Safe Axios request interceptor performs a hostname validation with isInternalAddress() before a later DNS lookup, creating a time‑of‑check/time‑of‑use window that allows DNS rebinding. An attacker can supply a URL that resolves to a public address during the initial validation and then to a loopback, private, link‑local, or metadata address when the HTTP connection is made, bypassing the guard. This flaw enables a server‑side request forgery that can reach internal resources hidden behind the app’s outbound filter. The weakness is classified as CWE‑918.

FastGPT, an open‑source large‑language‑model platform provided by Labrin, is impacted in all releases prior to version 4.15.2. The fix is included in release v4.15.2 and later. Any deployment using an older FastGPT version that exposes the safe‑Axios‑backed HTTP tool, workflow HTTP node, external file fetch, or server‑side integrations is affected.

The CVSS score of 6.3 indicates moderate severity. The EPSS score is not available, suggesting limited publicly known exploitation at this time, and the vulnerability is not listed in the CISA KEV catalog. However, the attack requires an authenticated user who can supply a URL to a safe‑Axios‑based endpoint. Once authenticated, the attacker can target internal network services that would normally be blocked by the internal address check. Although the exploitation requires legitimate access, the combination of SSRF with DNS rebinding and the potential to reach sensitive internal endpoints makes this a noteworthy risk that warrants prompt remediation.