For very long-running tasks, either (a) prompt Claude to verify its work with a background agent when it's done, (b) use an agent Stop hook to do that more deterministically, or (c) use the ralph-wiggum plugin (originally dreamt up by @GeoffreyHuntley).
12/ Verify Long-Running Tasks with Background Agents
When this pattern helps
The practical value is reducing ambiguity before implementation and making the result observable. Turn the expected behavior into checks that can pass or fail; a confident explanation is not a substitute for a reproducible test.
重点不是让提示词更长，而是先消除实现歧义，并把预期结果变成可通过或失败的检查。模型解释得很肯定，不等于测试已经通过。