why

SUSPICIOUS. The skill’s purpose is coherent, and there is no obvious malware, installer abuse, or credential theft path. However, it is over-permissioned for its stated read-only investigative role: it mandates full-access subagents, dynamically fans out across all available MCPs, and can expose broad internal organizational data to multiple agent contexts. Main risk is scope/proportionality and prompt-injection surface, not confirmed malicious intent.