global-governance-institutional-quality-data

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly instructs agents to call the AgentPMT MCP to fetch live schema/instructions at runtime (e.g., via agentpmt-tool-search-and-execution "get_schema"/"get_instructions" against https://api.agentpmt.com/mcp/), and those returned "live schema and instructions" are treated as authoritative guidance that directly control how the agent constructs prompts/behavior.