occ-use
Warn
Audited by Socket on Oct 12, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS: the skill's purpose is coherent, and the flagged autonomy-abuse issue is likely a false positive, but it asks the agent to execute a third-party reimplementation of Claude Code via npx and pass provider credentials to it. This is not clearly malicious, yet the install-trust and credential-forwarding footprint is higher risk than an official vendor CLI.
Confidence: 86%Severity: 58%
Audit Metadata