ai-promptinglisted

# AI Prompting (Haiku) **Purpose**: Quick tips and templates for effective prompt engineering - few-shot examples, chain-of-thought patterns, constraint specification, output formatting. **Model**: Haiku 4.5 (fast, practical prompt improvement) **When to Invoke Autonomously**: Use this skill when: 1. **Prompt Not Working Well**: "LLM output is inconsistent/poor quality" 2. **Need Examples**: "How to structure few-shot prompt for [task]?" 3. **Output Formatting**: "LLM not following desired format" 4. **Constraint Specification**: "How to enforce constraints in prompt?" 5. **Chain-of-Thought**: "Need step-by-step reasoning for [complex task]" **DO NOT invoke for**: - Complex prompt engineering (use AI LLM Expert agent) - Production prompt systems (use AI LLM Expert agent) - Already-working prompts ## Usage ``` /ai-prompting improve-prompt [current prompt] [issue] /ai-prompting few-shot [task description] /ai-prompting chain-of-thought [reasoning task] /ai-prompting output-format [desired structure] ``` ## Quick Prompt Patterns For detailed examples of each pattern, see [examples/prompt-patterns.md](examples/prompt-patterns.md). ### Common Patterns 1. **Few-Shot Examples**: Show 2-5 input→output examples (15-30% accuracy improvement) 2. **Chain-of-Thought**: "Think step-by-step" for reasoning tasks 3. **Output Format Specification**: Explicit structure with examples 4. **Constraint Enforcement**: Explicit boundaries, "DO NOT" statements 5. **Role-Based Prompting**: "