review-driven-development
Warn
Audited by Socket on Mar 18, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS: The skill is mostly coherent with its stated issue-to-merge purpose and uses official GitHub/GitLab tooling, so there is no strong malware or installer abuse signal. The main risk is proportionality and data flow: it can autonomously modify/push code and may send full diffs to external AI reviewers, which is plausible for review automation but increases confidentiality and prompt-injection risk.
Confidence: 86%Severity: 56%
Audit Metadata