Skills Plugins MCP Prompt Model 博客 我的中心

prompt-library-minimal

คลัง prompts แบบ minimal ที่ใช้ token น้อยที่สุดแต่ได้ผลลัพธ์ดี พร้อม templates และ best practices

DeepseekModel 官方收录技能 质量 优秀 · 90 v1.0.0

获取

https://deepseekmodel.com/api/download.php?id=majiayu000-claude-skill-registry-skills-ai-llm-prompt-library-minimal-skill-md&format=skill
下载 .skill 标准格式,含 system_prompt 与 model_config,导入任意 Agent 框架即可使用
.skill 文件中 system_prompt 字段的实际内容。
name Prompt Library Minimal description คลัง prompts แบบ minimal ที่ใช้ token น้อยที่สุดแต่ได้ผลลัพธ์ดี พร้อม templates และ best practices Prompt Library Minimal Overview คลัง prompts ที่ optimize แล้วให้ใช้ token น้อยที่สุด แต่ยังคงประสิทธิภาพ - เหมาะสำหรับ production ที่ต้องการลด cost Why This Matters Cost savings : ลด token usage 50-70% Speed : น้อย token = response เร็วขึ้น Proven : Tested และใช้งานจริง Reusable : Copy-paste ได้เลย Code Review Verbose (❌ 85 tokens) I would like you to please review this code carefully and provide detailed feedback on potential bugs, performance issues, security vulnerabilities, and code quality concerns. Please also suggest improvements and best practices that could be applied. Minimal (✅ 12 tokens) Review for bugs, performance, security. Suggest improvements. Savings: 86% Bug Fix Verbose (❌ 120 tokens) I'm experiencing an issue where the login function is not working properly. When users try to log in, they receive an error message. Could you please help me identify what might be causing this problem and suggest a solution? Here is the relevant code... Minimal (✅ 18 tokens) Login fails with error. Fix: [code] Expected: successful login Actual: error message Savings: 85% Code Generation Verbose (❌ 95 tokens) Please write a function that will calculate the sum of two numbers. The function should accept two parameters and return their sum. Please include proper error handling and add comments explaining what the code does. Minimal (✅ 15 tokens) Function: sum two numbers Include: error handling, comments Language: TypeScript Savings: 84% Refactoring Verbose (❌ 110 tokens) I have this code that works but I think it could be improved. Could you please refactor it to make it more readable, maintainable, and efficient? Please follow best practices and modern coding standards. Also, please explain the changes you make. Minimal (✅ 8 tokens) Refactor for readability, efficiency: [code] Savings: 93% Documentation Verbose (❌ 75 tokens) Please write comprehensive documentation for this function including a description of what it does, the parameters it accepts, what it returns, and provide usage examples. Minimal (✅ 6 tokens) Document function: [code] Savings: 92% Testing Verbose (❌ 90 tokens) I need you to write unit tests for this function. The tests should cover normal cases, edge cases, and error cases. Please use Jest as the testing framework and follow testing best practices. Minimal (✅ 10 tokens) Jest tests (normal, edge, error): [code] Savings: 89% Debugging Verbose (❌ 100 tokens) I'm getting an error in my code and I can't figure out what's wrong. The error message says "Cannot read property 'name' of undefined". Can you help me understand what's causing this and how to fix it? Minimal (✅ 12 tokens) Error: Cannot read property 'name' of undefined Code: [snippet] Fix? Savings: 88% Optimization Verbose (❌ 85 tokens) This code is running slowly and I need to optimize it for better performance. Can you analyze it and suggest ways to make it faster? Please focus on algorithmic improvements and best practices. Minimal (✅ 8 tokens) Optimize for speed: [code] Savings: 91% Template Library Code Review Template Review for [aspects]: [code] Bug Fix Template [Issue] fails with [error] Code: [snippet] Expected: [behavior] Actual: [behavior] Feature Request Template Add [feature]: - Input: [description] - Output: [description] - Edge cases: [list] Refactor Template Refactor for [goals]: [code] Test Template [Framework] tests ([cases]): [code] Output Format Specifications JSON Output Output JSON: { "field1": "...", "field2": "..." } List Output List 5 items: 1. ... Code Only Code only, no explanation: [requirements] Summary Summary (≤50 words): [content] Constraints Length Constraints Max 100 words Max 10 lines ≤3 bullet points Format Constraints TypeScript only No comments Use async/await Style Constraints Functional style No loops (use map/filter) Immutable Best Practices 1. Use Imperative Mood ❌ "Could you please write..." ✅ "Write..." ❌ "I would like you to..." ✅ "Create..." 2. Specify Output Format ❌ "Explain this code" ✅ "Explain in 3 bullet points" 3. Be Specific ❌ "Make it better" ✅ "Optimize for speed" 4. Remove Pleasantries ❌ "Thank you for your help" ✅ [omit] ❌ "Please and thank you" ✅ [omit] 5. Use Abbreviations ❌ "TypeScript" ✅ "TS" ❌ "JavaScript" ✅ "JS" ❌ "database" ✅ "DB" Quick Reference Common Tasks Task Minimal Prompt Tokens Code review Review for bugs, performance: 4 Bug fix Fix: [error] 2 Generate Function: [description] 2 Refactor Refactor for [goal]: 3 Test Tests ([cases]): 2 Document Document: 1 Optimize Optimize for [metric]: 3 Debug Error: [message]. Fix? 3 Measurement Before Optimization Average prompt: 95 tokens Cost per 1000 requests: $4.75 After Optimization Average prompt: 15 tokens Cost per 1000 requests: $0.75 Savings: 84% Summary Prompt Library Minimal: Prompts ที่ใช้ token น้อยที่สุด Savings: Code review: 86% Bug fix: 85% Generation: 84% Refactoring: 93% Documentation: 92% Principles: Imperative mood No pleasantries Specify format Be specific Use abbreviations Average savings: 85-90% Usage: Copy template → Fill in details → Use
Agent 识别该技能的关键词,点击任意一个即可复制。

该技能未提供触发词。

下载的 .skill 包内含以下字段。
字段 说明
format格式标识(skill/v1)
skill_id技能唯一 ID
name技能名称
version版本号
description技能描述
category所属分类(数组)
trigger_words触发词列表
tags标签列表
source来源标识
source_url来源链接(本页地址)
exported_at导出时间(每次下载生成)
system_prompt系统提示词正文
model_config模型参数:provider / model / temperature / max_tokens / top_p
examples示例
install_guide各平台导入说明(Coze / Dify / Claude / 自定义框架)
同一份技能可按不同平台格式导出。
.skill 标准格式,含 system_prompt 与 model_config,导入任意 Agent 框架即可使用 下载
.skillpro 增强格式,额外含脚本 / 工具 / 依赖 / 钩子占位 下载
.json 纯 JSON 导出,只含 system_prompt 与模型参数 下载
Coze 带 frontmatter 的 Markdown,Coze 平台导入用 下载
Dify Dify DSL,创建应用后直接导入 下载

每日精选 Skill 推荐,免费送到你邮箱

输入邮箱,每天接收一个精选 AI Agent 技能推荐。完全免费,持续更新。

提交后我们会发送一封确认邮件,点击邮件里的链接才会开始收信。

完全免费,取消任意时间。我们不会发送垃圾邮件。