&lt;/span&gt;&lt;span class="s"&gt;"true"&lt;/span&gt; &lt;span class="na"&gt;focusable&lt;/span&gt;&lt;span class="o"&gt;=&lt;/span&gt;&lt;span class="s"&gt;"false"&lt;/span&gt;&lt;span class="nt"&gt;&amp;gt;&lt;/span&gt;&lt;span class="c"&gt;&amp;lt;!-- icon --&amp;gt;&lt;/span&gt;&lt;span class="nt"&gt;&amp;lt;/svg&amp;gt;&lt;/span&gt;
  Continue to checkout
&lt;span class="nt"&gt;&amp;lt;/button&amp;gt;&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;This is an illustrative fix, not a customer assessment. A subsequent scan may stop reporting the naming issue, but that alone does not verify keyboard activation, focus behavior, or the resulting checkout state.&lt;/p&gt;

&lt;p&gt;Before closing the ticket, confirm that the button still exists and was included in the retest. Record the builds, URL, state, engine version and configuration. If those changed, explain the difference.&lt;/p&gt;

&lt;p&gt;Write “no longer detected in this scan” when that is what you established. Record keyboard and screen-reader observations separately; leave checks you did not perform marked as not tested.&lt;/p&gt;

&lt;h2&gt;
  
  
  Give the next reviewer a reproducible handoff
&lt;/h2&gt;



&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;Original report and finding:
Original build:
Updated build:
Page and starting state:
Element location and rule:
Automated retest result:
Keyboard observations:
Screen reader/browser and observations:
Checks not performed:
Reviewer and date:
Remaining issues:
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;GatekeeperQA's hosted review workflow can keep saved notes and prepare an editable GitHub issue draft. Its private finding links open the specific element in a retained report. Those links require the report owner's sign-in; pasting one into a team ticket does not grant teammates access. Include appropriately redacted evidence that your intended reviewer can actually read.&lt;/p&gt;

&lt;p&gt;You can also keep this information in your existing tracker and retain the local reports. The workflow is useful independently of hosted storage.&lt;/p&gt;

&lt;h2&gt;
  
  
  Recorded walkthroughs: baseline, retest and evidence
&lt;/h2&gt;

&lt;p&gt;&lt;em&gt;November 16, 2026 resource update.&lt;/em&gt; These two owned demonstrations show the steps described above:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;a href="https://youtu.be/3YSl3PQFHF4" rel="noopener noreferrer"&gt;Browser scan, saved report, review notes and same-page retest (4½ minutes)&lt;/a&gt;. &lt;a href="https://gatekeeperqa.com/guides/tester-walkthrough/" rel="noopener noreferrer"&gt;Captions and transcript&lt;/a&gt;.&lt;/li&gt;
&lt;li&gt;
&lt;a href="https://youtu.be/9RMC9lHsEHQ" rel="noopener noreferrer"&gt;CLI setup, separate reports and a manual GitHub Actions run (5 minutes)&lt;/a&gt;. &lt;a href="https://gatekeeperqa.com/guides/developer-walkthrough/" rel="noopener noreferrer"&gt;Commands, captions, transcript and workflow download&lt;/a&gt;.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;The videos use AI narration. The CLI demonstration uses version 0.3.0; it does not show the newer 0.3.1 diagnostics. Its GitHub template uses a manual trigger, not an automatic pull-request test. Treat the recorded results as examples, keep your own baseline and retest evidence, and record manual checks separately.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Where does your team's accessibility handoff break down: locating the element, reproducing the state, or deciding what has really been verified?&lt;/strong&gt;&lt;/p&gt;

&lt;p&gt;&lt;em&gt;Prepared with AI assistance and checked against the project's CLI documentation and reporting implementation. Automated testing covers only part of accessibility; this is not a completed accessibility assessment.&lt;/em&gt;&lt;/p&gt;

