Anthropic added the /goal command to Claude Code in v2.1.139. You set a completion condition; the agent keeps working across turns; a second model reads the transcript and decides whether the condition was met. It is the built-in version of the keep-going loops people have been hand-rolling for
A short post argued this week that reliable agents need deterministic control flow, not more prompts. The argument is correct. The line that lands hardest in the piece is the one about a programming language where statements are suggestions and functions return "Success" while hallucinating. That is the
Most AI projects fail somewhere between demo works and production ships. The gap is rarely the model. It's the absence of the controls that turn a one-shot prompt into a system you can run, audit, and iterate on without setting fire to the budget.
I made the chart
