enhance-skills

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

• Third-party content exposure detected (high risk: 0.80). This SKILL.md's "Dynamic Context Injection" section explicitly shows injecting outputs from commands like !gh pr diff and !gh pr view --comments, which pull user-generated GitHub PR diffs/comments (untrusted third‑party content) into the agent's context and could materially influence its subsequent actions.