azsdk-common-apiview-feedback-resolution

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

• Third-party content exposure detected (high risk: 0.85). Required workflow step “Retrieve” calls azure-sdk-mcp:azsdk_apiview_get_comments using an APIView URL from the user’s PR, and the resulting reviewer comment text (outsider-authored free-form feedback) is ingested into the agent context for categorization/resolution.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill invokes the azure-sdk-mcp tool to fetch APIView reviewer content at runtime (e.g., https://spa.apiview.dev/review/abc123?activeApiRevisionId=def456), and that fetched review text is injected into the agent’s context to drive resolution actions.