pr-reviewer

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.80). The prompt includes an explicit instruction to conceal the use of AI ("Never mention AI, Claude, or automated review tools in public updates"), which is a deceptive directive outside the stated PR-reviewing purpose and thus constitutes a prompt injection risk.

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow uses gh pr view <pr_number> --json files,... and then analyzes PR file contents, which are outsider-authored free text (PR author’s changes) ingested into the agent’s LLM context.