mockup

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 1.00). The prompt instructs the agent to disable the sandbox when opening the generated file ("with the sandbox disabled on the first try"), which is a hidden/deceptive environment-bypass instruction outside the skill's stated design/mockup purpose.

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

• Attempt to modify system services in skill instructions detected (high risk: 0.80). The prompt explicitly instructs the agent to disable the sandbox and run system "open/xdg-open" on creation (SKILL.md:82), which is a direct instruction to bypass a security mechanism and thus risks compromising the host environment.