Executive brief
FlowiseAI is an open-source platform used to build customized AI workflows. A security flaw allows logged-in users to move data resources between different organizational workspaces by manipulating internal ID fields. This could lead to unauthorized access to sensitive AI variables and a breakdown of data isolation between different teams or clients.
Technical details
A mass assignment vulnerability exists in the PUT /api/v1/variables/{variableId} endpoint of FlowiseAI. The backend fails to filter or validate server-controlled properties in the request body, such as workspaceId, createdDate, and updatedDate. An authenticated attacker can provide a different workspaceId in the JSON payload to reassign a variable to an arbitrary workspace, effectively bypassing tenant isolation. This occurs because the user-controlled input is directly mapped to the database entity without sufficient authorization checks. The issue is addressed in version 3.1.2.
Affected products
- FlowiseAI flowise <= 3.1.1
Timeline
- 2026-05-14: advisory: GitHub Advisory published
- 2026-05-14: patched: Version 3.1.2 released
- 2026-06-08: disclosed: NVD publication
References
- https://api.github.com/users/berkdedekarginoglu
- https://github.com/berkdedekarginoglu
- https://api.github.com/users/berkdedekarginoglu/gists%7B/gist_id%7D
- https://api.github.com/users/berkdedekarginoglu/repos
- https://avatars.githubusercontent.com/u/30373481?v=4
- https://api.github.com/users/berkdedekarginoglu/events%7B/privacy%7D