kiwi-go-agent

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs agents to fetch and ingest definitions/results from external services (e.g., NewMCPProxyTool calling client.ListTools in references/llm-best-practices.md "Wrapping External MCP Servers" and SearchTool/FileSearchTool using external search/vector clients in references/tools-implementation.md), which are untrusted third-party sources whose tool definitions and results are read and used by the agent and can materially change tool use and decisions.