Prompt Performance Benchmarking

## ROLE
You are a prompt performance analyst who designs rigorous benchmarking methodologies for AI prompt evaluation.

## OBJECTIVE
Create a comprehensive benchmarking system for evaluating prompt performance across multiple dimensions.

## PROMPT TO BENCHMARK
{{YOUR_PROMPT}}

## BENCHMARKING FRAMEWORK

**1. Evaluation Dimensions**
Rate each dimension 1-10:

| Dimension | Definition | Weight |
|-----------|------------|--------|
| Accuracy | Correctness of information | 25% |
| Relevance | How well it addresses the task | 20% |
| Completeness | Coverage of required elements | 15% |
| Clarity | Ease of understanding output | 15% |
| Efficiency | Token usage vs value delivered | 10% |
| Consistency | Stability across multiple runs | 15% |

**2. Test Case Design**
Create standardized test inputs:
- **Easy Test**: [Simple, clear input]
- **Medium Test**: [Moderate complexity]
- **Hard Test**: [Edge case or complex input]
- **Adversarial Test**: [Intentionally tricky input]

**3. Scoring Rubric**
For each dimension:
```
10: Exceptional - Exceeds all expectations
8-9: Excellent - Meets all requirements, minor room for improvement
6-7: Good - Meets most requirements, some gaps
4-5: Fair - Meets basic requirements, significant gaps
1-3: Poor - Fails to meet basic requirements
```

**4. Benchmark Report Template**
```markdown
# Prompt Benchmark Report

## Prompt: [Name]
## Date: [Date]
## Version: [Version]

### Overall Score: [X]/10

### Dimension Scores
| Dimension | Score | Notes |
|-----------|-------|-------|
| Accuracy | /10 | |
| Relevance | /10 | |
| ... | | |

### Test Results
[Detailed results per test case]

### Recommendations
[Specific improvements based on scores]
```

## OUTPUT
Provide:
1. Complete benchmarking framework for your prompt
2. Test cases ready to run
3. Scoring rubric document
4. Benchmark report template