</description>
      <category>a11y</category>
      <category>testing</category>
    </item>
    <item>
      <title>How to retest an accessibility fix and write useful evidence</title>
      <dc:creator>GatekeeperQA</dc:creator>
      <pubDate>Wed, 30 Sep 2026 16:35:05 +0000</pubDate>
      <link>https://dev.to/gatekeeperqa-team/how-to-retest-an-accessibility-fix-and-write-useful-evidence-o5f</link>
      <guid>https://dev.to/gatekeeperqa-team/how-to-retest-an-accessibility-fix-and-write-useful-evidence-o5f</guid>
      <description>&lt;p&gt;A developer fixes an unnamed button. The next scan no longer reports it. Before closing the ticket, what should you verify?&lt;/p&gt;

&lt;p&gt;Repeat the original check, test the interaction, and record the evidence. This fictional checkout example shows a focused retest workflow you can use in your existing issue tracker.&lt;/p&gt;

&lt;h2&gt;
  
  
  Start with one clear fix
&lt;/h2&gt;

&lt;p&gt;An icon-only button has no accessible name. Adding visible text makes its purpose available to users:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight html"&gt;&lt;code&gt;&lt;span class="c"&gt;&amp;lt;!-- Before --&amp;gt;&lt;/span&gt;
&lt;span class="nt"&gt;&amp;lt;button&lt;/span&gt; &lt;span class="na"&gt;type=&lt;/span&gt;&lt;span class="s"&gt;"button"&lt;/span&gt; &lt;span class="na"&gt;id=&lt;/span&gt;&lt;span class="s"&gt;"checkout"&lt;/span&gt;&lt;span class="nt"&gt;&amp;gt;&lt;/span&gt;
  &lt;span class="nt"&gt;&amp;lt;svg&lt;/span&gt; &lt;span class="na"&gt;aria-hidden=&lt;/span&gt;&lt;span class="s"&gt;"true"&lt;/span&gt; &lt;span class="na"&gt;focusable=&lt;/span&gt;&lt;span class="s"&gt;"false"&lt;/span&gt;&lt;span class="nt"&gt;&amp;gt;&lt;/span&gt;&lt;span class="c"&gt;&amp;lt;!-- Cart icon --&amp;gt;&lt;/span&gt;&lt;span class="nt"&gt;&amp;lt;/svg&amp;gt;&lt;/span&gt;
&lt;span class="nt"&gt;&amp;lt;/button&amp;gt;&lt;/span&gt;

&lt;span class="c"&gt;&amp;lt;!-- After --&amp;gt;&lt;/span&gt;
&lt;span class="nt"&gt;&amp;lt;button&lt;/span&gt; &lt;span class="na"&gt;type=&lt;/span&gt;&lt;span class="s"&gt;"button"&lt;/span&gt; &lt;span class="na"&gt;id=&lt;/span&gt;&lt;span class="s"&gt;"checkout"&lt;/span&gt;&lt;span class="nt"&gt;&amp;gt;&lt;/span&gt;
  &lt;span class="nt"&gt;&amp;lt;svg&lt;/span&gt; &lt;span class="na"&gt;aria-hidden=&lt;/span&gt;&lt;span class="s"&gt;"true"&lt;/span&gt; &lt;span class="na"&gt;focusable=&lt;/span&gt;&lt;span class="s"&gt;"false"&lt;/span&gt;&lt;span class="nt"&gt;&amp;gt;&lt;/span&gt;&lt;span class="c"&gt;&amp;lt;!-- Cart icon --&amp;gt;&lt;/span&gt;&lt;span class="nt"&gt;&amp;lt;/svg&amp;gt;&lt;/span&gt;
  Continue to checkout
&lt;span class="nt"&gt;&amp;lt;/button&amp;gt;&lt;/span&gt;
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;The label addresses the naming problem. The application still needs to handle activation correctly. W3C's &lt;a href="https://www.w3.org/WAI/ARIA/apg/patterns/button/" rel="noopener noreferrer"&gt;button pattern&lt;/a&gt; describes naming, keyboard activation and focus behavior.&lt;/p&gt;

