A server‑side request forgery (SSRF) flaw exists in the import‑from‑URL endpoint of HumanSignal Label Studio through 1.24.0.dev0, because the SSRF_PROTECTION_ENABLED flag is disabled by default. The vulnerability allows an authenticated user to supply any URL, including internal loopback addresses, which the server will fetch. This can expose internal services, cloud metadata endpoints, or other resources that should not be reachable from the public network. Depending on the exposed resource, an attacker could read sensitive data, manipulate services, or potentially execute code if the internal service is vulnerable. The impact is the ability to access and interact with internal network resources from an authenticated session.

The affected product is HumanSignal Label Studio, version 1.24.0.dev0 or earlier where the default configuration leaves SSRF_PROTECTION_ENABLED set to false. All installations that have not updated this setting or patched the software are susceptible, regardless of the upstream CPE representation.

The CVSS score of 6.5 indicates a moderate severity vulnerability. EPSS is not available and the vulnerability is not listed in the CISA KEV catalog. The likely attack vector requires an authenticated user with access to the import‑from‑URL endpoint; the attacker must supply a crafted URL targeting an internal endpoint. Successful exploitation would grant the attacker the same network privileges the server’s process has, enabling internal reconnaissance or further lateral movement. The absence of a public exploit or active campaigns suggests the exploitation probability is currently low, but the potential impact on internal assets warrants immediate remediation.