debug-like-expert

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly requires using web search, MCP/Context7 servers, GitHub issues, Stack Overflow and other public documentation as part of its "C. Gather External Knowledge" and "when-to-research.md" workflow, so the agent will fetch and read untrusted public third-party content that can materially influence its debugging decisions.