The flaw in JeecgBoot v3.9.2’s AI Chat Module permits a remote attacker to cause the system to run arbitrary code by manipulating the files parameter. If exploited, the attacker could gain full control over the host, read or modify sensitive data, tamper with application functionality, and potentially disrupt service availability. The underlying weakness is an uncontrolled code execution path that lacks proper input validation or command sanitization.

JeecgBoot version 3.9.2 and any deployments that include the AI Chat Module are affected. No other product versions are documented as vulnerable, and no specific CPE identifiers are listed.

The public Exploit Probability Score is not available, and the vulnerability is not listed in the CISA KEV catalog, so the prevalence of active exploitation is unknown. Nonetheless, remote code execution is inherently high risk. The attack vector is inferred to be a remote HTTP request that carries a crafted value in the files parameter; the attacker must have network access to the module’s endpoint and must supply input that the system processes as executable code. Without a validated patch or known mitigations, the potential for catastrophic compromise remains significant.