FastGPT versions before 4.14.17 contain an SSRF vulnerability (CWE-918) in the lafModule workflow node's fetchData function. This function uses axios to fetch URLs controlled by users but fails to properly enforce the internal network blocklist (isInternalAddress), enabling attackers to bypass SSRF protections and send HTTP requests to internal/private network addresses. Both unauthenticated attackers and authenticated users with App editing privileges can exploit this. The vulnerability has been addressed in FastGPT version 4.14.17.

An attacker exploiting this vulnerability can cause the FastGPT server to make arbitrary HTTP requests to internal or private network resources. This could potentially expose internal services or data not intended to be accessible externally. However, the CVSS score of 2.3 and the low severity rating indicate that the impact is limited and exploitation is not straightforward or highly damaging. There are no known exploits in the wild at this time.

This vulnerability is patched in FastGPT version 4.14.17. Users should upgrade to version 4.14.17 or later to remediate the issue. No other official remediation or temporary fixes are documented. Since this is not a cloud service, remediation depends on applying the vendor's patch.