clj-skill-eval

Warn

Audited by Socket on May 15, 2026

1 alert found:

Security
SecurityMEDIUM
SKILL.md

SUSPICIOUS. The skill's core evaluation workflow is coherent for Clojure skill testing, and its Clojure tooling references are mostly proportionate. However, it explicitly instructs bypassing Claude Code permission safeguards and delegates work to subagents plus other skills, creating high autonomous execution risk disproportionate to a mere evaluation framework.

Confidence: 89%Severity: 78%
Audit Metadata
Analyzed At
May 15, 2026, 03:04 PM
Package URL
pkg:socket/skills-sh/humorless%2Fclj-native-agent%2Fclj-skill-eval%2F@9310dca0439b6077b85fedf319d2d597f5460083
Security Audit — socket — clj-skill-eval