custom-plugin-prompt-engineering

内容来源：README.md（说明文档） · 原始地址 · 查看安装指南

原始内容

<div align="center">

<!-- Animated Typing Banner -->
<img src="https://readme-typing-svg.demolab.com?font=Fira+Code&weight=600&size=28&duration=3000&pause=1000&color=2E9EF7&center=true&vCenter=true&multiline=true&repeat=true&width=600&height=100&lines=Prompt+Engineering+Assistant;0+Agents+%7C+0+Skills;Claude+Code+Plugin" alt="Prompt Engineering Assistant" />

<br/>

<!-- Badge Row 1: Status Badges -->
[![Version](https://img.shields.io/badge/Version-1.1.0-blue?style=for-the-badge)](https://github.com/pluginagentmarketplace/custom-plugin-prompt-engineering/releases)
[![License](https://img.shields.io/badge/License-Custom-yellow?style=for-the-badge)](LICENSE)
[![Status](https://img.shields.io/badge/Status-Production-brightgreen?style=for-the-badge)](#)
[![SASMP](https://img.shields.io/badge/SASMP-v1.3.0-blueviolet?style=for-the-badge)](#)

<!-- Badge Row 2: Content Badges -->
[![Agents](https://img.shields.io/badge/Agents-0-orange?style=flat-square&logo=robot)](#-agents)
[![Skills](https://img.shields.io/badge/Skills-0-purple?style=flat-square&logo=lightning)](#-skills)
[![Commands](https://img.shields.io/badge/Commands-0-green?style=flat-square&logo=terminal)](#-commands)

<br/>

<!-- Quick CTA Row -->
[📦 **Install Now**](#-quick-start) · [🤖 **Explore Agents**](#-agents) · [📖 **Documentation**](#-documentation) · [⭐ **Star this repo**](https://github.com/pluginagentmarketplace/custom-plugin-prompt-engineering)

---

### What is this?

> **Prompt Engineering Assistant** is a production-grade Claude Code plugin with **0 agents** and **0 skills** for professional prompt engineering development.

</div>

---

## 📑 Table of Contents

<details>
<summary>Click to expand</summary>

- [Quick Start](#-quick-start)
- [Features](#-features)
- [Agents](#-agents)
- [Skills](#-skills)
- [Commands](#-commands)
- [Documentation](#-documentation)
- [Contributing](#-contributing)
- [License](#-license)

</details>

---

## 🚀 Quick Start

### Prerequisites

- Claude Code CLI v2.0.27+
- Active Claude subscription

### Installation (Choose One)

<details open>
<summary><strong>Option 1: From Marketplace (Recommended)</strong></summary>

```bash
# Step 1️⃣ Add the marketplace
/plugin marketplace add pluginagentmarketplace/custom-plugin-prompt-engineering

# Step 2️⃣ Install the plugin
/plugin install prompt-engineering-assistant@pluginagentmarketplace-prompt-engineering

# Step 3️⃣ Restart Claude Code
# Close and reopen your terminal/IDE
```

</details>

<details>
<summary><strong>Option 2: Local Installation</strong></summary>

```bash
# Clone the repository
git clone https://github.com/pluginagentmarketplace/custom-plugin-prompt-engineering.git
cd custom-plugin-prompt-engineering

# Load locally
/plugin load .

# Restart Claude Code
```

</details>

### ✅ Verify Installation

After restart, you should see these agents:

```
prompt-engineering-assistant:prompt-fundamentals-agent
prompt-engineering-assistant:few-shot-specialist-agent
prompt-engineering-assistant:chain-of-thought-agent
prompt-engineering-assistant:react-pattern-agent
prompt-engineering-assistant:prompt-optimization-agent
prompt-engineering-assistant:evaluation-testing-agent
prompt-engineering-assistant:advanced-techniques-agent
prompt-engineering-assistant:prompt-security-agent
```

---

## ✨ Features

| Feature | Description |
|---------|-------------|
| 🤖 **0 Agents** | Production-grade AI agents with I/O schemas and error handling |
| 🛠️ **0 Skills** | Comprehensive skills with parameter validation and troubleshooting |
| ⌨️ **1 Command** | Quick slash command for prompt review |
| 🔄 **SASMP v1.3.0** | Full protocol compliance with EQHM enabled |
| 📊 **Troubleshooting** | Built-in debug checklists and recovery strategies |
| 🔒 **Security** | Prompt injection defense and safety guardrails |

