codex-session-patcher
Warn
Audited by Socket on May 16, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the skill is internally coherent, but its purpose is to bypass model refusals and inject behavior-changing prompts into AI coding tools. The install path is only moderately risky, yet the main concern is deliberate tampering with session history and optional forwarding of session data plus API credentials to a configurable remote endpoint.
Confidence: 86%Severity: 74%
Audit Metadata