twitter-automation

SUSPICIOUS. The skill’s purpose matches social-media automation, but its footprint is high risk: it installs a remote same-org CLI via pipe-to-shell, routes X credentials and actions through inference.sh infrastructure instead of direct official API usage, enables autonomous posting/DM/follow actions, and encourages transitive skill installation. This is not confirmed malware, but it is a high-risk agent skill with third-party credential/action mediation.