OpenSift’s URL ingestion endpoint accepts user‑supplied URLs and fetches them unfiltered, enabling a server‑side request forgery vulnerability. An attacker that can supply a crafted URL can cause the OpenSift host process to contact internal or private network addresses, potentially exposing confidential data or services. The flaw is rooted in improper input validation (CWE‑20) and classic SSRF behavior (CWE‑918).

The issue affects OpenSift, a semantic‑search and generative‑AI study tool. Versions 1.1.2‑alpha and earlier are vulnerable. The problem was corrected in 1.1.3‑alpha. No other versions are reported as affected.

The CVSS score of 7.1 indicates moderate‑to‑high severity, while the EPSS of less than 1 % suggests a low current exploitation likelihood. The vulnerability is not listed in the CISA KEV catalog. Exploitation requires the attacker to reach the OpenSift ingestion endpoint with a malicious URL; the request is then performed by the host without authentication checks. If the endpoint is publicly reachable or accessible from untrusted networks, the path to internal resources is straightforward and does not require privileged credentials on the host. Official remediation is to upgrade to 1.1.3‑alpha or later. As an insecure workaround, the environment variable OPENSIFT_ALLOW_PRIVATE_URLS=true can be enabled to allow trusted private URLs, but this must be used with caution because it re‑introduces the SSRF risk for any incoming URLs.