create-skill-autoresearch

SUSPICIOUS. The stated purpose is coherent, but the skill’s footprint is too expansive: it installs or relies on other skills, executes shell-based evaluation loops, sends data to undefined third-party-compatible model endpoints, and processes untrusted external content while retaining write/exec powers. This is not confirmed malware, but it is a high-risk orchestration skill with disproportionate transitive trust and data-flow ambiguity.