The python-zeep library contains a flaw where the Settings.forbid_external flag is defined but not enforced during parsing of WSDL or XSD documents, allowing an attacker to provide documents that import or include external resources. When a Zeep client loads such a document, it automatically resolves the references to attacker‑chosen HTTP or HTTPS URLs, thereby performing outbound requests from the host that runs the library. This results in a server‑side request forgery (SSRF) that can be used to discover internal network services, exfiltrate data, or perform other actions that compromise confidentiality and integrity of the target environment.

All versions of mvantellingen's python-zeep from 4.0.0 up to but not including 4.3.3 are vulnerable because forbid_external is not enforced. Versions 4.3.3 and later include the fix that enforces the flag during parsing. The vulnerability is triggered whenever the application loads an untrusted WSDL or XSD, which could be supplied by an external party or by a malicious internal actor.

The CVSS score of 5.9 indicates moderate severity, while an EPSS score of < 1% suggests a very low probability of exploitation. The vulnerability is not listed in CISA's KEV catalog. Exploitation requires an attacker to supply or otherwise control the source of a WSDL or XSD document that the application processes. The likely attack vector is therefore a supply‑chain or application‑level compromise rather than a publicly reachable SSRF, which reduces the breadth of impact but still allows the host to make arbitrary outbound requests.