authentik Arbitrary Code Execution Vulnerability via Property Mapping or Expression Policy Test Endpoint

Vulnerability

A remote code execution vulnerability has been identified in authentik, an open-source identity provider. This issue affects versions 2021.3.1 prior to 2025.8.6, as well as 2025.10.4 and 2025.12.4. The vulnerability arises when delegated permissions are used, allowing a user with the 'Can view * Property Mapping' or 'Can view Expression Policy' permission to execute arbitrary code within the authentik server container. This exploitation occurs through the test endpoint, which is designed to preview how a property mapping or policy works.

Impact

Exploitation of this vulnerability allows users to execute arbitrary code with access to the entire authentik database and environment variables. This could lead to unauthorized permission escalation and complete takeover of the authentik instance.

Reproduction

The vulnerability can be reproduced by assigning a user the 'Can view * Property Mapping' or 'Can view Expression Policy' permissions. Once these permissions are granted, the user can send a POST request to the test endpoint of a property mapping or policy, including a payload that executes arbitrary code. The executed code can be crafted to, for example, read the output of the 'id' command and send it back in the response.

Remediation

Users can upgrade to authentik versions 2025.8.6, 2025.10.4, or 2025.12.4, all of which address this vulnerability. If an upgrade is not possible, it is recommended to revoke the 'Can view * Property Mapping' and 'Can view Expression Policy' permissions from users who do not require administrative rights, and to treat these permissions as equivalent to edit rights. Additionally, requests to the test endpoint can be blocked.

Added: Feb 12, 2026, 8:46 PM
Updated: Feb 12, 2026, 8:46 PM

Vulnerability Rating

Custom Algorithm
spread
2.2
impact
10.0
exploitability
6.4
remediation
8.3
relevance
2.7
threat
4.8
urgency
2.9
incentive
0.0

Our algorithm analyzes dozens of metrics to generate these 8 key vulnerability categories, which are then combined to calculate the overall risk score.