autoresearch

SUSPICIOUS: the skill’s main behavior is coherent with prompt optimization, but its footprint is high-risk because it grants an agent an open-ended autonomous loop that repeatedly executes and rewrites arbitrary other skills without pausing for approval. There is no strong evidence of malware or credential theft in the provided content, and the only external fetch is an official Chart.js CDN for the dashboard, but the autonomy and transitive execution surface make this a high-risk skill.