&lt;h2&gt;
  
  
  1 Repeat the automated check
&lt;/h2&gt;

&lt;p&gt;Open the same page and state, such as a cart containing an item. Record the original and updated build identifiers. Keep the engine version and scan settings comparable, or document what changed.&lt;/p&gt;

&lt;p&gt;Confirm the retest completed and included the affected button. An empty cart, missing component or failed scan cannot verify the fix.&lt;/p&gt;

&lt;p&gt;A useful result is specific: “The accessible-name finding was not detected for the checkout button in the populated-cart state.” Use that wording only when your evidence supports it. Keep both reports so another reviewer can compare them.&lt;/p&gt;

&lt;h2&gt;
  
  
  2 Test the interaction
&lt;/h2&gt;

&lt;p&gt;For this button, check:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;
&lt;strong&gt;Keyboard:&lt;/strong&gt; Reach it with Tab and Shift+Tab. Check visible focus and navigation order. Test Enter and Space separately from the same starting state; each should perform the intended action once.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Screen reader:&lt;/strong&gt; Check the announced name and button role. Record the browser, screen reader and versions used.&lt;/li&gt;
&lt;li&gt;
&lt;strong&gt;Resulting state:&lt;/strong&gt; Follow the checkout action. If it opens a dialog, check focus inside it and after closing it. Exercise applicable loading and error states, including how the user recovers.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;These checks verify a specific interaction. They do not establish whole-site accessibility. W3C explains why &lt;a href="https://www.w3.org/WAI/test-evaluate/" rel="noopener noreferrer"&gt;automated tools need human evaluation&lt;/a&gt;.&lt;/p&gt;

&lt;h2&gt;
  
  
  3 Keep the evidence with the ticket
&lt;/h2&gt;

&lt;p&gt;Record each check as passed, failed or not tested according to what you actually verified. These statuses describe the page assessment. Keep automated results and manual observations separate.&lt;/p&gt;

&lt;p&gt;Use this compact template:&lt;br&gt;
&lt;/p&gt;

&lt;div class="highlight js-code-highlight"&gt;
&lt;pre class="highlight plaintext"&gt;&lt;code&gt;Page, component and starting state:
Original issue and evidence:
Expected behavior:
Fix build or commit:
Retest date and reviewer:
Browser and assistive technology versions:
Scan engine, settings and result:
Keyboard and screen-reader observations:
Other states checked and coverage gaps:
Remaining issues and linked tickets:
Decision against acceptance criteria:
&lt;/code&gt;&lt;/pre&gt;

&lt;/div&gt;



&lt;p&gt;If the label is fixed but focus remains broken, record that remaining issue and assign it. Close the ticket against its acceptance criteria, with enough detail for another tester to repeat the check. Redact personal information and secrets before sharing evidence.&lt;/p&gt;

&lt;h2&gt;
  
  
  Put the workflow to use
&lt;/h2&gt;

&lt;p&gt;I maintain GatekeeperQA. &lt;strong&gt;GatekeeperQA is ready to use for accessibility reporting and retest tracking&lt;/strong&gt;, with findings, reviewer notes, ticket copying and retest comparisons.&lt;/p&gt;

&lt;p&gt;Explore the free &lt;a href="https://gatekeeperqa.com/interactive-sample" rel="noopener noreferrer"&gt;interactive sample&lt;/a&gt; without signing up, or &lt;a href="https://scan.gatekeeperqa.com/" rel="noopener noreferrer"&gt;scan a public page&lt;/a&gt; you own or have permission to test. Usage and shared-capacity limits apply.&lt;/p&gt;

&lt;p&gt;Try it in your workflow. Your feedback helps shape what we build next.&lt;/p&gt;

&lt;p&gt;&lt;em&gt;Disclosure: Drafted with AI assistance. This fictional example is not a client assessment.&lt;/em&gt;&lt;/p&gt;

</description>
      <category>a11y</category>
    </item>
  </channel>
</rss>
