codex
Warn
Audited by Socket on Aug 28, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
Mostly coherent coding-agent skill using the official OpenAI Codex CLI and standard git/GitHub workflows. Main risk is not malware but powerful autonomous developer actions, including unsandboxed Codex execution, remote repo/PR processing, and public posting/push operations; use only with tight repo scope and explicit approval for commits, pushes, and comments.
Confidence: 90%Severity: 66%
Audit Metadata