address-pr-feedback

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly fetches and ingests user-generated GitHub PR comments and reviews via gh api endpoints (e.g., gh api repos/{owner}/{repo}/pulls/{number}/comments, issues/{number}/comments, and gh pr view ...) and then parses and acts on that content to categorize feedback and drive fixes, commits, and PR comments, so untrusted third-party content can materially influence agent actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill calls GitHub API endpoints at runtime (e.g., gh api repos/{owner}/{repo}/pulls/{number}/comments and gh pr view {number} or the PR URL https://github.com/owner/repo/pull/42) to fetch reviewer comment bodies which are injected into the agent’s workflow and directly control what the agent does, making this a required remote input that controls prompts.