Afrexai Prompt Engineering

# Prompt Engineering Mastery

Complete methodology for writing, testing, and optimizing prompts that reliably produce high-quality outputs from any LLM. From first draft to production-grade prompt systems.

---

## Quick Health Check: /8

Run this diagnostic on any prompt:

| # | Check | Pass? |
|---|-------|-------|
| 1 | Clear task statement in first 2 sentences | |
| 2 | Output format explicitly specified | |
| 3 | At least one concrete example included | |
| 4 | Edge cases addressed | |
| 5 | Evaluation criteria defined | |
| 6 | No ambiguous pronouns or references | |
| 7 | Tested on 3+ diverse inputs | |
| 8 | Failure modes documented | |

Score: X/8. Below 6 = high risk of inconsistent outputs.

---

## Phase 1: Prompt Architecture

### The CRAFT Framework

Every effective prompt has five layers:

**C — Context**: What does the model need to know?
- Domain background, constraints, audience
- "You are reviewing legal contracts for a mid-market SaaS company"
- NOT "You are a helpful assistant" (too vague)

**R — Role**: Who should the model be?
- Specific expertise, experience level, perspective
- "You are a senior tax attorney with 15 years of cross-border M&A experience"
- Role selection guide:

| Task Type | Best Role | Why |
|-----------|-----------|-----|
| Technical writing | Senior technical writer at a developer tools company | Audience awareness |
| Code review | Staff engineer who's seen 10,000 PRs | Pattern recognition |
| Sales copy | Direct response copywriter (not "marketer") | Conversion focus |
| Analysis | Industry analyst at a top-3 consulting firm | Structured thinking |
| Creative | Genre-specific author (not "creative writer") | Voice consistency |

**A — Action**: What specifically should be done?
- Use imperative verbs: "Analyze", "Generate", "Compare", "Extract"
- One primary action per prompt (chain for multi-step)
- "Analyze this contract clause and identify: (1) risks to the buyer, (2) missing protections, (3) suggested redlines with...

# Prompt Engineering Mastery ⚡

The complete prompt engineering methodology — from first draft to production-grade prompt systems. Covers the CRAFT framework, 10+ advanced techniques, domain-specific patterns, testing protocols, and prompt management.

## Install

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

## What's Inside

- **CRAFT Framework** — 5-layer prompt architecture (Context, Role, Action, Format, Tests)
- **Core Techniques** — Chain-of-thought, few-shot design, constraint engineering, persona calibration
- **System Prompt Engineering** — Full agent/skill system prompt architecture with 100-point quality rubric
- **Advanced Techniques** — Prompt chaining, self-consistency, meta-prompting, RAG patterns, adversarial defense
- **Domain Patterns** — Ready-to-use templates for analysis, writing, code generation, extraction, evaluation
- **Testing Protocol** — 6-case test suite template, iteration methodology, failure pattern diagnosis
- **Production Management** — Versioning, monitoring, documentation, alert thresholds
- **Patterns Library** — Verifier, Decomposer, Devil's Advocate, Calibrator, Persona Switcher
- **Anti-Patterns** — 10 common mistakes with fixes

## Quick Start

Ask your agent:
- "Write a prompt for [any task]" — builds using CRAFT framework
- "Review this prompt" — scores against 100-point rubric
- "Test this prompt" — generates diverse test suite
- "Debug this prompt" — diagnoses failure patterns

## Why This Over Others

The existing prompt engineering skills on ClawHub are 1-2KB of bullet-point advice. This is 28KB of actionable methodology with templates, rubrics, decision matrices, and real examples. Zero dependencies.

---

## ⚡ Level Up

Get the full **AI Agent Context Pack** for your industry — 5 files of production-ready agent configurations, workflows, and domain knowledge:

👉 [Browse Context Packs ($47)](https://afrexai-cto.github.io/context-packs/)

## 🔗 More Free Skills by AfrexAI

- [afrexai-code-reviewer](https:/...