review-deep

SUSPICIOUS. The skill’s core behavior matches its stated purpose, but its footprint is high-risk: it automates large-scale outbound code review to multiple external AI providers using local credentials/config, runs many unattended subprocesses, and acknowledges a Serena path-isolation weakness. No clear credential theft, covert exfiltration endpoint, or malicious mismatch is present, so this is not confirmed malware; the main concerns are broad autonomous execution, sensitive code disclosure to third-party models, and tool-chain trust inherited from opencode/Serena.