fix-coderabbit-review

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

• Third-party content exposure detected (high risk: 0.85). Outsider-authored free text is ingested into the agent/LLM context via the exporter’s runtime fetch of GitHub PR review comment bodies (REST pulls.listReviewComments and GraphQL reviewThreads.comments.body), which are then written into ai-docs/reviews-pr-<PR_NUMBER>/issues/*.md and _summary.md and subsequently used by the agent for remediation.