address-review-bots

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

• Third-party content exposure detected (high risk: 0.85). This skill runs scripts/review-bot-snapshot.mjs --wait --since ... --json, which fetches GitHub PR review/inline/issue comment bodies (outsider-authored bot text) and renders them into .context/review-bot-snapshot.json/markdown, which the agent then incorporates into the LLM context for classification and fixes.