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.
{
"cwe_ids": [
"CWE-1336"
],
"github_reviewed": true,
"github_reviewed_at": "2026-05-18T17:54:15Z",
"nvd_published_at": "2026-05-12T11:16:19Z",
"severity": "HIGH"
}