agentiko-hermes

SUSPICIOUS: the skill is internally consistent as a Hermes operations guide, but its actual footprint is a high-risk offensive automation playbook. Same-org Hermes provenance reduces supply-chain concern, yet the skill materially increases risk by enabling autonomous scans, disabling approval gates, and guiding red-team activity through an SSH worker and scheduled jobs. Not confirmed malware, but dangerous by design for an AI agent.