codex-autoresearch-loop
Warn
Audited by Socket on Mar 21, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the skill’s core purpose is autonomous code modification, but its footprint is unusually broad and high-risk: unbounded unattended execution, automatic git actions, external web-search escalation, and transitive remote skill installation. The ara.so branding versus third-party GitHub install source further weakens trust, so this should not be treated as a benign documentation-only skill.
Confidence: 90%Severity: 86%
Audit Metadata