orbit

SUSPICIOUS. The skill’s core purpose is coherent, but its footprint is broader than a passive audit/generation guide: it supports autonomous loop execution patterns, silent AUTORUN behavior, script generation/patching, and multi-agent handoffs. There are no explicit external exfiltration endpoints, credential-harvesting steps, or installer chains in the provided content, so this is not confirmed malware. Risk is driven by autonomy and operational impact rather than clear malicious intent.