Executive brief
PraisonAI Platform, a system for managing multi-agent AI teams, contains a security flaw that allows any authorized workspace member to take full control of their workspace. By exploiting a weakness in how the platform checks user permissions, a low-privilege user can promote themselves to an 'owner' role. Once escalated, the attacker can add or remove other users, delete the workspace, or access sensitive data, effectively locking out the original owners and disrupting operations.
Technical details
A vertical privilege escalation vulnerability exists in PraisonAI Platform due to improper authorization checks in workspace management routes. The `require_workspace_member` dependency in `praisonai_platform/api/deps.py` defaults to a `min_role` of 'member'. Several administrative endpoints in `workspaces.py` (including PATCH, POST, and DELETE methods for workspace and member management) utilize this dependency without overriding the default role requirement. Consequently, any authenticated user with basic workspace access can call these endpoints to modify their own role to 'owner', add new owners, or remove existing members, including the original workspace creator. The issue is resolved in version 0.1.4 by enforcing stricter role requirements (admin/owner) on management routes.
Affected products
- MervinPraison PraisonAI Platform < 0.1.4
Timeline
- 2026-05-19: patched: Fix committed and advisory published by maintainer
- 2026-07-21: disclosed: CVE published to NVD