This vulnerability occurs when the Network‑AI ApprovalInbox GET routes do not enforce the configured authorization check, allowing anyone who can reach the application to retrieve detailed approval request information. The exposed endpoints—such as /approvals/?status=all, /approvals/:id, /approvals/stats, and /approvals/sse—return the full ApprovalEntry payload, which includes action and target shell‑command strings, file paths, justifications, and risk levels. Additionally, all responses carry a hardcoded Access‑Control‑Allow‑Origin: * header, allowing cross‑origin disclosure from any website the operator visits. These leaks expose sensitive administrative approval data and shell command strings, facilitating planning of further attacks. The weakness corresponds to CWE‑862 (Missing Authorization).

The affected product is Jovancoding:Network‑AI, versions 5.12.2 through 5.13.3 are explicitly named in the description. Based on the description, it is inferred that no other minor or patch releases outside this range are affected, and no alternative versions are listed.

The CVSS score of 9.3 indicates a critical impact, while the EPSS score of less than 1% suggests a low probability of exploitation at present. The vulnerability is not listed in the CISA KEV catalog, but the lack of authentication allows arbitrary read of sensitive data. An attacker only needs network access to the exposed endpoints; no credentials are required. Exploitation involves sending simple HTTP GET requests to the vulnerable routes, receiving the approval payloads and optionally leaking shell command strings. The hardcoded Access‑Control‑Allow‑Origin: * header further enables cross‑origin disclosure, allowing a malicious website to read the data if the user visits it. This increases the potential impact to remote clients that load the application.