activity-push

SUSPICIOUS. The skill’s core workflow is coherent for activity extraction and WeCom pushing, and its WeCom endpoints match the stated purpose. The main risks are a user-configurable MP_API_HOST that can receive feed data and optional API keys, raw env-file secret handling, unverifiable local helper-script provenance, and autonomous outbound posting to internal groups. This looks more like a high-risk automation skill than confirmed malware.