The vulnerability arises from the use of Python’s re.match() to validate the Origin header against the allow_origin_pat configuration in Jupyter Server. Because re.match() only anchors at the start of the string, a pattern intended to match a trusted domain (e.g., trusted.example.com) also matches any origin that begins with that domain and continues with additional characters (e.g., trusted.example.com.evil.com). This results in a CORS origin validation bypass, allowing an attacker controlling a domain that extends the trusted domain to make cross‑origin requests to the Jupyter Server API from an untrusted site. The attacker can therefore read or modify data and perform operations through the API, potentially compromising confidentiality, integrity, and availability of the user’s notebook environment. This weakness is identified as CWE-625 and CWE-777.

The defect is present in Jupyter Server version 2.17.0 and earlier. Versions 2.18.0 and newer contain the fix and are not affected.

The CVSS score of 7.6 indicates high severity. The EPSS score is shown as less than 1% (approximately 0.00029), indicating a very low but nonzero likelihood of exploitation, and the vulnerability is not listed in the CISA KEV catalog. The likely attack vector is web‑based: an attacker who controls a domain that satisfies the prefix condition can create malicious content that sends cross‑origin requests to the Jupyter Server API, bypassing the intended origin restriction. Successful exploitation requires the attacker to host a site under a domain that matches the trusted prefix in allow_origin_pat and to deliver traffic that invokes API calls to the server, enabling information disclosure or modification.