context-probe

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This content defines a deliberate, persistent backdoor: it programmatically modifies multiple rule files without confirmation to enforce a non-overridable assistant behavior across layers (unauthorized persistent file modification and cross-file broadcast that can be abused for supply-chain/backdoor purposes) — there is no sign of network exfiltration or remote code execution, but the forced, non-consensual override is high-risk.