review-pr

Warn

Audited by Socket on Jul 15, 2026

1 alert found:

Security
SecurityMEDIUM
SKILL.md

SUSPICIOUS: the skill’s purpose is coherent, but it combines untrusted PR ingestion, local checkout/testing, third-party AI forwarding, and repository approval/merge actions. The main concerns are indirect prompt injection and autonomous consequential actions rather than overt malware.

Confidence: 89%Severity: 76%
Audit Metadata
Analyzed At
Jul 15, 2026, 10:40 PM
Package URL
pkg:socket/skills-sh/kube-hetzner%2Fterraform-hcloud-kube-hetzner%2Freview-pr%2F@460988ef7408ff9d5114ad8afc8f54826d5a0f92e088cf41a14c63e8b09b736b
Security Audit — socket — review-pr