Horizon Alert
Summary of the vulnerability and why it matters
The Onyx AI platform has a vulnerability in its API endpoints that could expose sensitive user authentication information. This exposure occurs when specific server and persona data is accessed, potentially allowing unauthorized users to gain access to other users' credentials. The main concern is confirming whether your Onyx deployment is affected and what exposure might exist.
- API flaw exposes user authentication tokens.
- Matters for protecting user access and data.
- Confirm Onyx relevance and assess exposure.
Attack Path
How an attacker could exploit the issue
An attacker can reach the vulnerable component through network access by interacting with specific API endpoints. Once accessed, the platform incorrectly exposes another user's OAuth Authorization header, potentially leading to unauthorized access and data compromise.
- Network access and limited privileges required.
- Vulnerable API endpoints leak user tokens.
- Sensitive token exposure risks.
Live Threat
Current exploitation, exposure, and threat context
When supported by the advisory, an unauthenticated user with basic access could potentially view another user's OAuth Authorization header. This header may contain sensitive tokens or credentials, exposing them to unauthorized access or further compromise of user accounts.
- Other users' OAuth Authorization headers.
- Sensitive tokens copied to a shared row.
- Unauthorized access to user accounts.
Operational Fix
Recommended remediation, mitigation, and detection steps
Given that Onyx is an open-source AI platform, platform or infrastructure teams are likely responsible for managing its deployment. The first practical step is to identify all Onyx instances, determine their reachability and criticality, and then locate the accountable owner for each instance to plan remediation based on risk.
- Platform or infrastructure teams own this issue.
- Verify Onyx instance reachability and criticality.
- Plan remediation with accountable owners.
