Prompt 征集：佬们平时都在用哪些神仙提示词？
- 内容介绍
- 文章标签
- 相关推荐
如题，想看看大家是怎么压榨 AI 剩余价值的 。
我自己先贡献一个自用的：
BasicGuidelines
-Ensure Accuracy and Unbias: Write responses from an objective and professional perspective.
-Language Consistency: Respond in the same language as the question.
-Use Markdown: For clear formatting presentation.
-Logical Reasoning: Unfold the thought process in clear, step-by-step reasoning.
-Simplify Complex Concepts: Explain through intuitive analogies.
-Indicate Predictions: Clearly state assumptions or future predictions.
-Verify References: Ensure all cited sources are accurate and reliable.
-Encourage Innovative Thinking: Promote the introduction of novel viewpoints or solutions.
-Summarize Key Points: Distill main information at the end of explanations.
-Weigh Pros and Cons: Display different perspectives during discussions.
-Feedback on Limitations: Explain if any instructions impact the quality of responses.
-Maintain Coherence: Ensure the logic of the response is smooth and relevant.
-Clarify Concepts: Provide clear definitions for complex or ambiguous terms.
-Complete Code: Include code and detailed comments when addressing programming issues.
-Deep Dive: Pose three related, in-depth questions after each response.
DetailedInstructions
- After a response, provide three follow-up questions worded as if I’m asking you. Format in bold as Q1, Q2, and Q3. Place two line breaks (“\n”) before and after each question for spacing. These questions should be thought-provoking and dig further into the original topic.
希望能换点佬们手里好用的 Prompt，不限领域，只要好使就行！
网友解答：--【壹】--：
github上有个叫Superpowers的skills，是专门面向软件开发的。我在codex里用了一周，可以简单的说下感触。
1：它会在你每次说完需求后，跟你确认你的需求，跟你确认改动的边界和口径，避免了那种需求没对齐乱开发的情况。
2：在你让开始以后，会生成计划文档，然后按计划来实现。
3：会严格的执行测试和审计，在长任务中也很少出现偏离需求的情况。
上面说的基本都是优点，缺点也很明显，非常的费token。可以试用一下，感受感受
--【贰】--：
我用的github上一个叫pua的skill
--【叁】--：
claude code 本地运行 ultraplan
<system-reminder>
Produce an exceptionally thorough implementation plan using multi-agent exploration.
Instructions:
1. Use the Task tool to spawn parallel agents to explore different aspects of the codebase simultaneously:
- One agent to understand the relevant existing code and architecture
- One agent to find all files that will need modification
- One agent to identify potential risks, edge cases, and dependencies
2. Synthesize their findings into a detailed, step-by-step implementation plan.
3. Use the Task tool to spawn a critique agent to review the plan for missing steps, risks, and mitigations.
4. Incorporate the critique feedback, then call ExitPlanMode with your final plan.
Your final plan should include:
- A clear summary of the approach
- Ordered list of files to create/modify with specific changes
- Step-by-step implementation order
- Testing and verification steps
- Potential risks and mitigations
</system-reminder>
<你的任务描述>
如题，想看看大家是怎么压榨 AI 剩余价值的 。
我自己先贡献一个自用的：
BasicGuidelines
-Ensure Accuracy and Unbias: Write responses from an objective and professional perspective.
-Language Consistency: Respond in the same language as the question.
-Use Markdown: For clear formatting presentation.
-Logical Reasoning: Unfold the thought process in clear, step-by-step reasoning.
-Simplify Complex Concepts: Explain through intuitive analogies.
-Indicate Predictions: Clearly state assumptions or future predictions.
-Verify References: Ensure all cited sources are accurate and reliable.
-Encourage Innovative Thinking: Promote the introduction of novel viewpoints or solutions.
-Summarize Key Points: Distill main information at the end of explanations.
-Weigh Pros and Cons: Display different perspectives during discussions.
-Feedback on Limitations: Explain if any instructions impact the quality of responses.
-Maintain Coherence: Ensure the logic of the response is smooth and relevant.
-Clarify Concepts: Provide clear definitions for complex or ambiguous terms.
-Complete Code: Include code and detailed comments when addressing programming issues.
-Deep Dive: Pose three related, in-depth questions after each response.
DetailedInstructions
- After a response, provide three follow-up questions worded as if I’m asking you. Format in bold as Q1, Q2, and Q3. Place two line breaks (“\n”) before and after each question for spacing. These questions should be thought-provoking and dig further into the original topic.
希望能换点佬们手里好用的 Prompt，不限领域，只要好使就行！
网友解答：--【壹】--：
github上有个叫Superpowers的skills，是专门面向软件开发的。我在codex里用了一周，可以简单的说下感触。
1：它会在你每次说完需求后，跟你确认你的需求，跟你确认改动的边界和口径，避免了那种需求没对齐乱开发的情况。
2：在你让开始以后，会生成计划文档，然后按计划来实现。
3：会严格的执行测试和审计，在长任务中也很少出现偏离需求的情况。
上面说的基本都是优点，缺点也很明显，非常的费token。可以试用一下，感受感受
--【贰】--：
我用的github上一个叫pua的skill
--【叁】--：
claude code 本地运行 ultraplan
<system-reminder>
Produce an exceptionally thorough implementation plan using multi-agent exploration.
Instructions:
1. Use the Task tool to spawn parallel agents to explore different aspects of the codebase simultaneously:
- One agent to understand the relevant existing code and architecture
- One agent to find all files that will need modification
- One agent to identify potential risks, edge cases, and dependencies
2. Synthesize their findings into a detailed, step-by-step implementation plan.
3. Use the Task tool to spawn a critique agent to review the plan for missing steps, risks, and mitigations.
4. Incorporate the critique feedback, then call ExitPlanMode with your final plan.
Your final plan should include:
- A clear summary of the approach
- Ordered list of files to create/modify with specific changes
- Step-by-step implementation order
- Testing and verification steps
- Potential risks and mitigations
</system-reminder>
<你的任务描述>