GHSA-r32r-j5jq-3w4m
HIGHGHSA-r32r-j5jq-3w4m is a high-severity (CVSS 8.3) Improper Privilege Management vulnerability in vaultwarden. O3 Security confirms whether GHSA-r32r-j5jq-3w4m is actually reachable in your code before you act, and blocks exploitation at runtime until you patch.
Vaultwarden has Privilege Escalation via Bulk Permission Update to Unauthorized Collections by Manager
Blast Radius
vaultwardenReal-time download stats are indexed for npm and PyPI packages. This vulnerability affects crates.io packages — download data is not available via public APIs for these ecosystems.
Description
Summary
A Manager account (access_all=false) was able to escalate privileges by directly invoking the bulk-access API against collections that were not originally assigned to them.
The API allowed changing assigned=false to assigned=true, resulting in unauthorized access.
Additionally, prior to the bulk-access call, the regular single-update API correctly returned 401 Unauthorized for the same collection. After executing the bulk-access API, the same update API returned 200 OK, confirming an authorization gap at the HTTP level.
Description
-
The endpoint accepts
ManagerHeadersLooseand does not validate access rights for the specifiedcollectionIds. src/api/core/organizations.rs:551headers: ManagerHeadersLoose, -
The received
collection_idsare processed directly without per-collection authorization checks. src/api/core/organizations.rs:564for col_id in data.collection_ids { -
Existing group assignments for the collection are deleted. src/api/core/organizations.rs:583
CollectionGroup::delete_all_by_collection(&col_id, &conn).await?; -
Existing user assignments for the collection are deleted. src/api/core/organizations.rs:590
CollectionUser::delete_all_by_collection(&col_id, &conn).await?; -
By comparison, another bulk-processing endpoint performs per-collection validation using
from_loose. src/api/core/organizations.rs:787let headers = ManagerHeaders::from_loose(headers, &collections, &conn).await?; -
The actual access control logic is implemented in
can_access_collection, which is not invoked in the bulk-access endpoint. src/auth.rs:911if !Collection::can_access_collection(&h.membership, col_id, conn).await {
Preconditions
- The attacker possesses a valid Manager account within the target organization.
- The organization contains collections that are not assigned to the attacker.
- The attacker can authenticate through the standard API login process (Owner/Admin privileges are not required).
Steps to Reproduce
-
Log in as a Manager and obtain a Bearer token.
<img width="4016" height="1690" alt="image" src="https://github.com/user-attachments/assets/218f05e2-6a2e-4066-8f8d-6bbef1cc5858" /> -
Confirm the current values of
<img width="4026" height="1694" alt="image" src="https://github.com/user-attachments/assets/a6d2fc70-5370-4984-85bd-a6f74febdfa3" />assigned,manage,readOnly, andhidePasswordsfor the target collection. -
Verify that the standard update API returns 401 Unauthorized when attempting to modify the unassigned collection.
<img width="4030" height="1708" alt="image" src="https://github.com/user-attachments/assets/802f0d2b-d474-44d2-beef-b4f7f3335225" /> -
Invoke the bulk-access API, including:
<img width="4036" height="1120" alt="image" src="https://github.com/user-attachments/assets/1d3caa01-3ac2-4636-9ed0-189e5923c986" />collectionIdscontaining the target collectionuserscontaining the attacker’s ownmembership_idConfirm that the API returns 200 OK.
-
Re-run the standard update API. Confirm that it now succeeds and that the previously unauthorized modification is applied.
<img width="4040" height="1440" alt="image" src="https://github.com/user-attachments/assets/340e9676-d802-404c-b894-9986a176360a" />
Required Minimum Privileges
- Manager role within the target organization
(the issue occurs even when
access_all=false)
Attack Scenario
A delegated administrator or department-level Manager within an organization directly calls the API to add themselves to unauthorized collections and gain access to confidential information.
Because the bulk update process deletes and reassigns existing permissions, the attacker can also remove other users’ access, enabling denial-of-service or sabotage within the organization.
Potential Impact
- Confidentiality: Unauthorized access to sensitive information within restricted collections.
- Integrity: Unauthorized modification of collection permission settings and arbitrary changes to access controls.
- Availability: Deletion of existing assignments may cause legitimate users to lose access.
Affected Packages
| Ecosystem | Package | Vulnerable range | Fix |
|---|---|---|---|
| 🦀crates.io | vaultwarden | all versions | 1.35.4 |
Detection & mitigation playbook
Open-source dependencyDetect
Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for vaultwarden. 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 vaultwarden to 1.35.4 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms GHSA-r32r-j5jq-3w4m 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-r32r-j5jq-3w4m 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-r32r-j5jq-3w4m. 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-r32r-j5jq-3w4m in your dependencies?
O3 detects GHSA-r32r-j5jq-3w4m across crates.io dependencies and uses function-level reachability to confirm whether the vulnerable code path is actually reachable — not just present. No false positives.