A team builds an MCP tool that manages Kubernetes clusters. The tool includes operations like…

Question

A team builds an MCP tool that manages Kubernetes clusters. The tool includes operations like scale_deployment, delete_pod, and drain_node. These are high-impact operations that should require confirmation. How should the tool design handle dangerous operations?

Accepted Answer

D. Implement a two-phase execution: the tool first returns a preview of what will happen (dry run), and requires a second call with a confirmation token to execute

Answer

A. Add 'DANGEROUS:' prefix to tool descriptions and rely on Claude to warn the user

Answer

B. Separate dangerous tools into a different MCP server that requires admin credentials

Answer

C. Require the user to type 'CONFIRM' before each dangerous operation

Question 262