The vulnerability is a Server-Side Request Forgery (SSRF) bypass in pyLoad's download functionality. It exists because of an incomplete fix for a previous vulnerability (CVE-2026-33992).
The root cause is a combination of two factors:
- An insufficient SSRF check in the
BaseDownloader.download function, which only validates the IP address of the initial download URL.
- The configuration of the underlying
pycurl library in HTTPRequest.init_handle to automatically follow HTTP redirects (FOLLOWLOCATION=1).
An authenticated attacker with 'ADD' permissions can exploit this by providing a URL that points to a server under their control. This initial URL will pass the SSRF check if it resolves to a public IP address. The attacker's server then issues an HTTP redirect (e.g., 302) to a restricted internal address, such as a cloud metadata endpoint (169.254.169.254) or an internal service. Because pycurl is configured to follow redirects automatically, it will then make a request to the internal address, bypassing the security check entirely.
The identified vulnerable functions are BaseDownloader.download, which contains the flawed validation logic, and HTTPRequest.init_handle, which enables the insecure redirect following. During exploitation, a runtime profile would show calls to BaseDownloader.download followed by the internal workings of the HTTPRequest class as it processes the download and the subsequent malicious redirect.