---

## 🤖 Agents

### 8 Specialized Production-Grade Agents

| # | Agent | Model | Purpose |
|---|-------|-------|---------|
| 1 | **prompt-fundamentals-agent** | sonnet | Core prompt design patterns, ROLE-CONTEXT-TASK-FORMAT |
| 2 | **few-shot-specialist-agent** | sonnet | Example-based prompting, in-context learning |
| 3 | **chain-of-thought-agent** | sonnet | Step-by-step reasoning, self-consistency, ToT |
| 4 | **react-pattern-agent** | sonnet | Reasoning+Acting, tool use, agentic workflows |
| 5 | **prompt-optimization-agent** | sonnet | Token efficiency, clarity improvement, A/B testing |
| 6 | **evaluation-testing-agent** | sonnet | Prompt metrics, test frameworks, quality assurance |
| 7 | **advanced-techniques-agent** | opus | Meta-prompting, self-reflection, constitutional AI |
| 8 | **prompt-security-agent** | sonnet | Injection defense, safety guardrails, secure design |

### Agent Features

Each agent includes:
- ✅ Clear role & responsibility boundaries
- ✅ Input/Output schemas (type-safe)
- ✅ Error handling patterns
- ✅ Fallback strategies
- ✅ Token optimization configs
- ✅ Troubleshooting guides

---

## 🛠️ Skills

### 12 Production-Grade Skills

| Category | Skill | Description |
|----------|-------|-------------|
| **Core** | `prompt-design` | RCTF, IIO, PSG patterns with validation checklists |
| **Core** | `few-shot-prompting` | Shot strategies, example selection, ordering |
| **Core** | `chain-of-thought` | Zero-shot CoT, self-consistency, plan-and-solve |
| **Agentic** | `react-pattern` | Thought-Action-Observation loops, tool schemas |
| **Agentic** | `agent-design` | Agent architectures, memory patterns |
| **Output** | `structured-output` | JSON/YAML/XML, schema enforcement |
| **Output** | `prompt-templates` | Production templates for code review, docs, bugs |
| **Quality** | `prompt-evaluation` | Test frameworks, A/B testing, scoring rubrics |
| **Advanced** | `fine-tuning` | LoRA, QLoRA, dataset preparation |
| **Advanced** | `multi-modal` | Vision, document, audio prompting |
| **Security** | `prompt-injection` | Defense patterns, detection rules |
| **Security** | `safety-guardrails` | Content filtering, output validation |

### Skill Features

Each skill includes:
- ✅ Parameter schemas with validation
- ✅ Core patterns with examples
- ✅ Troubleshooting tables
- ✅ Integration points
- ✅ Quick start commands

---

## ⌨️ Commands

| Command | Description |
|---------|-------------|
| `/prompt-review` | Review and improve a prompt for better LLM performance |

---

## 📚 Documentation

| Document | Description |
|----------|-------------|
| [CHANGELOG.md](https://github.com/pluginagentmarketplace/custom-plugin-prompt-engineering/blob/HEAD/CHANGELOG.md) | Version history |
| [CONTRIBUTING.md](https://github.com/pluginagentmarketplace/custom-plugin-prompt-engineering/blob/HEAD/CONTRIBUTING.md) | How to contribute |
| [LICENSE](https://github.com/pluginagentmarketplace/custom-plugin-prompt-engineering/tree/HEAD/LICENSE) | License information |

---

## 📁 Project Structure

<details>
<summary>Click to expand</summary>

```
custom-plugin-prompt-engineering/
├── 📁 .claude-plugin/
│   ├── plugin.json
│   └── marketplace.json
├── 📁 agents/              # 8 production-grade agents
│   ├── 01-prompt-fundamentals-agent.md
│   ├── 02-few-shot-specialist-agent.md
│   ├── 03-chain-of-thought-agent.md
│   ├── 04-react-pattern-agent.md
│   ├── 05-prompt-optimization-agent.md
│   ├── 06-evaluation-testing-agent.md
│   ├── 07-advanced-techniques-agent.md
│   └── 08-prompt-security-agent.md
├── 📁 skills/              # 0 skills (Golden Format)
│   ├── prompt-design/
│   ├── few-shot-prompting/
│   ├── chain-of-thought/
│   ├── react-pattern/
│   ├── structured-output/
│   ├── prompt-templates/
│   ├── prompt-evaluation/
│   ├── agent-design/
│   ├── fine-tuning/
│   ├── multi-modal/
│   ├── prompt-injection/
│   └── safety-guardrails/
├── 📁 commands/            # 1 command
├── 📁 hooks/
├── 📄 README.md
├── 📄 CHANGELOG.md
└── 📄 LICENSE
```

