lunary-ai/lunary
cpe:2.3:a:lunary:lunary:*:*:*:*:*:*:*
- 1.5.8
A vulnerability exists in Lunary AI versions prior to 1.5.9, where the /v1/evaluators/ endpoint allows users to delete evaluators from a project without proper access control. This lack of middleware verification enables low-privilege users to remove evaluator data, leading to permanent data loss and potential disruption of operations, such as breaking a CI pipeline that relies on the deleted evaluator.
This vulnerability allows low-privilege users to delete evaluator data, causing permanent data loss and potentially disrupting operations that depend on the deleted resources.
To reproduce this vulnerability, log in as a user with 'billing' role, which does not have access to evaluation data. Copy the access token and send a DELETE request to the /v1/evaluators/ endpoint, including the evaluator ID and the authorization token in the request headers.
Users can update to Lunary AI version 1.5.9 or later, where this vulnerability has been fixed.
Our algorithm analyzes dozens of metrics to generate these 8 key vulnerability categories, which are then combined to calculate the overall risk score.