Description

Yuxi is a large-model-based intelligent knowledge base and knowledge graph agent development platform. Prior to version 0.6.2, the project's authentication mechanism contains a flaw. In affected versions, the system does not sufficiently validate the identity token in the Authorization header — only performing a validity check. This allows an administrator token generated in another deployment instance or local testing environment to be used to access the backend management interfaces of a different affected instance. An attacker who obtains or constructs an acceptable administrator Authorization token may bypass normal login authentication and gain administrator privileges. This vulnerability could allow an attacker to access system configurations, invoke backend management APIs, create administrator accounts, and ultimately take over the system backend. This issue has been fixed in version 0.6.2. Before upgrading, users are advised to implement the following temporary measures: Set the environment variable `JWT_SECRET_KEY` to a non-default value, and configure a unique, sufficiently strong JWT/authentication key for each deployment instance; and/or avoid exposing backend management interfaces directly to the public network.

Executive Summary

Technical Details

Mitigations

References

Additional Metadata

{
  "score": "9.4",
  "pubDate": "2026-08-12T15:17:38.307Z",
  "pubdate": "2026-08-12T15:17:38.307Z",
  "executiveSummary": "Yuxi, a large-model-based intelligent knowledge base and knowledge graph agent development platform prior to version 0.6.2, suffers from an authentication mechanism vulnerability. The flaw stems from insufficient validation of the identity token within the Authorization header, where the system performs only a basic validity check without verifying the issuing context.\nThis architectural oversight allows an administrative token generated in an entirely separate deployment instance or a local testing environment to be successfully accepted across different vulnerable target instances.\nAn external attacker who obtains or constructs a valid administrator Authorization token can completely bypass standard login authentication controls and acquire elevated administrator privileges.\nThe risk implications are severe, as successful exploitation enables malicious actors to access sensitive system configurations, invoke restricted backend management APIs, create arbitrary administrator accounts, and execute a full system backend takeover.\nExploitation requires the acquisition or generation of an acceptable administrator token combined with network access to the backend management interfaces of the targeted Yuxi deployment.",
  "technicalDetails": "The vulnerability resides in the authentication and session validation mechanisms of Yuxi prior to version 0.6.2, specifically within the handling of JSON Web Tokens (JWT) or identity tokens supplied via the HTTP Authorization header.\nThe root cause of the flaw is that the application validates tokens based solely on cryptographic signature validity or structural integrity without adequately tying the token to a specific deployment instance, environment, or unique secret key configuration.\nBecause deployments frequently rely on default cryptographic keys or shared secrets across independent installations, an administrator token generated in a local testing environment or a separate staging/production instance remains cryptographically valid when presented to a different affected Yuxi instance.\nThe attack flow proceeds as follows: First, the attacker obtains an administrator-level token, which may be acquired through a separate compromised deployment, local development testing, or default key derivation. Second, the attacker issues HTTP requests targeting the backend management interfaces of the victim Yuxi instance, injecting the harvested token into the Authorization header. Third, the vulnerable backend processes the header, executes a superficial validity check, and accepts the token as authentic due to the lack of instance-specific validation or unique key enforcement. Finally, the application grants full administrative privileges to the session.\nThe vulnerable component handles backend authentication checks and API request authorization. Affected versions include all Yuxi deployments prior to version 0.6.2.\nRegarding prerequisites, exploitation requires network exposure to backend management interfaces and possession of an acceptable administrator token. No prior low-privilege account is strictly necessary on the target instance if an external administrator token is supplied.\nPost-exploitation capabilities are critical, granting the attacker full control to query system configurations, execute privileged backend management APIs, provision new unauthorized administrator accounts, and achieve complete backend system takeover."
}