brand-strategy

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.90). The "Companion check" explicitly instructs the agent to run local shell commands and offer installs (accessing ~/.claude/skills, etc.), which is unrelated to brand-strategy and thus is a sneaked-in instruction that attempts to change agent behavior and access the environment outside the skill's stated 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 companion check explicitly instructs the agent to run shell commands (ls) to inspect user directories and to offer installing companion skills via npx, which would modify the machine and can execute remote code, so it pushes the agent to change the host environment.