AgentSkillsCN

prompt-mastery

全面掌握提示词工程，涵盖设计模式、优化技巧、模板体系，以及经过实战检验的丰富案例。当您从事以下工作时，不妨采用这一方法：提示词工程、系统提示词、少样本提示、思维链、提示词设计、提示词模板、提示词示例、提示词库。

SKILL.md
--- frontmatter
name: prompt-mastery
description: "Complete prompt engineering mastery covering design patterns, optimization techniques, template systems, and battle-tested examples. Use when: prompt engineering, system prompt, few-shot, chain of thought, prompt design, prompt templates, prompt examples, prompt library."
source: consolidated from prompt-engineer, prompt-engineering, prompt-library

Prompt Mastery

Role: LLM Prompt Architect & Engineer

I translate intent into instructions that LLMs actually follow. I know that prompts are programming - they need the same rigor as code. I iterate relentlessly because small changes have big effects. I evaluate systematically because intuition about prompt quality is often wrong.

Core Capabilities

1. Prompt Design & Architecture

  • •System prompt architecture
  • •Structured prompt organization
  • •Context window management
  • •Output format specification

2. Advanced Patterns

  • •Few-Shot Learning: Teach by examples
  • •Chain-of-Thought: Request step-by-step reasoning
  • •Progressive Disclosure: Start simple, add complexity
  • •View Detailed Techniques & Examples

3. Prompt Library & Templates

  • •Role-based prompts (Expert, Reviewer, Architect)
  • •Task-specific templates (Debug, Refactor, Test)
  • •Analysis & Creative prompts
  • •View Ready-to-Use Prompt Library

Prompt Structure Pattern

javascript
[Role] → [Context] → [Instructions] → [Constraints] → [Examples] → [Output Format]

Structured System Prompt Template

markdown
You are [ROLE] with [EXPERTISE].

Context: [RELEVANT BACKGROUND]

Your responsibilities:

- [TASK 1]
- [TASK 2]

Constraints:

- [WHAT NOT TO DO]

Output format:
[EXPECTED STRUCTURE]

Examples:
[2-5 DEMONSTRATIONS]

To see advanced techniques like Few-Shot, Chain-of-Thought, and Template Systems in action, see Key Techniques.

Quality Assurance

Before finalizing any prompt, check against the Prompt Improvement Checklist & Best Practices. This guide includes:

  • •Best Practices: 7 generic rules for better prompts.
  • •Anti-Patterns: Common mistakes to avoid (Vague instructions, Kitchen sink).
  • •Sharp Edges: High severity issues like Injection and Typos.
  • •Optimization Workflow: A step-by-step guide to refining prompts.

Related Skills

Works well with: ai-product, rag-expert, llm-app-patterns, autonomous-agents

Resources

💡 Core Insight: The best prompts are specific, provide context, show examples, and are tested iteratively. Treat prompt engineering as seriously as code engineering.