Summary
Kimai: Timesheet PATCH/POST allows assigning to project outside user's team via query_builder OR-bypass
The Timesheet API PATCH /api/timesheets/{id} and POST /api/timesheets endpoints accept a user-supplied project ID and resolve it through a Symfony EntityType whose query_builder allows the submitted ID to satisfy the access predicate via an unconditional OR branch. As a result, any authenticated user can re-assign their own timesheet to any project in the database, including projects that belong to teams or customers they have no membership in and cannot otherwise see. The user can then read serialized project/customer details via GET /api/timesheets/{id}?full=true, leaking metadata (name, currency, customer hierarchy) that would otherwise be filtered out by the team ACL.
Details
Entry point, only ownership is checked in src/API/TimesheetController.php:317-355
#[IsGranted('edit', 'timesheet')]
#[Route(methods: ['PATCH'], path: '/{id}', name: 'patch_timesheet', requirements: ['id' => '\d+'])]
public function patchAction(Request $request, Timesheet $timesheet): Response
{
...
$form = $this->createForm(TimesheetApiEditForm::class, $timesheet, [...]);
$form->setData($timesheet);
$form->submit($request->request->all(), false);
if (false === $form->isValid()) { ... }
$this->service->saveTimesheet($timesheet);
...
}
src/Voter/TimesheetVoter.php:134-142:
if ($subject->getUser()?->getId() === $user->getId()) {
return $this->permissionManager->hasRolePermission($user, $permission . '_own_timesheet');
}
if (!$this->permissionManager->checkTeamAccessTimesheet($subject, $user)) {
return false;
}
For an own-timesheet, only edit_own_timesheet is required. The voter does not look at the new project being submitted; it only validates the existing record's ownership.
Form replays user-controlled project ID into the access query
src/Form/TimesheetEditForm.php:60-71:
$isNew = true;
if (isset($options['data']) && $options['data'] instanceof Timesheet) {
...
if (null !== $entry->getId()) {
$isNew = false;
}
...
}
$this->addProject($builder, $isNew, $project, $customer);
src/Form/FormTrait.php:59-100:
$builder->addEventListener(
FormEvents::PRE_SUBMIT,
function (FormEvent $event) use ($builder, $project, $customer, $isNew, $options): void {
$data = $event->getData();
$customer = \array_key_exists('customer', $data) && $data['customer'] !== '' ? $data['customer'] : null;
$project = \array_key_exists('project', $data) && $data['project'] !== '' ? $data['project'] : $project;
$event->getForm()->add('project', ProjectType::class, array_merge($options, [
'group_by' => null,
'query_builder' => function (ProjectRepository $repo) use ($builder, $project, $customer, $isNew) {
$project = \is_string($project) ? (int) $project : $project;
...
if ($isNew && \is_int($project)) {
$project = $repo->find($project);
if ($project !== null) {
if (!$project->getCustomer()->isVisible()) { ... $project = null; }
elseif (!$project->isVisible()) { $project = null; }
}
}
...
$query = new ProjectFormTypeQuery($project, $customer);
$query->setUser($builder->getOption('user'));
$query->setWithCustomer(true);
return $repo->getQueryBuilderForFormType($query);
},
]));
}
);
Two problems compound:
- The visibility re-check on line 73 is gated on
$isNew. For PATCH,$isNew = false, so the closure passes the attacker-supplied ID straight through. - Even when
$isNew = true(POST), the re-check only validatesisVisible(), it does not validate team membership.
The query-builder unconditionally accepts the submitted ID
src/Repository/ProjectRepository.php:150-208:
public function getQueryBuilderForFormType(ProjectFormTypeQuery $query): QueryBuilder
{
...
$mainQuery = $qb->expr()->andX();
$mainQuery->add($qb->expr()->eq('p.visible', ':visible'));
$mainQuery->add($qb->expr()->eq('c.visible', ':customer_visible'));
if (!$query->isIgnoreDate()) { ... }
if ($query->hasCustomers()) { ... }
$permissions = $this->getPermissionCriteria($qb, $query->getUser(), $query->getTeams());
if ($permissions->count() > 0) {
$mainQuery->add($permissions);
}
$outerQuery = $qb->expr()->orX();
if ($query->hasProjects()) {
$outerQuery->add($qb->expr()->in('p.id', ':project')); // <-- unconditional
$qb->setParameter('project', $query->getProjects());
}
...
