Know every vulnerabilitybefore it knows you.
DevGuard continuously monitors your dependencies and alerts you when CVEs like this one affect your stack — with real-time threat intelligence built for developers.
GHSA-7cm9-v848-cfh2
No affected components available
Summary
The blacklist (ban) note parameter in UserController::ajax_blackList_post() is stored in the database without sanitization and rendered into an HTML data-note attribute without escaping. An admin with blacklist privileges can inject arbitrary JavaScript that executes in the browser of any other admin who views the user management page.
Details
In modules/Users/Controllers/UserController.php, the ajax_blackList_post() method (line 344-362) accepts a note POST parameter with only a required validation rule:
// Line 347 — validation only checks 'required', no sanitization
$valData = (['note' => ['label' => lang('Backend.notes'), 'rules' => 'required'],
'uid' => ['label' => 'uid', 'rules' => 'required|is_natural_no_zero']]);
// Line 352 — raw user input passed directly to ban()
$user->ban($this->request->getPost('note'));
Shield's Bannable::ban() trait stores the message as-is:
// vendor/codeigniter4/shield/src/Traits/Bannable.php
public function ban(?string $message = null): self
{
$this->status = 'banned';
$this->status_message = $message; // No escaping
// ...
}
In the users() method (line 13-91), when building the DataTables response, the status_message is concatenated directly into HTML without escaping:
// Line 55 — esc() IS used here (correct)
$result->fullname = esc($result->firstname) . ' ' . esc($result->surname);
// Line 58-59 — NO esc() on status_message (vulnerable)
if ($result->status == 'banned'):
$result->actions .= '<button ... data-note="' . $result->status_message . '">'
The HTML string is returned as JSON (line 90) and DataTables renders it into the DOM. CSP is disabled ($CSPEnabled = false in App.php), and no SecureHeaders filter is applied.
PoC
Step 1 — Store XSS payload via ban endpoint:
curl -X POST 'https://TARGET/backend/users/blackList' \
-H 'X-Requested-With: XMLHttpRequest' \
-H 'Cookie: ci_session=ADMIN_SESSION_WITH_UPDATE_PERM' \
-d 'uid=2¬e=%22+onmouseover%3D%22alert(document.cookie)%22+x%3D%22'
Expected response: {"result":true,"error":{"type":"success","message":"..."}}
Step 2 — Trigger payload:
Any admin navigating to /backend/users will receive HTML containing:
<button ... data-note="" onmouseover="alert(document.cookie)" x="">
The XSS fires when the admin hovers over the blacklist button for the banned user.
Alternative immediate-execution payload:
note="><img src=x onerror=alert(document.cookie)>
Impact
- Session hijacking: An attacker with blacklist privileges can steal session cookies of other admins (including superadmins who view the user list but are themselves protected from being banned).
- Privilege escalation: A lower-privileged admin could use stolen superadmin sessions to gain full control.
- Persistent: The payload persists in the database and fires every time the user list is loaded, affecting all admins who view the page.
Recommended Fix
Wrap status_message with esc() to match the escaping already applied to other user fields on line 55:
// In users() method, line 58-59 — change:
$result->actions .= '<button type="button" class="btn btn-outline-dark btn-sm open-blacklist-modal"
data-id="' . $result->id . '" data-status="' . $result->status . '" data-note="' . esc($result->status_message) . '"><i
The vulnerability can be exploited over the network without needing physical access. It is easy for an attacker to exploit this vulnerability. An attacker needs high-level or administrative privileges. The attacker needs the user to perform some action, like clicking a link. The vulnerability can affect other systems as well, not just the initial system. There is a low impact on the confidentiality of the information. There is a low impact on the integrity of the data.
Exploitation attempts have been detected. Elevated vigilance and prompt remediation are advised.
The exploit probability is very low. The vulnerability is unlikely to be exploited in the next 30 days.
We did not find any exploit available. Neither in GitHub repositories nor in the Exploit-Database.
Browse More
Continuously monitor your dependencies and get alerted when vulnerabilities like this one affect your stack.
Checkout DevGuard