ultrathink

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

• Potentially malicious external URL detected (high risk: 0.90). The skill relies on an external Sequential Thinking MCP that is invoked at runtime and explicitly instructs installing/executing remote code via "npx -y @modelcontextprotocol/server-sequential-thinking" (reference: https://github.com/modelcontextprotocol/servers/tree/main/src/sequentialthinking), which would fetch and run external code that directly controls the agent's prompts/thought chain.