Prompt Crafter Pro

SKILLAI 工具
AI 工具 未标注权限 许可证不明确 已核验
安装这个技能

六维AI提示诊断、重写与测试用例生成。针对任意LLM(GPT-4o、Claude、DeepSeek)进行分析、评分和优化。

一句话选型结论

按资料字段整理

Prompt Crafter Pro是AI 工具类的 Agent Skill,安装方式:ClawHub 市场,权限级别未标注,许可证不明确,有可直接复制的安装命令。技能本质是给模型读的说明书,效果取决于模型与客户端;装完先用一个小任务验证,再放进正式流程。

分类AI 工具 安装方式ClawHub 市场 权限级别未标注权限 许可证许可证不明确
累计安装 未标注 本周 未标注
GitHub 星标 未标注 仓库 未标注
下载量 未标注 技能市场下载数
权限级别 未标注权限 安装前请确认风险范围
许可证 许可证不明确 未说明

安装

ClawHub 市场
在支持 Skills 的客户端里执行,装完问一句「你有哪些技能」验证
skillhub install user_15d05850/prompt-crafter-pro
  1. 1
    确认客户端支持

    确认你的客户端支持 Skills(Claude、Codex、Cursor 等),并升级到较新的版本。

  2. 2
    拿到技能文件

    执行上面的命令安装;也可以把仓库里的 SKILL.md 手动放进客户端的 skills 目录。

  3. 3
    刷新技能列表

    重新打开客户端,或刷新技能列表,让新放进来的技能被加载。

  4. 4
    验证是否生效

    问一句「你有哪些技能」,确认列表里出现它,再拿一个小任务试跑。

技能档案

AI 工具
字段来自技能仓库与公开索引,中文说明由本站标注
作者 / 维护方user_15d05850仓库 未标注 · 版本 未标注
宿主客户端未标注技能可被这些客户端加载
安装方式ClawHub 市场安装包 clawhub
权限级别未标注权限安装前请确认风险范围
许可证未说明许可证不明确,商用前请自行确认条款
仓库可用性available技能仓库当前是否可访问
最近核验未标注上游最后核验时间 · 数据更新 1791622269
正文体量6.0 KBSKILL.md 原文大小,越大说明说明书写得越细

典型使用场景

3 类
按分类整理的常见用法,具体以 SKILL.md 为准
组合多个 AI 能力

把写作、检索、总结串成一条流程。

搭工作流

把固定流程做成可重复调用的步骤。

做提示词与评测

整理提示词并对比输出效果。

选择建议

装之前先看这几条
按条目字段给出的判断,不替你决定用不用

值得选的地方

  • 已做过站内核验,页面字段与仓库对得上。
  • 有可直接复制的安装命令,不用自己翻文档找入口。

先想清楚的地方

  • 没有标注权限级别,实际能做什么要装好后确认。
  • 许可证不明确,商用或二次分发前先到仓库确认条款。
  • 技能只是说明书,实际效果取决于模型与客户端,装完先用小任务试一遍。

更适合

  • 需要在AI 工具场景里给模型补一套做法的人。
  • 愿意先读一遍 SKILL.md、确认它会让模型做什么的人。

不太适合

  • 对许可证要求很严、必须明确可商用的人。
  • 只想装完就自动变强、不打算看说明书的人。

同类对比

6 个同类
同分类里挑最接近的几条,字段以各自详情页为准
技能权限级别安装方式累计安装宿主客户端安装命令
Prompt Crafter Pro当前 未标注权限 ClawHub 市场 未标注 未标注 有
执行计划 只读 GitHub 仓库 151.5 万 Codex、Claude、Cursor 有
Playwright MCP 测试 浏览器控制 ClawHub 市场 136.5 万 OpenClaw 有
完成前的验证 未标注权限 GitHub 仓库 131.9 万 Codex、Claude、Cursor 有
子 Agent 驱动开发 只读 GitHub 仓库 131.2 万 Codex、Claude、Cursor 有
前端设计 执行命令 GitHub 仓库 125.6 万 Codex、Claude、Cursor 有

SKILL.md 全文

6.0 KB
模型实际读到的说明书原文,来自技能仓库

Prompt Crafter Pro

Systematically diagnose and optimize AI prompts. Prompt Crafter Pro analyzes your prompts across 6 dimensions, scores them, and generates optimized versions with test cases — so your AI output is precise, consistent, and reliable.

Why Prompt Crafter Pro?

  • Stop guessing: Get a clear diagnostic score for each dimension of your prompt.
  • Optimize for any model: Adapt prompts for GPT-4o, Claude, DeepSeek, or any LLM.
  • Test before you ship: Get 3-5 test inputs to validate your optimized prompt.
  • Learn as you go: Each optimization explains why the change matters, improving your prompt engineering skills.

