resolve-pr-comments

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

• Third-party content exposure detected (medium risk: 0.30). At runtime the required workflow uses GitHub CLI to fetch unresolved PR review threads and issue/PR conversation comments (GraphQL reviewThreads(...){ nodes { ... comments{ ... body ... }}} and REST issues/{PR_NUMBER}/comments) where the body text is authored by external users/bots and ingested by the LLM for triage and fix planning.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill issues runtime gh api calls (e.g., "repos/{OWNER}/{REPO_NAME}/issues/{PR_NUMBER}/comments?per_page=100") to fetch PR review comments and threads which are then injected into generated fix plans and subagent prompts, so external content directly controls agent instructions.