snap-grill-me

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 1.00). The skill's SKILL.md explicitly instructs the agent to fetch GitHub issues (and referenced parent PRDs/epics) with gh issue view, which are untrusted, user-generated third-party pages that the agent will read and use to drive its questioning and decisions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly directs the agent at runtime to fetch GitHub issues (e.g., https://github.com/.../issues/...) using "gh issue view", which would inject remote issue/PRD text into the agent's context and thus directly control prompts.