Executive brief
PraisonAI Platform, a system for managing multi-agent AI teams, contains a security flaw that allows users to access data from other organizations' workspaces. By exploiting this vulnerability, an authorized user in one workspace can read private comments and post new comments in any other workspace if they know the specific issue ID. This could lead to the exposure of sensitive internal discussions, customer data, or the injection of malicious links into other teams' workflows.
Technical details
An Insecure Direct Object Reference (IDOR) exists in the comment endpoints of the PraisonAI Platform. The application validates that a user belongs to the `workspace_id` provided in the URL path but fails to verify that the requested `issue_id` actually belongs to that workspace. Specifically, `CommentService.create` and `CommentService.list_for_issue` perform database operations using only the `issue_id` without a workspace-scoped join or filter. An authenticated attacker can provide their own valid `workspace_id` to pass the membership check and then supply a target `issue_id` from a different workspace to read or create comments. This vulnerability is patched in version 0.1.4.
Affected products
- MervinPraison praisonai-platform < 0.1.4
Timeline
- 2026-05-19: patched: Fix committed to repository
- 2026-07-21: disclosed: Security advisory published