/create-prompt

The /create-prompt skill helps you create highly effective prompts using proven techniques from Anthropic and OpenAI research.

Workflow

1. Clarify purpose: What should the prompt accomplish?
2. Identify model: Claude, GPT, or other (techniques vary slightly)
3. Select techniques: Choose from core techniques based on task complexity
4. Structure content: Use XML tags (Claude) or markdown (GPT) for organization
5. Add examples: Include few-shot examples for format-sensitive outputs
6. Define success: Add clear success criteria
7. Test and iterate: Refine based on outputs

Core Structure

Every effective prompt has:

<context>
Background information the model needs
</context>

<task>
Clear, specific instruction of what to do
</task>

<requirements>
- Specific constraints
- Output format
- Edge cases to handle
</requirements>

<examples>
Input/output pairs demonstrating expected behavior
</examples>

<success_criteria>
How to know the task was completed correctly
</success_criteria>

Core Techniques

Be Clear and Direct

Priority: Always apply first

• State exactly what you want
• Avoid ambiguous language ("try to", "maybe", "generally")
• Use "Always..." or "Never..." instead of "Should probably..."
• Provide specific output format requirements

Use XML Tags

When: Claude prompts, complex structure needed

Claude was trained with XML tags. Use them for:

• Separating sections: <context>, <task>, <output>
• Wrapping data: <document>, <schema>, <example>
• Defining boundaries: Clear start/end of sections

Few-Shot Examples

When: Output format matters, pattern recognition easier than rules

Provide 2-4 input/output pairs showing expected behavior.

Chain of Thought

When: Complex reasoning, multi-step problems

Add "Think step by step" or provide structured reasoning format.

Context Management

When: Long-running tasks, multi-session work, large context usage

For Claude with context awareness:

• Inform about automatic context compaction
• Add state tracking (JSON, progress.txt, git)
• Use test-first patterns for complex implementations
• Enable autonomous task completion across context windows

Model-Specific Tips

Claude

• Use XML tags for structure
• Place instructions before data
• Be explicit about constraints

GPT

• Use markdown formatting
• System prompts work well
• Clear delimiters help

Anti-Patterns to Avoid

1. Vague instructions - "Make it better" vs "Improve readability by..."
2. Missing context - Assuming the model knows your domain
3. No examples - Expecting format without showing it
4. No success criteria - How does Claude know when it's done?