/

CVE-2025-66001

Severity:

HIGH RISK

(8.8)

Improper Certificate Validation in NeuVector OpenID Connect

CVE-2025-66001

Deferred

Vulnerability type

Category:

Improper Certificate Validation

CWE:

CWE-295

Impact:

Authentication Compromise via Man-in-the-Middle

Severity:

HIGH RISK

(8.8)

CVSS version:

3.1

Catch this CVE while code is being written

Detect vulnerable code directly in pull requests, so CVEs are fixed before merge.

Summary

NeuVector's OpenID Connect (OIDC) login flow does not enforce TLS certificate verification by default when communicating with the identity provider.

In environments where this default is left unchanged, an attacker positioned on the network path could perform a man-in-the-middle (MITM) attack against OIDC traffic, potentially intercepting or manipulating authentication data.

Affected context

This vulnerability affects NeuVector deployments configured to use OpenID Connect for authentication where TLS verification of the OIDC provider is not explicitly enforced.

It is most relevant in environments where:

  • NeuVector communicates with an external or remote identity provider over TLS
  • Network paths between NeuVector and the OIDC provider could be monitored or controlled by an attacker (e.g., shared infrastructure, misconfigured networks, or hostile network segments)
  • The default TLS verification behavior has not been hardened or overridden by deployment-specific configuration

Exploit preconditions

To exploit this issue, the following conditions typically apply:

  • Attack surface: Network path between NeuVector and the configured OIDC identity provider
  • Required access: Ability to intercept, route, or proxy traffic (e.g., control over a gateway, compromised router/switch, DNS spoofing, or ARP poisoning in the same network segment)
  • Trigger point: NeuVector initiating an OIDC login authentication flow and contacting the identity provider over TLS without enforcing strict certificate verification
  • Why exploitation is possible: Because TLS certificate validation is not enforced by default, NeuVector may accept a spoofed or untrusted TLS endpoint as the OIDC provider, enabling an attacker to impersonate the identity provider or tamper with the OIDC protocol messages

Why this happens

The underlying issue maps to CWE-295: Improper Certificate Validation.

TLS is designed to provide endpoint authentication and integrity guarantees based on certificate chains and hostname verification.

In this case, NeuVector's OIDC integration does not enforce TLS verification by default when connecting to the OIDC provider.

This means the client connection either skips SSL/TLS certificate checks or relaxes them sufficiently that any presented certificate (including self-signed or attacker-controlled certificates) may be accepted.

As a result, the security assumption that "OIDC traffic goes only to the genuine identity provider" no longer holds, and an attacker who can position themselves on the network path can impersonate that provider or decrypt and modify traffic.

Potential impact

  • An attacker may intercept OIDC authentication traffic and observe sensitive authentication data and tokens.
  • An attacker could impersonate the legitimate OIDC identity provider and trick NeuVector into trusting forged identity assertions.
  • User sessions and access tokens could be issued under attacker-controlled identities, leading to unauthorized access within NeuVector.
  • Integrity of authentication decisions can be compromised, allowing privilege escalation or lateral movement within the protected environment.
  • Audit trails based on identity information may become unreliable due to forged or manipulated identity data.

General mitigation guidance

Effective mitigation focuses on enforcing strict TLS verification and hardening trust boundaries around identity providers:

  • Ensure that TLS certificate verification is explicitly enabled and enforced for all communications with the OIDC provider, including certificate chain validation and hostname checking.
  • Configure NeuVector and its environment to trust only a well-defined set of Certificate Authorities (CAs) that issue certificates for the identity provider, reducing exposure to rogue or misused CAs.
  • Pin expected certificates or public keys for the OIDC provider where operationally feasible, to make impersonation significantly harder.
  • Deploy NeuVector and the OIDC provider on network segments designed to minimize exposure to untrusted intermediaries, using secure routing and isolation.
  • Apply the principle of secure-by-default in configuration management: treat relaxed TLS options (e.g., "skip verification", "insecure", or "trust all certificates") as temporary exceptions that require explicit justification and monitoring.

How teams detect issues like this

Organizations can detect and manage this class of issue using a combination of configuration review and automated checks:

  • Perform configuration audits of NeuVector and related services to identify any TLS settings that disable or weaken certificate verification for OIDC endpoints.
  • Use static configuration scanning or policy-as-code tools (e.g., for Kubernetes manifests, Helm charts, or infrastructure-as-code) to flag patterns such as insecureSkipVerify, "trust all", or unset verification flags for identity provider URLs.
  • Incorporate security-focused code and configuration reviews for authentication flows, explicitly checking that TLS is validated for all identity and token endpoints.
  • Monitor network security tools or TLS inspection logs for connections that negotiate TLS without proper certificate validation or that accept untrusted/self-signed certificates in production paths.
  • Integrate automated security testing in CI/CD to enforce organizational policies that forbid disabling TLS verification for external identity providers.

Sources

SourceLink
bugzilla.suse.combugzilla.suse.com/show_bug.cgi?id=CVE-2025-66001
github.comgithub.com/neuvector/neuvector/security/advisories/GHSA-4jj9-cgqc-x9h5

Frequently asked

questions

Yes, in scenarios where an attacker can control or influence the network path between NeuVector and the OpenID Connect provider. This often requires network-level access (e.g., same LAN, compromised router, or malicious gateway), but does not necessarily require local access to the NeuVector host itself.

The attacker generally needs the ability to perform man-in-the-middle attacks on the TLS connection, such as DNS spoofing, ARP poisoning, or control over a proxy or routing device. They do not need valid user credentials if they can successfully impersonate the OIDC provider or tamper with authentication traffic.

The most robust fix pattern is to enforce strict TLS verification for all OIDC-related connections: validate certificates against trusted CAs, verify hostnames, and avoid permissive or "insecure" TLS options. Where possible, combining this with certificate or key pinning adds an additional layer of assurance.

Adopting a security posture where TLS verification is secure-by-default for all external service integrations is key. This includes standardized configurations for TLS, automated policy checks that block disabling certificate validation, and documentation that clearly flags any deviation from full TLS verification as a high-risk exception requiring compensating controls and explicit approval.

Find out what's already
exploitable in your codebase.