okx-buildx-hackathon-agent-track

SUSPICIOUS: the skill’s hackathon purpose broadly matches its Moltbook/OKX actions, but its footprint is high-risk for an AI agent because it combines remote install/update chains, transitive skill installation, untrusted content ingestion, and autonomous public actions (posting, deleting, commenting, voting). Data flows mostly target official OKX/Moltbook endpoints, so this is not confirmed credential theft, but the operational scope is broader and riskier than a normal documentation skill.