modify-agent-prompts

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

• Third-party content exposure detected (high risk: 0.90). The skill's required workflow and tool list (SKILL.md and references/TOOL-CREATION.md) explicitly include MCP tools and an mcpConfig example that runs a web-search server (e.g., @modelcontextprotocol/server-web-search) plus a knowledge_base_query tool, indicating the agent will fetch and read arbitrary web/third-party content which can influence its decisions and tool usage.