$outerQuery->add($mainQuery);
$qb->andWhere($outerQuery);
return $qb;
}
The final WHERE clause is roughly:
WHERE (p.id IN (:project)) OR (p.visible AND c.visible AND <date> AND <team-ACL>)
Because :project is the submitted ID itself, the first branch matches unconditionally, completely bypassing the team-ACL applied by getPermissionCriteria. Symfony's EntityType happily resolves the foreign Project entity, the form passes validation, and the timesheet is persisted with the new project_id.
No downstream validation closes the gap
TimesheetService::saveTimesheet→updateTimesheet(src/Timesheet/TimesheetService.php:154-177) is explicitly documented as not validating.TimesheetBasicValidatoronly validates begin/end and project/activity coherence.TimesheetDeactivatedValidator::validateActivityAndProject(src/Validator/Constraints/TimesheetDeactivatedValidator.php:36-42) returns early for non-running existing timesheets.- No validator anywhere in the timesheet pipeline checks that the project's team membership intersects the acting user's teams.
A PoC was provided, but removed for security reasons.
Impact
- Integrity: any authenticated user can attribute their own tracked time to any project ID in the database, including projects belonging to teams/customers they cannot see. This pollutes per-project budgets, billing exports and reports for other teams. There is no in-app warning that records belonging to outsiders have been added.
- Confidentiality: by reading the timesheet back via
?full=true, the attacker obtains serialized project and customer details (name, currency, start/end dates, customer hierarchy) which would normally be filtered by the team ACL. - Privilege model: the
edit_own_timesheetpermission is part of the default ROLE_USER, so the bypass is reachable by every regular user without any administrator action.
The blast radius is bounded by what an attacker can persist (their own timesheet rows) and what the ?full=true serializer exposes, there is no direct ability to modify other teams' existing data.
Affected versions
Security releases
Kodem intelligence
Severity tells you how bad this could be in the worst case. It does not tell you whether you are exposed. Exploitability and impact are functions of runtime truth: whether the vulnerable code is present, reachable, and actually executes in your application. A vulnerable package can sit in your dependency tree and never run.
Kodem, an Intelligent Application Security platform, uses runtime intelligence to reveal which vulnerabilities actually execute in production, so teams prioritize the ones that genuinely matter. Kodem's runtime-powered SCA identifies whether this CVE is reachable in your applications.
Already deployed Kodem?
See it in your environmentNew to Kodem? Get a demo →Remediation advice
- The FormTrait was updated to only pass the project forward for new timesheets
- A new
TimesheetTeamAccessValidatorwas added, which checks ifprojectoractivitywere changed. If that is the case, the team access permission is checked first
Find out more at https://www.kimai.org/en/security/ghsa-vrr2-g9gh-c3jc
Frequently Asked Questions
- What is CVE-2026-52820? CVE-2026-52820 is a medium-severity security vulnerability in kimai/kimai (composer), affecting versions <= 2.56.0. It is fixed in 2.57.0.
- Which versions of kimai/kimai are affected by CVE-2026-52820? kimai/kimai (composer) versions <= 2.56.0 is affected.
- Is there a fix for CVE-2026-52820? Yes. CVE-2026-52820 is fixed in 2.57.0. Upgrade to this version or later.
- Is CVE-2026-52820 exploitable, and should I be worried? Whether CVE-2026-52820 is exploitable in your environment depends on whether the vulnerable code is present and reachable. A CVSS score is a worst-case rating; it does not account for your specific deployment, configuration, or usage patterns. Kodem, an Intelligent Application Security platform, uses runtime intelligence to show which vulnerabilities actually execute in production, so you can focus on the ones that represent real risk. Get a demo
- What actually determines whether CVE-2026-52820 is exploitable, and how bad it is? Exploitability and impact are not fixed properties of a CVE. They depend on runtime truth: whether the vulnerable code is present, reachable, and actually executes in your application. A high CVSS score on a dependency that never runs is not the same as real risk. Kodem, an Intelligent Application Security platform, uses runtime intelligence to reveal which vulnerabilities actually execute in production, so teams prioritize the ones that genuinely matter.
- How do I fix CVE-2026-52820? Upgrade
kimai/kimaito 2.57.0 or later.