CVE-2025-11864

7.3

NucleoidAI · Nucleoid

A server-side request forgery vulnerability exists in the Nucleoid outbound request handler, allowing remote unauthenticated attackers to manipulate network requests via the extension.apply function.

Executive summary

NucleoidAI Nucleoid is vulnerable to a server-side request forgery flaw that allows remote, unauthenticated attackers to force the application to perform unauthorized outbound requests.

Vulnerability

The vulnerability exists in the extension.apply function within the src/cluster.ts file. Unauthenticated attackers can manipulate arguments for host, IP, port, path, and headers to trigger server-side request forgery.

Business impact

Successful exploitation allows an attacker to interact with internal network resources or external services that the server can access, potentially bypassing perimeter security controls. With a CVSS score of 7.3, this high-severity flaw poses a significant risk to internal infrastructure confidentiality and integrity, as it facilitates unauthorized scanning or data exfiltration from the server environment.

Remediation

Immediate Action: Review the NucleoidAI vendor advisory for the release of a patched version and apply the update immediately upon availability.

Proactive Monitoring: Monitor outbound network traffic from the Nucleoid server for unusual destination IP addresses or unexpected HTTP request patterns that deviate from standard application behavior.

Compensating Controls: Implement strict egress filtering on the server hosting the application to prevent unauthorized outbound connections to internal or sensitive external networks.

Exploitation status

Public Exploit Available: Unknown.

Analyst recommendation

Given the ability for remote, unauthenticated attackers to abuse this functionality, immediate attention is required. Organizations should prioritize updating the Nucleoid component as soon as the vendor releases a fix and verify that egress traffic is restricted to known, trusted destinations to mitigate the potential for unauthorized network reconnaissance.

Sources

Originally found and disclosed by lakshay12311 (VulDB User), per the CVE Program record.