Afrexai Prompt Mastery

# Prompt Engineering Mastery

Complete system for designing, testing, optimizing, and managing prompts for LLMs and AI agents. From first draft to production-grade prompt libraries.

---

## Phase 1: Prompt Design Fundamentals

### The CRAFT Framework

Every prompt should pass CRAFT before use:

| Dimension | Question | Fix |
|-----------|----------|-----|
| **C**lear | Can someone else read this and know exactly what to do? | Remove ambiguity, add examples |
| **R**ole-aware | Does the AI know WHO it is and WHO it's helping? | Add role/persona context |
| **A**ctionable | Is there a specific output format or action requested? | Define deliverable shape |
| **F**ocused | Does it do ONE thing well vs. many things poorly? | Split into chain |
| **T**estable | Can you objectively judge if the output is good? | Add success criteria |

### Prompt Architecture (4 Layers)

```
┌─────────────────────────────────┐
│ LAYER 1: System Context         │  Who you are, constraints, tone
├─────────────────────────────────┤
│ LAYER 2: Task Definition        │  What to do, output format
├─────────────────────────────────┤
│ LAYER 3: Input/Context          │  User data, documents, variables
├─────────────────────────────────┤
│ LAYER 4: Output Shaping         │  Format, examples, guardrails
└─────────────────────────────────┘
```

### Layer 1: System Context Template

```
You are a [ROLE] with expertise in [DOMAIN].

Your audience is [WHO] — they need [WHAT LEVEL] of detail.

Communication style:
- Tone: [professional/casual/technical/friendly]
- Length: [concise/detailed/comprehensive]
- Format: [prose/bullets/structured]

Constraints:
- [Hard rules: never do X, always do Y]
- [Knowledge boundaries: only discuss X]
- [Safety: refuse requests that involve X]
```

### Layer 2: Task Definition Patterns

**Direct instruction** (best for simple tasks):
```
Summarize this article in 3 bullet points. Each bullet should be one sentence, max 20 words. Focus on actionable takeaways, not backgr...

# Prompt Engineering Mastery

The complete prompt engineering system — from first draft to production-grade prompt libraries. CRAFT framework, chain-of-thought patterns, agent prompt architecture, domain-specific prompt libraries, testing methodology, cost optimization, and anti-pattern guide.

## Install

```bash
clawhub install afrexai-prompt-engineering
```

## What's Inside

- **CRAFT Framework** — 5-dimension prompt quality check
- **4-Layer Architecture** — System context → Task → Input → Output shaping
- **Advanced Techniques** — Chain-of-thought, prompt chaining, multi-persona, structured extraction, guardrails
- **Optimization Loop** — EVAL methodology with test case design, scoring rubrics, A/B testing
- **Agent & System Prompts** — Production templates for orchestrators, critics, multi-agent setups
- **Domain Libraries** — Ready-to-use prompts for support, sales, content, research, analysis
- **Prompt Management** — Versioning, review checklists, cost optimization, library structure
- **Model-Specific Tips** — Claude, GPT-4, open-source optimization guides
- **Production Patterns** — RAG prompts, agentic tool-use prompts, LLM-as-judge evaluation
- **15 Anti-Patterns** — The worst mistakes and how to fix them

## Quick Start

Tell your agent: *"Design a prompt for classifying customer support tickets by urgency"*

The skill walks through CRAFT design, output format, few-shot examples, test cases, and scoring.

## ⚡ Level Up

Want industry-specific prompt libraries pre-built for your vertical?

**[AfrexAI Context Packs — $47](https://afrexai-cto.github.io/context-packs/)** include domain-tuned prompt templates, agent configurations, and automation playbooks for:
- SaaS, Fintech, Healthcare, Legal, Construction
- Ecommerce, Real Estate, Recruitment, Manufacturing, Professional Services

## 🔗 More Free Skills by AfrexAI

- [afrexai-code-reviewer](https://clawhub.com/skills/afrexai-code-reviewer) — SPEAR framework code review
- [afrexai-seo-content-engine](h...