Splunk AI Toolkit versions prior to 6.0.0 generate or store credentials for connected container services using predictable or hard‑coded values. Users with non‑admin or power roles can read these credentials, which can then be used to access the container services directly. This flaw allows attackers to gain unauthorized access to external container endpoints that are intended to be protected by proper authentication mechanisms, potentially leading to further compromise if those containers host sensitive data or services.

The vulnerability affects the Splunk AI Toolkit application across all installations where the version is older than 6.0.0. Any deployment of this application that connects to external container services is potentially impacted, regardless of the overall Splunk installation version.

The flaw carries a CVSS score of 5.4, indicating a moderate severity. No EPSS score is available, and the vulnerability is not listed in the CISA KEV catalog. Based on the description, the likely attack vector is through the Splunk web interface or API where the AI Toolkit is accessed, or by directly interacting with the AI Toolkit’s stored configuration. An attacker with read permissions to the application configuration can obtain the default credentials and then authenticate to the target container services, potentially escalating privileges or exfiltrating data. While no public exploit is documented, the use of hard‑coded credentials presents a clear avenue for credential abuse when combined with knowledge of the container service endpoints.