Use when you have a written implementation plan to execute in a separate session with review checkpoints
npx skills add sheikh-mohammad/project-a1-extract-your-human-job-into-skills --skill "prompt-engineering"
Install specific skill from multi-skill repository
# Description
|-
# SKILL.md
name: prompt-engineering
description: |-
This skill transforms raw prompts into clear, effective prompts using proven frameworks and best practices. It applies the six-part prompting framework and other advanced techniques to enhance prompt quality, specificity, and effectiveness. This skill should be used when users need to convert vague or basic prompts into professional-grade prompts that yield better results from AI models.
Prompt Engineering Skill
This skill enhances raw prompts by transforming them into clear, structured prompts using industry-standard frameworks and best practices. It leverages the six-part prompting framework and other advanced techniques to improve prompt effectiveness and output quality.
Core Capabilities
- Framework Application: Applies the six-part prompting framework to structure prompts effectively
- Best Practice Integration: Incorporates proven prompt engineering techniques
- Quality Enhancement: Improves clarity, specificity, and effectiveness of prompts
- Template Generation: Creates professional-grade prompts with optimal structure
The Six-Part Prompting Framework
The skill implements the comprehensive six-part framework for optimal prompt engineering:
1. Command: Start Strong, Not Soft
- Begins with clear, direct commands using strong action verbs
- Uses specific action verbs: analyze, create, design, recommend, generate, evaluate
- Avoids weak words like "give" or "help"
- Sets a professional, focused tone
2. Context: More is Always Better
- Provides comprehensive background information to narrow AI's interpretation range
- Implements the Rule of Three Framework:
- Who: Age, profession, experience level, situation
- What: Specific goal, constraints, requirements
- When: Timeline, deadlines, urgency
- Scales context appropriately based on complexity and importance
3. Logic: Define the Output Structure
- Specifies exactly how the AI should think and respond
- Defines clear output requirements and structure
- May include tables, checklists, or specific formatting
4. Roleplay: Transform Generic into Expert-Level
- Has the AI adopt a specific professional identity to improve response quality
- Assigns field-specific roles based on the prompt's domain
- Examples: "You are a certified financial advisor with 15 years of experience..."
5. Formatting: Structure for Success
- Organizes information in immediately useful and actionable ways
- Offers popular formatting options:
- Numbered lists for sequential steps
- Bullet points for equal-weight items
- Tables for comparing options
- Sections with headers for complex topics
- Summary boxes for key takeaways
6. Questions: The Secret Sauce
- Ends prompts with question requests to identify gaps and refine results
- Implements iterative questioning for deeper refinement
- Uses follow-up questions to enhance precision
Additional Prompt Engineering Techniques
- Zero-Shot: Direct requests without examples
- Few-Shot: Multiple examples to establish pattern (typically 3-5 optimal)
- Chain of Thought (CoT): Encourages step-by-step reasoning
- System Prompting: Sets overall context and behavior guidelines
- Role Prompting: Assigns specific expertise to the AI
Configuration Settings Guidance
- Temperature (0-1): Low (0-0.3) for focused responses, high (0.8-1.0) for creative outputs
- Output Length/Token Limits: Controls maximum response length
- Top-K and Top-P: Limit choices to top most likely tokens based on probability
Input Parameters
- Raw Prompt: The initial, potentially vague or basic prompt to be enhanced
- Target Domain: The specific field or application area for the prompt (optional)
- Output Requirements: Specific format or structure requirements (optional)
Output
An enhanced, professionally-structured prompt in JSON format that incorporates the six-part framework and other best practices, designed to produce superior results from AI models. The JSON output includes:
originalPrompt: The input prompt provided by the userenhancedPrompt: The improved prompt following the six-part frameworkframeworkApplied: Details of which framework elements were appliedimprovementNotes: Explanation of the enhancements madeexpectedBenefits: Expected improvements from the enhanced prompt
Before Implementation
Gather context to ensure successful implementation:
| Source | Gather |
|---|---|
| Codebase | Any existing prompt templates or guidelines in the project |
| Conversation | User's specific prompt to enhance and target requirements |
| Skill References | Framework details and best practices from references/ |
| User Guidelines | Any specific constraints or preferences for the enhanced prompt |
Ensure all required context is gathered before implementing.
Only ask user for THEIR specific requirements (domain expertise is in this skill).
# Supported AI Coding Agents
This skill is compatible with the SKILL.md standard and works with all major AI coding agents:
Learn more about the SKILL.md standard and how to use these skills with your preferred AI coding agent.