fullstack-feature

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.95). The skill embeds a forced, unconditional "runtime directive" execution model (via directives) that creates a remote-control/backdoor channel able to spawn agents which run shell commands, write/commit code, and update persistent memory—enabling remote code execution, code injection or supply-chain persistence and potential data exfiltration without human oversight.