CVE-2026-32715: AnythingLLM is an application that turns pieces of content into context that any LLM can use as references during chatting. In 1.11.1 and earlier, The two generic system-preferences endpoints allow...

Added to the CISA Known Exploited Vulnerabilities catalog on 16 Mar 2026. Vendor: AI/ML. Product: LLM. CVSS score: 3.8.

AnythingLLM is an application that turns pieces of content into context that any LLM can use as references during chatting. In 1.11.1 and earlier, The two generic system-preferences endpoints allow manager role access, while every other surface that touches the same settings is restricted to admi...

Required action: Review and patch if applicable to your AI infrastructure.