Executive brief
PraisonAI Platform, a system for managing multi-agent AI teams, contains a security flaw that allows users to access data belonging to other organizations or teams. By exploiting this vulnerability, an authorized user in one workspace can view, modify, or delete issues and tasks in any other workspace if they know the specific ID of that item. This could lead to the exposure of confidential project details, unauthorized changes to internal roadmaps, or the permanent deletion of team data.
Technical details
An Insecure Direct Object Reference (IDOR) exists in the issue CRUD endpoints (`GET / PATCH / DELETE /workspaces/{workspace_id}/issues/{issue_id}`). While the application verifies that a user is a member of the `workspace_id` provided in the URL, the underlying `IssueService` performs lookups using only the `issue_id` primary key without enforcing a workspace ownership constraint. Consequently, an authenticated user belonging to any workspace can manipulate issues in a different workspace by supplying their own valid workspace ID in the path and the target issue's UUID in the object parameter. This allows for unauthorized reading of issue descriptions, modification of project assignments, and deletion of records. The issue is resolved in version 0.1.4 by threading the workspace constraint through the service layer.
Affected products
- MervinPraison praisonai-platform < 0.1.4
Timeline
- 2026-05-19: patched: Fix committed to repository
- 2026-07-21: disclosed: CVE published to NVD