pr-review
Warn
Audited by Socket on Apr 27, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS: the core PR-review behavior is legitimate and aligned with its purpose, and the official gh/glab tooling is coherent. The main risk comes from transitive skill dependencies plus processing untrusted PR content while retaining write/action capabilities, which creates meaningful prompt-injection and data-flow uncertainty. No clear credential harvesting or confirmed malicious behavior is present.
Confidence: 85%Severity: 64%
Audit Metadata