SSRF Vulnerability in Next AI Draw.io

Next AI Draw.io versions 0.4.16 and earlier contain a server-side request forgery (SSRF) vulnerability in the POST /api/parse-url endpoint. The vulnerability arises because the application performs hostname validation using static string pattern matching rather than resolving the provided input via DNS. This allows an unauthenticated attacker to supply a crafted hostname that bypasses the initial string-based filter but resolves to internal infrastructure addresses upon execution. Successful exploitation enables an attacker to reach arbitrary internal HTTP services, potentially exfiltrating sensitive data, internal service responses, or cloud instance metadata (e.g., AWS/GCP/Azure IMDS). The issue was identified as a security risk where the lack of proper DNS-based validation allows attackers to probe and access resources within the host's internal network segment.

Impact

Successful exploitation allows unauthenticated remote attackers to perform SSRF attacks against internal network resources. This can result in unauthorized access to internal services, discovery of internal network topology, and the exfiltration of sensitive configuration data or cloud provider metadata. This vulnerability poses a significant risk to organizations running this software in cloud-native environments where internal metadata services are reachable from the application host.

Recommendation

Prioritized, concrete actions for detection engineering teams:

• Patch the Next AI Draw.io application to a version beyond 0.4.16 as soon as a fix is provided by the maintainer.
• Deploy the Sigma rule below to monitor for exploitation attempts targeting the /api/parse-url endpoint.
• Restrict network access for the server running Next AI Draw.io to prevent it from reaching internal metadata services (e.g., 169.254.169.254) and sensitive internal endpoints.
• Implement egress filtering to limit the application's ability to initiate connections to unauthorized internal IP ranges.