recipe-pr-review

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

• Third-party content exposure detected (high risk: 0.90). Yes—scripts/collect-pr-context.py uses gh pr view and gh_api_paginated to fetch outsider-authored PR body, diff, prior comments/reviews, then scripts/get-review-material.py reads them from context_dir and scripts/run-review.py passes them (via MATERIAL_SCRIPT instructions) into the reviewer LLM prompt context.