An attacker with only a GitHub account can plant a JavaScript payload in a craftcms/cms issue title. When a Craft admin uses the CraftSupport widget’s "Give feedback" screen and types a search term that returns the poisoned issue, the payload executes in the admin’s control panel session.
No control panel account or elevated privileges are required on the attacker’s side.
CraftSupportWidget.js lines 382-392:
$('<a>', {
href: this.getSearchResultUrl(results[i]),
target: '_blank',
html:
'<span class="status ' +
this.getSearchResultStatus(results[i]) +
'"></span>' +
this.getSearchResultText(results[i]),
})
FeedbackScreen.getSearchResultText (line 669-671) returns result.title verbatim from the GitHub API response. The jQuery html: option sets the element’s innerHTML, so a title containing <img src=x onerror=...> executes immediately on render.
The GitHub API returns issue titles as raw JSON strings with no HTML encoding. The widget makes this request directly from the browser, without a Craft proxy or any sanitization step.
HelpScreen (Stack Exchange) is not affected because the Stack Exchange API HTML-encodes titles before returning them.
Plant (attacker, GitHub account only):
https://github.com/craftcms/cms/issues/new.<img src=x onerror=alert(document.domain)> cannot upload files
Trigger (victim, Craft admin):
cannot upload files in the search box.alert(document.domain) fires in the admin's session.XSS in the admin control panel session. The payload has access to Craft.csrfTokenName and Craft.csrfTokenValue and can send same-origin action requests as the admin without any further interaction.
https://github.com/craftcms/cms/commit/6bbb66038a268552180ca5c8eed9f46ea25a4417
{
"cwe_ids": [
"CWE-79"
],
"github_reviewed": true,
"github_reviewed_at": "2026-07-06T21:29:16Z",
"nvd_published_at": "2026-07-01T23:16:52Z",
"severity": "HIGH"
}