AutoGPT's POST /api/integrations/webhooks/{webhook_id}/ping endpoint does not verify that the webhook belongs to the authenticated user before fetching it by primary key. This improper access control (CWE-284) enables authenticated users to enumerate webhooks, leak OAuth provider information, and in some cases trigger webhook pings on behalf of other users. The vulnerability is identified as CVE-2026-56823 with a CVSS 3.1 base score of 5.4 (AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:N). No vendor advisory or patch details are provided, and the affected versions are unspecified.

An authenticated user can confirm the existence of arbitrary webhooks belonging to other users, leak information about the webhook's OAuth provider type, and potentially trigger webhook ping deliveries on behalf of other users. This leads to information disclosure and limited integrity impact. There is no indication of availability impact. The vulnerability requires low attack complexity and privileges but no user interaction.

Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. Until an official fix is available, restrict access to the webhook ping endpoint to only allow operations on webhooks owned by the authenticated user. Monitor for unusual webhook activity and consider implementing additional access controls to prevent unauthorized webhook enumeration and triggering.