codex-session-patcher
Warn
Audited by Socket on Mar 29, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The skill’s footprint is coherent with its stated purpose, but that purpose is to weaken or bypass AI safety behavior across multiple tools. It persistently alters local session/config files and can forward session content plus API credentials to configurable external LLM endpoints, creating meaningful offensive-use and data-exposure risk.
Confidence: 95%Severity: 83%
Audit Metadata