Tensei through 0.11.5 contains a privilege escalation vulnerability in @tensei/auth that allows unauthenticated or authenticated users to assign themselves arbitrary roles via the exposed User Roles field. Attackers can supply a roles array during registration, a POST to the users route, or a PATCH to their own record to gain administrator permissions.
Casky was already ahead
This CVE exploits attack patterns that Casky's 0matched skills already investigate — long before this vulnerability was disclosed. Claude's reasoning model maps these techniques to MITRE ATT&CK, so practitioners who ran these skills have already seen the threat behaviour in their findings.
CVE-2026-129550 is a critical privilege escalation vulnerability in Tensei's @tensei/auth component that enables both unauthenticated and authenticated users to elevate their privileges to administrator level. The flaw stems from improper input validation on the User Roles field, which remains exposed during user registration, POST requests to user creation endpoints, and PATCH operations on user records. Any attacker can inject a roles array containing administrative permissions, completely bypassing authorization controls. This affects all Tensei deployments through version 0.11.5 and represents a severe risk to organizations using this framework, as it allows immediate lateral movement and system compromise with minimal attack effort.
While CVE-2026-129550 does not map directly to specific MITRE ATT&CK techniques, Casky's Claude-powered analysis engine would detect the attack patterns associated with this vulnerability across related threat behaviors. A practitioner using Casky would identify findings aligned with T1548 (Abuse Elevation Control Mechanism) and T1078 (Valid Accounts), as the attack abuses trust in the authentication system to gain elevated access. The extended reasoning capability in Claude would correlate suspicious patterns such as: role field manipulation in API requests, mismatched user privilege levels versus account creation timestamps, and configuration changes occurring immediately after user registration. Though zero matching skills exist in Casky's current 754-skill library for this specific CVE, practitioners would see recommendations to monitor authentication logs for unauthorized role assignments, implement strict input validation on user-supplied role parameters, and enforce server-side authorization checks independent of client submissions.
Composite risk scoring from EPSS, CISA KEV, Shodan, and GreyNoise — 21 security APIs correlated into a single Casky Risk Score. Coming in Casky Pro. Join early access →
Casky has 0 skills that investigate the attack patterns behind CVE-2026-129550. Run one and get CVSS-scored findings in 3 minutes.
Run the skill that detects this →© 2026 Casky.AI, Inc. · AI Security Investigation