ai-agent-redteam

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This content is an explicit offensive red-team toolkit (not defensive-only code) containing deliberate, high-risk attack patterns — e.g., data exfiltration via markdown/image/OAST callbacks and base64 smuggling, SSRF to cloud metadata for credential theft, MCP tool-poisoning and rug‑pulls that can enable RCE/backdoors, persistent memory poisoning backdoors (MINJA/Chroma), and multiple obfuscation techniques — indicating clear intent and capability for abuse.