gh

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

• Third-party content exposure detected (medium risk: 0.30). In SKILL.md’s described required workflow for the GitHub CLI, the agent can be made to read issue/PR/discussion content (including comment bodies and discussion bodies) from GitHub at runtime via commands like gh discussion view/list with --comments and gh api .../pulls/{n}/comments, where outsider-authored text from users is ingested through the selected item.