codex-session-patcher
Warn
Audited by Socket on Sep 16, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS: the skill is internally coherent, but its stated purpose is to bypass AI refusals and inject persistent CTF prompts, which is a high-risk misuse-oriented workflow. No clear credential theft or malware is shown, yet provenance is mixed and the optional AI cleaning feature can transmit session content to an external API.
Confidence: 83%Severity: 62%
Audit Metadata