github

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

• Third-party content exposure detected (high risk: 0.80). SKILL.md explicitly instructs the agent to fetch and inspect content from GitHub (e.g., using "gh pr", "gh issue", "gh run", and "gh api" to retrieve PRs, issues, and logs) which are public, user-generated sources that the agent is expected to read/interpret and that could contain instructions influencing later actions.