/

CVE-2026-90509

Severity:

HIGH RISK

(7.3)

Hard-Coded Credentials (CWE-259) in Orion Visor API

CVE-2026-90509

Deferred

Vulnerability type

Category:

Hard-Coded Credentials

CWE:

CWE-259, CWE-798

Impact:

Unauthorized Access and Credential Compromise

Severity:

HIGH RISK

(7.3)

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

A high-severity weakness in dromara Orion Visor through version 2.5.7 involves hard-coded credentials in the ExposeApiAspect.beforeExposeApi function. Because the affected logic may be reachable remotely, an attacker could use exposed credentials to access protected functionality or services when those credentials remain valid and insufficiently restricted.

Affected context

The issue occurs in Orion Visor deployments using the affected API exposure aspect and its credential-handling logic. Risk depends on whether the embedded credentials are active, what systems or endpoints they protect, and whether the affected function can be reached through a remotely accessible application path.

Exploit preconditions

  • Attack surface: A remotely accessible Orion Visor application or API path invoking ExposeApiAspect.beforeExposeApi
  • Required access: Network access to the exposed service; valid application authentication may or may not be required depending on deployment controls
  • Trigger point: Processing of an API exposure request through the affected aspect
  • Why exploitation is possible: Credentials embedded in application logic may be discoverable or reusable, allowing access beyond the intended request context when they are trusted by downstream services

Why this happens

The underlying bug is the storage or use of authentication material directly within application code. This creates a fixed trust secret that can be extracted from source, compiled artifacts, logs, or runtime behavior and may be shared across deployments. The issue maps primarily to CWE-259: Use of Hard-coded Password, with CWE-798: Use of Hard-coded Credentials also describing the broader insecure pattern.

Potential impact

  • An attacker may gain unauthorized access to protected APIs or backend services.
  • Exposed credentials could allow impersonation of an application or service account.
  • Attackers could access, alter, or remove data permitted to the compromised identity.
  • Compromised credentials may enable lateral movement into connected systems.
  • Credential reuse across environments could increase the scope of a compromise.

General mitigation guidance

Mitigation generally focuses on removing authentication material from source code and reducing the trust placed in static secrets.

Common defensive patterns include:

  • Obtaining credentials from a managed secret store or protected runtime configuration.
  • Using short-lived, scoped, and rotatable credentials rather than shared static values.
  • Applying least privilege to service identities and restricting access by environment and function.
  • Preventing secrets from appearing in source repositories, build artifacts, logs, and diagnostic output.
  • Rotating any potentially exposed credentials and reviewing dependent systems for unauthorized use.
  • Enforcing secure-by-default deployment policies that fail safely when required secrets are absent.

How teams detect issues like this

Organizations can identify hard-coded credential risks through:

  • Static analysis rules for passwords, tokens, API keys, and credential-like literals in source code.
  • Secret-scanning checks across repositories, build outputs, container images, and configuration files.
  • Code review signals such as authentication values passed directly from constants or embedded in aspect and middleware logic.
  • Runtime monitoring for unusual use of service identities, unexpected API access, and authentication from unfamiliar locations.
  • CI and repository policies that block commits containing probable secrets and require approved secret-management integrations.

Automated code review and secret-detection tools can help surface these patterns, but findings should be validated to distinguish test data from active credentials.

Sources

SourceLink
github.comgithub.com/dromara/orion-visor/
github.comgithub.com/dromara/orion-visor/issues/170
github.comgithub.com/sumo166/CVE-apply/blob/main/dromara-orion-visor/ExposeApi%20Hardcoded%20Default%20Token%20Authentication%20Bypass%20(CWE-798)_en.md
vuldb.comvuldb.com/cve/CVE-2026-90509
vuldb.comvuldb.com/submit/911864
vuldb.comvuldb.com/vuln/403097
vuldb.comvuldb.com/vuln/403097/cti

Frequently asked

questions

The reported weakness is described as remotely exploitable, but practical exploitation depends on network exposure, the execution path reaching the affected function, and whether the embedded credentials remain valid.

Not necessarily. The required access depends on Orion Visor's deployment and API controls. Network reachability may be sufficient if the affected endpoint is exposed and does not require prior authentication.

Remove credentials from application code, replace them if exposure is possible, and retrieve narrowly scoped secrets through a protected secret-management mechanism with rotation support.

Combine secret scanning, secure code review, least-privilege service accounts, short-lived credentials, repository protection, and CI policies that prevent credential literals from entering production code.

Find out what's already
exploitable in your codebase.