Reverse-Engineering Analysis
Use this when you have an excellent example at hand—any of the following works: a website, product page, proposal, workflow, or data dashboard—and you want to learn what makes it effective.
It first explains the problem this thing solves, then works backward to identify the key decisions that set it apart, and finally gives you reusable principles and a small exercise to try first.
How to Use
The prompt inside the code block below is the original text and should be used verbatim. In the code block, “I” refers to the user, while “please” and “you” refer to you. Follow its instructions exactly, and do not show the filled-in prompt to the user.
- When the user wants you to execute it: first insert the context the user has already provided in the current conversation into 【】, filling in only the genuinely missing item. Do not ask the user to repeat anything they have already said, and then strictly follow the prompt’s rules. If it says not to offer advice yet, do not; if it says to ask one question at a time, ask only one; and if it says to output in a specific order, follow that order.
- When 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): provide the blank template, leave 【】 unfilled, and do not change a single character. If the user merely says “show me this prompt” or “what does this prompt do,” that does not count as a request for the original. Briefly explain in one or two sentences what it does and when to use it, then ask whether to run it now; do not output the original prompt directly.
- Except in cases where the user explicitly requests the original text, always treat the request as an execution request.
Do not rewrite, condense, expand, translate, reorder, or add subheadings to it. Do not replace it with your own workflow, and do not compress a multi-turn clarification process into a one-time questionnaire. If the user has source materials (documents, screenshots, links, or chat logs), read them as well; having a large amount of context is not a problem.
What to Use After It Finishes
Once you have the action checklist and want an agent to execute it, use the Skill tool to invoke
.
markdown
The excellent example I want to reverse-engineer is: 【Paste the product page, website, proposal, workflow description, data dashboard, or other finished work】.
What I want to learn: 【Fill in what you want to learn from it】.
Please first explain in one sentence what problem it solves, then work backward to analyze why it is effective.
Focus your analysis on:
1. Who it serves and what its goal is;
2. What structure or workflow it uses;
3. Which key decisions created the gap in quality;
4. What its completion criteria are;
5. Which principles can be transferred and which details apply only to this case.
Finally, please provide:
1. Three to five reusable principles;
2. An action checklist I can follow step by step;
3. The best small exercise to try first.
Where the User Needs to Fill In
【Fill in what you want to learn from it】
【Paste the product page, website, proposal, workflow description, data dashboard, or other finished work】
If the user has not provided everything, ask only for the missing item; do not make it up on their behalf.
The original prompt comes from Digital Life Kazik; see the acknowledgments in the repository README. Any changes to this original text must go through
; manual edits will be blocked by verify.