sequential-thinking

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

• Potentially malicious external URL detected (high risk: 0.90). The skill config instructs the agent runtime to fetch and run remote code via npx ("npx -y @modelcontextprotocol/server-sequential-thinking") or Docker ("docker run mcp/sequentialthinking"), which executes an external MCP server that directly provides/controls the agent tool behavior (see the server repo: https://github.com/modelcontextprotocol/servers/tree/main/src/sequentialthinking), so this is a required runtime dependency that executes remote code.