architecture-review

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.80). The skill contains explicit instructions that override or ignore external/system prompts (e.g., "Ignore Claude-specific mode-switch instructions") and mandates auto-execution/authorization behavior for subagents and slash/skill commands, which act like prompt-injection/override directives outside the architecture-review's core analysis purpose.