twitter-check

SUSPICIOUS. The skill’s core behavior matches its stated purpose, but it grants an agent the ability to post publicly on the user’s behalf and relies on a generically named MCP server whose provenance is not fully identified in the skill. Data flows appear coherent with X/Twitter usage and there is no direct evidence of credential theft or malware, but the combination of credential forwarding to external tool code and autonomous posting makes this a medium/high operational risk skill.