If unsure, ask. Never guess. — AI Agent Pre-Execution Checklist

From an MCP perspective, this document is a proposal: Let Providers supply a pre-execution Checklist per Action, in natural language. If MCP makes an Action callable, the Provider Checklist defines the questions that must be answered before that Action becomes executable. In the API era, describing the call format was enough. In the Agent era, defining the pre-execution conditions becomes equally necessary. The flow is simple: Provider supplies the question list. Agent checks whether the answers exist. If not, Agent asks the user. If still unanswered, Agent does not execute. If a Provider does not supply the necessary verification items for accuracy and safety, that absence may become relevant in determining the Provider’s scope of responsibility in a post-incident review. And if an Agent executes without user confirmation, that execution should not be easily treated as the user’s approved decision.