command-builder

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

• Third-party content exposure detected (high risk: 0.90). The skill's "Review Pull Request Command" explicitly tells the agent to fetch PR details and diffs via gh pr view {pr-number} (and the frontmatter allows tools like Read/WebFetch), which pulls user-generated content from GitHub/public repos that the agent will read and interpret, exposing it to untrusted third-party input.