map-generator-with-markers

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

• Potentially malicious external URL detected (high risk: 0.90). This skill explicitly directs runtime calls to the AgentPMT MCP server (e.g., https://api.agentpmt.com/mcp/) to fetch live schema and get_instructions, and it instructs agents to treat the returned instructions as authoritative, meaning remote content fetched at runtime can directly control agent prompts.