The vulnerability resides in Flowise’s handling of DatasetRow creation and updates, where mass‑assignment lacks proper workspace isolation. An authenticated user can manipulate request data to inject or modify DatasetRow objects belonging to another workspace, thereby gaining unauthorized access to, altering, or deleting that workspace’s data. This flaw directly jeopardizes the confidentiality, integrity, and availability of data managed by separate workspaces and is classified as CWE‑915.

Any deployment of Flowise AI’s Flowise application running a version older than 3.1.2 is affected, including community and self‑hosted instances. Users of earlier releases should verify the version of Flowise installed and consider reducing the scope of API access where possible.

The CVSS score of 7.7 indicates a high severity level. While no EPSS score is currently available, the documented exploit path does not require special privileges beyond legitimate workspace membership, and the flaw can be exercised via standard API or UI interactions. The vulnerability is not listed in the CISA KEV catalog, but its impact on data isolation means it should be treated with urgency. Attackers could target any user with workspace access and leverage API endpoints that process DatasetRow data.