Your Agent Needs a Stop Condition: A Step‑by‑Step Build

Описание и применение

Создан скрипт‑стоп‑гейт, который предотвращает нежелательные действия агента, проверяя ограничения и уровень уверенности.

Задача

Agents can keep working after a task is finished, potentially making unwanted changes to files.

Результат

The stop gate script successfully intercepts unwanted agent actions, providing a clear human‑review step and preventing accidental file modifications.

Рабочий процесс

• 1. Clone MonkeyCode repo and obtain free access. 2. Create a project directory and install OpenAI SDK. 3. Write gate.mjs with system prompt, limits, and checks. 4. Run the gate script with a task argument. 5. If the agent outputs STOP or violates limits, hand back to human; otherwise, output is safe to show.
• 6. Verify the gate by testing with various tasks and observing the console logs.

Доказательства

An agent summarized it, then rewrote it. Nobody asked for a rewrite.

Источник

The fix is not a better prompt. The fix is a boundary you can test.

Источник