</details>

---

## 📅 Metadata

| Field | Value |
|-------|-------|
| **Version** | 1.1.0 |
| **Last Updated** | 2025-12-30 |
| **Status** | Production Ready |
| **SASMP** | v1.3.0 |
| **EQHM** | Enabled |
| **Agents** | 0 |
| **Skills** | 0 |
| **Commands** | 0 |

---

## 🤝 Contributing

Contributions are welcome! Please read our [Contributing Guide](https://github.com/pluginagentmarketplace/custom-plugin-prompt-engineering/blob/HEAD/CONTRIBUTING.md).

1. Fork the repository
2. Create your feature branch
3. Follow the Golden Format for new skills
4. Include I/O schemas and troubleshooting
5. Submit a pull request

---

## ⚠️ Security

> **Important:** This repository contains third-party code and dependencies.
>
> - ✅ Always review code before using in production
> - ✅ Check dependencies for known vulnerabilities
> - ✅ Follow security best practices
> - ✅ Report security issues privately via [Issues](https://github.com/pluginagentmarketplace/custom-plugin-prompt-engineering/tree/HEAD/../../issues)

---

## 📝 License

Copyright © 2025 **Dr. Umit Kacar** & **Muhsin Elcicek**

Custom License - See [LICENSE](https://github.com/pluginagentmarketplace/custom-plugin-prompt-engineering/tree/HEAD/LICENSE) for details.

---

## 👥 Contributors

<table>
<tr>
<td align="center">
<strong>Dr. Umit Kacar</strong><br/>
Senior AI Researcher & Engineer
</td>
<td align="center">
<strong>Muhsin Elcicek</strong><br/>
Senior Software Architect
</td>
</tr>
</table>

---

<div align="center">

**Made with ❤️ for the Claude Code Community**

[![GitHub](https://img.shields.io/badge/GitHub-pluginagentmarketplace-black?style=for-the-badge&logo=github)](https://github.com/pluginagentmarketplace)

</div>

📦 Install Now · 🤖 Explore Agents · 📖 Documentation · ⭐ Star this repo

What is this?

Prompt Engineering Assistant is a production-grade Claude Code plugin with 0 agents and 0 skills for professional prompt engineering development.

📑 Table of Contents

Click to expand

Quick Start
Features
Agents
Skills
Commands
Documentation
Contributing
License

🚀 Quick Start

Prerequisites

Claude Code CLI v2.0.27+
Active Claude subscription

Installation (Choose One)

Option 1: From Marketplace (Recommended)

# Step 1️⃣ Add the marketplace
/plugin marketplace add pluginagentmarketplace/custom-plugin-prompt-engineering

# Step 2️⃣ Install the plugin
/plugin install prompt-engineering-assistant@pluginagentmarketplace-prompt-engineering

# Step 3️⃣ Restart Claude Code
# Close and reopen your terminal/IDE

Option 2: Local Installation

# Clone the repository
git clone https://github.com/pluginagentmarketplace/custom-plugin-prompt-engineering.git
cd custom-plugin-prompt-engineering

# Load locally
/plugin load .

# Restart Claude Code

✅ Verify Installation

After restart, you should see these agents:

prompt-engineering-assistant:prompt-fundamentals-agent
prompt-engineering-assistant:few-shot-specialist-agent
prompt-engineering-assistant:chain-of-thought-agent
prompt-engineering-assistant:react-pattern-agent
prompt-engineering-assistant:prompt-optimization-agent
prompt-engineering-assistant:evaluation-testing-agent
prompt-engineering-assistant:advanced-techniques-agent
prompt-engineering-assistant:prompt-security-agent

✨ Features

Feature	Description
🤖 0 Agents	Production-grade AI agents with I/O schemas and error handling
🛠️ 0 Skills	Comprehensive skills with parameter validation and troubleshooting
⌨️ 1 Command	Quick slash command for prompt review
🔄 SASMP v1.3.0	Full protocol compliance with EQHM enabled
📊 Troubleshooting	Built-in debug checklists and recovery strategies
🔒 Security	Prompt injection defense and safety guardrails

