code-review:review-pr
Warn
Audited by Socket on Mar 22, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS: The skill’s core purpose is coherent, and its GitHub data flow is to official endpoints, not a proxy. The main risk is not malware but autonomous action: it can post review comments and modify PR descriptions after processing untrusted PR content with parallel agents, creating medium-high operational risk for an AI agent.
Confidence: 90%Severity: 61%
Audit Metadata