Back to Details

minimum-experiment

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

最小实验

Minimum Experiment

纸上谈兵到头的时候，需要的是现实世界的反馈和数据。
它先找出决定背后最需要验证的三个假设，挑出最可能改变结论的那个，围绕它设计一个低成本、可逆、七天内能做完的实验，最后告诉你明天就能开始的第一个动作。
When armchair theorizing has reached its limit, you need feedback and data from the real world.
It first identifies the three hypotheses behind the decision that most need validation, selects the one most likely to change the conclusion, designs a low-cost and reversible experiment around it that can be completed within seven days, and finally tells you the first action you can start tomorrow.

怎么用

How to Use

下面代码块里的提示语是原文，逐字使用。代码块里的「我」指用户，「请你」「你」指你自己，照它做事，不要把填好的提示语打给用户看。
  • 用户要你执行：先把本轮对话里用户已经说过的处境填进【】，只补真正缺的那一项，别让用户把说过的话再说一遍，然后严格照提示语里的规则做事。它说先别给建议就先别给，说一次只问一个问题就一次只问一个，说按顺序输出就按那个顺序输出。
  • 用户要原文（明确说要复制、要粘贴、发我原文、给我提示语这类话才算）：打空白版，【】保持不填，一个字不改。用户只是说「看看这条提示语」「这条是干嘛的」，不算要原文，用一两句说清它干什么、什么时候用，再问要不要现在跑，别直接倒原文。
  • 除了明确要原文的情况，其余一律当执行处理。
禁止改写、精简、扩写、翻译、重排、加小标题，禁止把它替换成你自己的流程，也禁止把多轮追问压缩成一次性问卷。用户手上有原始材料（文档、截图、链接、聊天记录）就一并读进来，上下文多不是问题。
The prompt in the code block below is the original text and must be used verbatim. In the code block, 「我」 refers to the user, while 「请你」 and 「你」 refer to you. Follow its instructions as written, and do not send the completed prompt to the user.
  • If the user asks you to execute it: First fill the situation the user has already described in this conversation into 【】, filling in only the one item that is genuinely missing. Do not ask the user to repeat what they have already said. Then strictly follow the rules in the prompt. If it says not to give advice yet, do not give advice; if it says to ask one question at a time, ask only one question; if it says to output things in order, follow that order.
  • If the user asks for the original text (only explicit requests such as “send me the original,” “I want to copy it,” “paste it for me,” or “give me the prompt” count): Output the blank version, leave 【】 unfilled, and do not change a single word. If the user merely says “show me this prompt” or “what does this do?”, that does not count as asking for the original text. Explain in one or two sentences what it does and when to use it, then ask whether to run it now. Do not simply dump the original text.
  • Unless the user explicitly asks for the original text, always treat the request as an instruction to execute it.
Do not rewrite, condense, expand, translate, reorder, or add subheadings. Do not replace it with your own workflow, and do not compress a multi-turn questioning sequence into a one-time questionnaire. If the user has source materials (documents, screenshots, links, or chat logs), read them as well; having extensive context is not a problem.

跑完接哪条

Next Step After Completion

实验设计好了要交给 agent 跑，用 Skill 工具唤起 
kaz-leader
写任务书。
markdown
我正在纠结的是：【填写你的选择或想法】。

请先找出这个决定背后最需要验证的3个假设，再选出最可能改变最终结论的那一个。

围绕这个假设，帮我设计一个低成本、可逆、能在【7天或你能接受的周期】内完成的最小实验。

请写清：
1. 具体要做什么；
2. 需要投入多少时间和资源；
3. 观察什么指标；
4. 什么结果支持继续；
5. 什么结果提醒我停止；
6. 实验结束后能获得什么新信息。

最后告诉我，明天就能开始的第一个动作是什么。
Once the experiment design is ready, hand it to an agent to run. Use the Skill tool to invoke 
kaz-leader
to write the task brief.
markdown
我正在纠结的是：【填写你的选择或想法】。

请先找出这个决定背后最需要验证的3个假设，再选出最可能改变最终结论的那一个。

围绕这个假设，帮我设计一个低成本、可逆、能在【7天或你能接受的周期】内完成的最小实验。

请写清：
1. 具体要做什么；
2. 需要投入多少时间和资源；
3. 观察什么指标；
4. 什么结果支持继续；
5. 什么结果提醒我停止；
6. 实验结束后能获得什么新信息。

最后告诉我，明天就能开始的第一个动作是什么。

用户要填的位置

Items the User Needs to Fill In

  • 【7天或你能接受的周期】
  • 【填写你的选择或想法】
用户没给全就先问缺的那一条，别自己替他编。
提示语原文来自数字生命卡兹克，出处见仓库 README 的致谢。改动这段原文要走 
scripts/sync.py
，手改会被 verify 拦下。
  • 【7天或你能接受的周期】
  • 【填写你的选择或想法】
If the user has not provided all of them, first ask for the missing one. Do not invent it for them.
The original prompt comes from Digital Life Kazik; see the acknowledgments in the repository README for its source. To modify this original text, use 
scripts/sync.py
; manual edits will be blocked by verify.