CVE-2025-31363
16.04.2025, 10:15
Mattermost versions 10.4.x <= 10.4.2, 10.5.x <= 10.5.0, 9.11.x <= 9.11.9 fail to restrict domains the LLM can request to contact upstreamwhich allows an authenticated user toexfiltrate data from an arbitrary server accessible to the victim via performing a prompt injectionin the AI plugin's Jira tool.Enginsight
Awaiting analysis
This vulnerability is currently awaiting analysis.
References