Prompt Injection via Memory Poisoning in PromptChatMemoryAdvisor
21Vexday Risk Score
No sign of exploitation. No public exploitation artifact known so far.
ssvc Trackcvss 8.2epss 0.2%
exploitation probability
0.2%top 88% of all CVEs
observed exploitation
nono source reports it
A malicious user could craft input that is stored in conversation memory and later interpreted by the model in an unintended way. Applications using the affected advisor with user-controlled input may be susceptible to manipulation of model behavior across conversation turns.
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:H/A:N
Affected products
VMware · Spring AI