codex-router-external-models
Warn
Audited by Socket on Sep 16, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the skill’s core purpose is coherent, but its trust model is weak. It asks users to run remote installer scripts from a personal GitHub repo, centralizes multiple provider credentials in a local proxy, and supports custom base-URL overrides that could redirect credentials; these risks are significant even though the documented functionality appears legitimate.
Confidence: 85%Severity: 72%
Audit Metadata