A flaw in the LLMClient.chat function of the LLM Configuration Handler allows an attacker to manipulate the base_url parameter, which the application uses to send outgoing HTTP requests. This leads to server‑side request forgery, enabling the attacker to make the server reach arbitrary internal or external resources. The impact is that the application can be coerced into contacting malicious hosts, leaking internal network information or facilitating further network‑based attacks. The weakness aligns with CWE‑918, which describes insecure construction of URLs.

The vulnerability affects ptbsare anx‑calibre‑manager up to version 1.13.12. No other products or versions are listed in the CVE entry. If an organization is using a version within that range, the server is susceptible to SSRF via the exposed LLMClient.chat endpoint.

The CVSS score of 5.3 indicates moderate risk. EPSS is not available, and the vulnerability is not listed in the CISA KEV catalog, so no known active exploitation campaigns are documented in public sources. However, the description states that an exploit has been published and may be used, and that the attack can be initiated remotely. Because SSRF can be triggered through remote inputs to the LLMClient.chat API, the likely attack vector is a client that can invoke that endpoint with a crafted base_url. Mitigation requires patching or restricting the URL handling logic.