iterate-pr

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly queries GitHub reviewThreads via gh api graphql and triggers the /fixing-pr-comments sub-skill (with instructions to "triage, fix, commit, push, reply, resolve"), so it ingests and acts on user-generated PR review comments from third-party sources which could contain instructions that influence behavior.