gh-address-comments

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

• Third-party content exposure detected (medium risk: 0.30). scripts/fetch_comments.py calls GitHub (via gh pr view and gh api graphql) to read PR conversation comments, review bodies, and inline review thread comment body fields, which are outsider-authored free text.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The skill explicitly instructs the agent to run commands with "elevated network access" and to re-run gh auth status with sandbox_permissions=require_escalated, i.e. to bypass sandboxing and request escalated privileges, which is a direct instruction to compromise the agent's execution environment.