optimize-agentic-mcp

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

• Third-party content exposure detected (high risk: 0.80). The skill's "Dispatching Subagents" workflow explicitly instructs launching an "internet-researcher" for current spec/vendor pricing/platform quirks/CVEs, which implies fetching and ingesting untrusted public web content (third‑party sites, forums, CVE pages) that can materially influence architecture and tool-selection decisions.