Executive brief
The Drupal AI Agents module, which allows users to automate website tasks like creating content or fields using AI, contains a security flaw. An attacker with access to an AI agent could bypass security restrictions to view sensitive content they are not authorized to see. This could lead to the exposure of private website data or configuration details.
Technical details
A missing authorization vulnerability (CWE-862) exists in the Drupal AI Agents module. The module fails to sufficiently validate permissions when an AI agent's tool loads content entities. An authenticated attacker with access to a configured agent can exploit this to perform forceful browsing and access content they should not be permitted to view. The vulnerability is mitigated by the requirement that the attacker must have access to an agent specifically configured to use the affected tool. Patches are available in versions 1.1.4, 1.2.5, and 1.3.1.
Affected products
- Drupal AI Agents 0.0.0 to 1.1.4, 1.2.0 to 1.2.5, 1.3.0 to 1.3.1
Timeline
- 2026-06-24: advisory: Drupal security advisory SA-CONTRIB-2026-056 published
- 2026-07-10: disclosed: CVE-2026-13236 published to NVD
- 2026-07-06: patched: Version 1.3.1 and 1.3.2 released