lev-workshop

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.90). The code snippet sets ClaudeAgentOptions(permission_mode="bypassPermissions"), which is an explicit instruction to override agent permission controls and thus constitutes a hidden/deceptive directive outside the workshop's stated lifecycle purpose.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.70). Although most actions target user/project files, the prompt explicitly sets ClaudeAgentOptions(permission_mode="bypassPermissions"), which instructs the agent to bypass permission/security controls and therefore pushes toward compromising machine security.