Singleshot Prompt Testing & Optimization Skill

Description

Prompt cost testing with single shot

Installation

javascript

▼ 展开

  Or: cargo install singleshot

When to Use

• Testing new prompts before openclaw implementation
• Benchmarking prompt variations for token efficiency
• Comparing model performance and costs
• Validating prompt outputs before production

Core Commands

Always use -d (detail) and -r (report) flags for efficiency analysis:

javascript

▼ 展开

Report Analysis Workflow

1. Generate Baseline

javascript

▼ 展开

2. Optimize & Compare

javascript

▼ 展开

Report Metrics

Reports contain:

javascript

▼ 展开

Optimization Strategies

1. 1. Test with cheaper models first:

javascript

▼ 展开

Reduce tokens:

• Shorten system prompts
• Use --max-tokens to limit output
• Add "be concise" to system prompt

Test locally (free):

javascript

▼ 展开

Example: Full Optimization

javascript

▼ 展开

Quick Reference

javascript

▼ 展开

Environment Variables

javascript

▼ 展开

Best Practices

1. 1. Always use -d for detailed token metrics
2. 2. Always use -r to save reports
3. 3. Always cat reports to analyze metrics
4. 4. Test variations and compare costs
5. 5. Set --max-tokens to control costs
6. 6. Use gpt-4o-mini for testing (cheaper)

Troubleshooting

• No metrics: Ensure -d flag is used
• No report file: Ensure -r flag is used
• High costs: Switch to gpt-4o-mini or Ollama
• Connection issues: Run singleshot ping -P <provider>

Singleshot Prompt Testing & Optimization Skill

Description

Prompt cost testing with single shot

Installation

javascript

▼ 展开

  Or: cargo install singleshot

When to Use

• Testing new prompts before openclaw implementation
• Benchmarking prompt variations for token efficiency
• Comparing model performance and costs
• Validating prompt outputs before production

Core Commands

Always use -d (detail) and -r (report) flags for efficiency analysis:

javascript

▼ 展开

Report Analysis Workflow

1. Generate Baseline

javascript

▼ 展开

2. Optimize & Compare

javascript

▼ 展开

Report Metrics

Reports contain:

javascript

▼ 展开

Optimization Strategies

1. 1. Test with cheaper models first:

javascript

▼ 展开

Reduce tokens:

• Shorten system prompts
• Use --max-tokens to limit output
• Add "be concise" to system prompt

Test locally (free):

javascript

▼ 展开

Example: Full Optimization

javascript

▼ 展开

Quick Reference

javascript

▼ 展开

Environment Variables

javascript

▼ 展开

Best Practices

1. 1. Always use -d for detailed token metrics
2. 2. Always use -r to save reports
3. 3. Always cat reports to analyze metrics
4. 4. Test variations and compare costs
5. 5. Set --max-tokens to control costs
6. 6. Use gpt-4o-mini for testing (cheaper)

Troubleshooting

• No metrics: Ensure -d flag is used
• No report file: Ensure -r flag is used
• High costs: Switch to gpt-4o-mini or Ollama
• Connection issues: Run singleshot ping -P <provider>