zero-day

SUSPICIOUS: the skill is internally consistent, but it equips an AI agent with high-risk offensive security capabilities, including exploit research and active target probing. There is little evidence of credential theft or covert exfiltration, so this is not confirmed malware, but the overall security risk is high due to the agentic exploit-development and testing workflow.