Description

IBM API Connect 12.1.0.0 through 12.1.1.2 could allow a remote authenticated attacker to obtain sensitive information due to server-side request forgery. When service definitions are imported, references to external resources are not sufficiently restricted, so the server could be made to request resources it should not access.

Executive Summary

Technical Details

Mitigations

References

Additional Metadata

{
  "score": "8.5",
  "pubDate": "2026-11-07T05:18:29.640Z",
  "pubdate": "2026-11-07T05:18:29.640Z",
  "executiveSummary": "This vulnerability is identified as a Server-Side Request Forgery (SSRF) flaw residing within IBM API Connect.\nThe vulnerability affects versions 12.1.0.0 through 12.1.1.2 of the product.\nThe issue arises from insufficient validation of external resource references during the import process of service definitions.\nA remote authenticated attacker can leverage this flaw to coerce the server into performing unauthorized requests to internal or external network resources.\nThe impact includes the potential exposure of sensitive information, such as metadata, internal service data, or credentials, which could facilitate further unauthorized access or reconnaissance within the host network.\nSuccessful exploitation requires the attacker to have authenticated access to the API Connect management interface or the specific component responsible for service definition imports.\nThis represents a significant security risk, as it circumvents network boundary protections by using the API Connect instance as a proxy to probe internal infrastructure.",
  "technicalDetails": "The root cause of this SSRF vulnerability is the inadequate sanitization and validation of Uniform Resource Locators (URLs) and resource pointers parsed during the ingestion of service definitions, such as OpenAPI or WSDL files.\nWhen a user imports a service definition, the IBM API Connect server attempts to resolve and fetch referenced external resources to populate the API configuration.\nBecause the server does not enforce a strict allow-list or perform sufficient validation on the target URI schemes or destination hosts, an attacker can manipulate these references to point to arbitrary endpoints.\nThe attack flow begins with an authenticated attacker submitting a malicious service definition file containing embedded URIs that target internal network services, private IP addresses (e.g., 127.0.0.1, 169.254.169.254), or unauthorized external resources.\nUpon processing the file, the server's backend logic initiates an outbound HTTP/HTTPS request to the specified, unauthorized address.\nThe server effectively acts as a proxy, sending the request from its privileged position within the internal network. The response from the target resource is then processed or reflected by the API Connect system.\nIf the target resource returns sensitive data—such as internal server metadata, configuration files, or authentication tokens—the attacker can potentially gain visibility into these details via the error messages or the resulting API configuration state.\nThe vulnerability is exposed through the administrative or developer interface utilized for API management and lifecycle operations. The primary component at risk is the resource resolution/import engine within the IBM API Connect stack.\nAuthentication is a prerequisite for exploitation, meaning the attacker must already possess valid credentials to access the relevant interface. However, once authenticated, no additional privilege escalation is typically required to trigger the SSRF.\nPost-exploitation impact extends beyond mere information disclosure; an attacker could use this vulnerability to perform internal port scanning, interact with unauthenticated internal APIs, or bypass firewalls that would otherwise restrict direct access from the external environment."
}