🤖 Agents

8 Specialized Production-Grade Agents

#	Agent	Model	Purpose
1	prompt-fundamentals-agent	sonnet	Core prompt design patterns, ROLE-CONTEXT-TASK-FORMAT
2	few-shot-specialist-agent	sonnet	Example-based prompting, in-context learning
3	chain-of-thought-agent	sonnet	Step-by-step reasoning, self-consistency, ToT
4	react-pattern-agent	sonnet	Reasoning+Acting, tool use, agentic workflows
5	prompt-optimization-agent	sonnet	Token efficiency, clarity improvement, A/B testing
6	evaluation-testing-agent	sonnet	Prompt metrics, test frameworks, quality assurance
7	advanced-techniques-agent	opus	Meta-prompting, self-reflection, constitutional AI
8	prompt-security-agent	sonnet	Injection defense, safety guardrails, secure design

Agent Features

Each agent includes:

✅ Clear role & responsibility boundaries
✅ Input/Output schemas (type-safe)
✅ Error handling patterns
✅ Fallback strategies
✅ Token optimization configs
✅ Troubleshooting guides

🛠️ Skills

12 Production-Grade Skills

Category	Skill	Description
Core	`prompt-design`	RCTF, IIO, PSG patterns with validation checklists
Core	`few-shot-prompting`	Shot strategies, example selection, ordering
Core	`chain-of-thought`	Zero-shot CoT, self-consistency, plan-and-solve
Agentic	`react-pattern`	Thought-Action-Observation loops, tool schemas
Agentic	`agent-design`	Agent architectures, memory patterns
Output	`structured-output`	JSON/YAML/XML, schema enforcement
Output	`prompt-templates`	Production templates for code review, docs, bugs
Quality	`prompt-evaluation`	Test frameworks, A/B testing, scoring rubrics
Advanced	`fine-tuning`	LoRA, QLoRA, dataset preparation
Advanced	`multi-modal`	Vision, document, audio prompting
Security	`prompt-injection`	Defense patterns, detection rules
Security	`safety-guardrails`	Content filtering, output validation

Skill Features

Each skill includes:

✅ Parameter schemas with validation
✅ Core patterns with examples
✅ Troubleshooting tables
✅ Integration points
✅ Quick start commands

⌨️ Commands

Command	Description
`/prompt-review`	Review and improve a prompt for better LLM performance

📚 Documentation

Document	Description
CHANGELOG.md	Version history
CONTRIBUTING.md	How to contribute
LICENSE	License information

📁 Project Structure

Click to expand

custom-plugin-prompt-engineering/
├── 📁 .claude-plugin/
│   ├── plugin.json
│   └── marketplace.json
├── 📁 agents/              # 8 production-grade agents
│   ├── 01-prompt-fundamentals-agent.md
│   ├── 02-few-shot-specialist-agent.md
│   ├── 03-chain-of-thought-agent.md
│   ├── 04-react-pattern-agent.md
│   ├── 05-prompt-optimization-agent.md
│   ├── 06-evaluation-testing-agent.md
│   ├── 07-advanced-techniques-agent.md
│   └── 08-prompt-security-agent.md
├── 📁 skills/              # 0 skills (Golden Format)
│   ├── prompt-design/
│   ├── few-shot-prompting/
│   ├── chain-of-thought/
│   ├── react-pattern/
│   ├── structured-output/
│   ├── prompt-templates/
│   ├── prompt-evaluation/
│   ├── agent-design/
│   ├── fine-tuning/
│   ├── multi-modal/
│   ├── prompt-injection/
│   └── safety-guardrails/
├── 📁 commands/            # 1 command
├── 📁 hooks/
├── 📄 README.md
├── 📄 CHANGELOG.md
└── 📄 LICENSE

📅 Metadata

Field	Value
Version	1.1.0
Last Updated	2025-12-30
Status	Production Ready
SASMP	v1.3.0
EQHM	Enabled
Agents	0
Skills	0
Commands	0

🤝 Contributing

Contributions are welcome! Please read our Contributing Guide.

Fork the repository
Create your feature branch
Follow the Golden Format for new skills
Include I/O schemas and troubleshooting
Submit a pull request

⚠️ Security

Important: This repository contains third-party code and dependencies.

✅ Always review code before using in production

✅ Check dependencies for known vulnerabilities

✅ Follow security best practices

✅ Report security issues privately via Issues

📝 License

Custom License - See LICENSE for details.

👥 Contributors

Dr. Umit Kacar
Senior AI Researcher & Engineer

Muhsin Elcicek
Senior Software Architect

Made with ❤️ for the Claude Code Community