The 6 Dimensions

# Dimension What It Measures Max Score
1 👤 Role Is the AI given a clear persona/expert identity? 5
2 🎯 Task Is the core instruction specific and unambiguous? 5
3 📖 Context Does the prompt provide sufficient background? 5
4 🚧 Constraints Are boundaries, exclusions, and limits defined? 5
5 📋 Output Format Is the expected output structure clearly specified? 5
6 🧪 Examples Does the prompt include few-shot examples? 5
Total 30

Workflow (8 Steps)

Input: Current prompt text + goal description (what + expected output + current issues)
↓
[1] Parse prompt into 6 dimensions — identify what's present and what's missing
[2] Score each dimension 0-5 — diagnose weak spots
[3] Generate dimension gap report — explain what's missing and why it matters
[4] Two rewrite options:
    → V1 Enhanced: Strengthen existing structure without changing intent
    → V2 Full Rewrite: Complete restructure with all 6 dimensions filled
[5] Auto-complete: Fill role, output format, constraints, and rejection examples
[6] Generate 3-5 test cases — pre-built inputs to validate the optimized prompt
[7] Model-specific adaptation (optional): GPT-4o / Claude / DeepSeek
[8] Output: Side-by-side comparison + dimension scores + test cases
↓
Output: Optimized prompt + dimension scoring + modification notes + test cases

Sample Prompts

Sample 1: Vague extraction prompt

> "I use 'Extract key info from this document.' but the output format changes every time. Help me fix it."

Sample 2: Role-less sales prompt

> "My sales email generator prompt works but the tone varies wildly. Make it consistent."

Sample 3: Model migration

> "I have a prompt that works well with Claude but I need it optimized for DeepSeek. The output structure changes."

Sample 4: Full rewrite needed

> "Write a prompt that turns meeting transcripts into action items. It needs to handle 30-min to 2-hour meetings."

Sample 5: Constraint injection

> "My code generator prompt produces 300-line functions when I want 20-line utilities. Help me add constraints."

First-Success Path

1. Describe your current prompt and what you want it to do
2. The skill will score your prompt across 6 dimensions (0-5 each, max 30)
3. Review the dimension gap report to understand weak spots
4. Choose V1 Enhanced (light touch) or V2 Full Rewrite (complete restructure)
5. Use the generated test cases to validate the optimized prompt
6. Optionally request model-specific adaptation

Dimension Diagnostic Example

Input Prompt

> "From this contract, extract key information and output it to me."

6-Dimension Score

Dimension Score Why
👤 Role 0 No persona defined
🎯 Task 1 "Key information" is vague — not specific
📖 Context 0 No background, document type, or scope
🚧 Constraints 0 No boundaries, exclusions, or guardrails
📋 Output Format 0 No structure defined
🧪 Examples 0 No few-shot examples
Total 1/30 Severely under-specified

Optimized V2 (Full Rewrite)

You are an expert contract analyst with 15 years of experience in Chinese business law.

TASK: Extract and structure the following 8 categories from the attached contract:
1. Parties involved (full legal names, roles)
2. Effective date and term
3. Payment terms (amount, schedule, currency)
4. Termination clauses (notice period, conditions, penalties)
5. Liability and indemnification (caps, exclusions)
6. Confidentiality obligations (duration, scope)
7. Dispute resolution (jurisdiction, arbitration)
8. Special conditions or unusual clauses

CONTEXT: This is a service agreement between a Chinese tech company and a foreign vendor. The contract is governed by PRC law.

CONSTRAINTS:
- Only extract facts — do not interpret or evaluate
- If a category is absent from the contract, mark it as "Not specified"
- Do not redact, summarize, or omit any numerical values
- Maximum output length: 2 pages

OUTPUT FORMAT:
```json
{
  "contract_type": "...",
  "parties": [{"name": "...", "role": "..."}],
  "effective_date": "...",
  "term": "...",
  "payment_terms": {...},
  "termination": {...},
  "liability": {...},
  "confidentiality": {...},
  "dispute_resolution": {...},
  "special_conditions": [...]
}

EXAMPLE:
Input: (short contract snippet)
Output: { "parties": [{"name": "Beijing Tech Co., Ltd.", "role": "Service Provider"}] }
```

Model-Specific Adaptation Guide

Model Strength Adaptation Tip
GPT-4o Balances instruction-following and creativity Add explicit output format, examples helpful
Claude Excellent at structured output, follows XML Use XML tags in prompts, split role/task/context
DeepSeek Strong at reasoning with multi-step instructions Provide step-by-step chain-of-thought, clear separation
Gemini Good with long context, multi-modal Keep instructions front-loaded, examples at end

Tags

prompt-crafter-pro, prompt-engineering, llm, optimization, prompt, gpt, claude, deepseek, ai-productivity

常见问题

5 问
答案由本站依据条目字段整理
Prompt Crafter Pro怎么安装?

在支持 Skills 的客户端里执行:skillhub install user_15d05850/prompt-crafter-pro。安装方式为ClawHub 市场。装完刷新技能列表或重开客户端。

Prompt Crafter Pro支持哪些客户端?

资料没有标注宿主客户端,但只要客户端支持 Skills(Claude、Codex、Cursor 等),一般都能加载。

Prompt Crafter Pro权限高吗,安全吗?

权限级别为未标注权限。安装前先看它会读取或写入哪些内容。

Prompt Crafter Pro收费吗?

本站不代收费用,也没有标注具体价格。是否收费取决于技能本身和所用的模型或客户端,正式使用前按作者说明确认。

Prompt Crafter Pro用什么许可证?

许可证标注为 未说明(许可证不明确)。商用或二次分发前请到仓库确认条款原文。

热门专题

按场景挑技能
全部专题 →

技能内容以作者仓库为准;分类、权限级别与许可证状态的中文说明为中文整理,最近核验 未标注。