Install any skill in seconds. Free to start, no credit card required.
Get Started Free →CLI command development specialist. Use when creating commands, designing argument parsing, automating tasks, or implementing CLI best practices. Specializes in command design patterns and user experience.
.claude/skills/aiskillstore-command-optimization/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-17 | ✗→✓ | ▲ Improved | 111% | 0% |
| case-05 | ✓→✗ | ▼ Worse | 84% | 0% |
| case-12 | ✓→✓ | = Same ✓ | 43% | 0% |
| case-01 | ✓→✓ | = Same ✓ | 55% | 0% |
| case-03 | ✓→✓ | = Same ✓ | 134% | 0% |
This skill specializes in creating, designing, and optimizing command-line interfaces. It focuses on command design patterns, argument parsing, task automation, and CLI best practices.
Create a command to generate React componentsDesign a CLI command for database migrationsImprove the user experience of this commandStructure:
Example:
bash# Good: Clear and actionable npm run generate:component UserProfile # Bad: Unclear purpose npm run make UserProfile
Structure:
Example:
bash# Good: Clear analysis command npm run analyze:dependencies -- --outdated # Bad: Vague command npm run check
Pattern:
Example:
bash# Single file lint src/utils.js # Directory lint src/ # Glob pattern lint "src/**/*.js"
Common Options:
--config: Custom configuration file--output: Output directory or format--verbose: Detailed logging--dry-run: Preview without execution--force: Override safety checksCheck:
Good Error Messages:
Example:
bash# Good Error: Configuration file not found: .eslintrc.json Run 'npm run init:config' to create it. # Bad Error: File not found
Input: Create a command to generate React components
Output:
markdown## Command: generate:component ### Usage
npm run generate:component ComponentName options]
### Options
- `--path <path>`: Output path (default: src/components)
- `--type <type>`: Component type (functional|class) (default: functional)
- `--with-tests`: Generate test file
- `--with-styles`: Generate stylesheet
### Examples
npm run generate:component UserCard
npm run generate:component UserCard --path src/features/users --with-tests --with-styles
### Implementation
// scripts/generate-component.js const componentName = process.argv2]; const options = parseArgs(process.argv.slice(3));
if (!componentName) { console.error('Error: Component name required'); console.log('Usage: npm run generate:component <Name>'); process.exit(1); }
generateComponent(componentName, options);
### Example 2: Database Migration Command
**Input**: Create a database migration command
**Output**:
bashnpm run db:migrate [migration-name] [options]
--create <table>: Create new migration for table--rollback: Rollback last migration--status: Show migration status--to <version>: Migrate to specific versionbash# Create migration npm run db:migrate --create users # Run migrations npm run db:migrate # Check status npm run db:migrate --status # Rollback npm run db:migrate --rollback
## Reference Files
- **`references/SLASH_COMMAND.template.md`** - Slash command template with frontmatter, dynamic context, and workflow structure
## Best Practices
### Command Design
1. **Clear Names**: Use descriptive, action-oriented names
2. **Consistent Patterns**: Follow project conventions
3. **Helpful Defaults**: Sensible defaults for common use cases
4. **Good Feedback**: Clear output and progress indicators
5. **Error Handling**: Graceful failure with helpful messages
### User Experience
- **Progressive Disclosure**: Show basic usage, advanced options in help
- **Validation**: Validate inputs early with clear errors
- **Confirmation**: Ask for confirmation on destructive operations
- **Dry Run**: Support --dry-run for preview
- **Verbose Mode**: --verbose for detailed output
### Documentation
- **Help Text**: Clear, concise help for each command
- **Examples**: Include practical examples
- **Error Messages**: Explain what went wrong and how to fix
- **README**: Document commands in project README
## Related Use Cases
- Creating project-specific commands
- Automating development tasks
- Building CLI tools
- Improving command usability
- Standardizing command patterns| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-12 | pass→pass | 11,937 | 17,079 | +43% | 1 | 1 | 0% | 1,921 | 2,754 | +43% | 0 | 0 | — |
case-01 | pass→pass | 10,671 | 9,474 | -11% | 1 | 1 | 0% | 1,826 | 2,839 | +55% | 0 | 0 | — |
case-02 | fail→fail | 11,923 | 12,635 | +6% | 1 | 1 | 0% | 2,057 | 3,549 | +73% | 0 | 0 | — |
case-03 | pass→pass | 6,365 | 6,467 | +2% | 1 | 1 | 0% | 986 | 2,306 | +134% | 0 | 0 | — |
case-04 | pass→pass | 6,176 | 16,059 | +160% | 1 | 1 | 0% | 991 | 2,572 | +160% | 0 | 0 | — |
case-05 | pass→fail | 9,220 | 8,950 | -3% | 1 | 1 | 0% | 1,410 | 2,599 | +84% | 0 | 0 | — |
case-06 | pass→pass | 7,679 | 4,527 | -41% | 1 | 1 | 0% | 1,304 | 2,033 | +56% | 0 | 0 | — |
case-07 | pass→pass | 9,006 | 8,009 | -11% | 1 | 1 | 0% | 1,524 | 2,603 | +71% | 0 | 0 | — |
case-08 | pass→pass | 9,415 | 15,662 | +66% | 1 | 1 | 0% | 1,503 | 2,571 | +71% | 0 | 0 | — |
case-09 | pass→pass | 17,605 | 14,399 | -18% | 1 | 1 | 0% | 2,649 | 3,565 | +35% | 0 | 0 | — |
case-10 | pass→pass | 10,823 | 8,225 | -24% | 1 | 1 | 0% | 1,812 | 2,554 | +41% | 0 | 0 | — |
case-11 | pass→pass | 16,395 | 23,221 | +42% | 1 | 1 | 0% | 2,440 | 4,070 | +67% | 0 | 0 | — |
case-13 | pass→pass | 3,835 | 3,304 | -14% | 1 | 1 | 0% | 496 | 1,742 | +251% | 0 | 0 | — |
case-14 | pass→pass | 3,878 | 6,314 | +63% | 1 | 1 | 0% | 600 | 2,229 | +272% | 0 | 0 | — |
case-15 | pass→pass | 13,211 | 8,215 | -38% | 1 | 1 | 0% | 1,514 | 2,464 | +63% | 0 | 0 | — |
case-16 | pass→pass | 11,020 | 10,583 | -4% | 1 | 1 | 0% | 1,801 | 2,939 | +63% | 0 | 0 | — |
case-17 | fail→pass | 8,207 | 9,265 | +13% | 1 | 1 | 0% | 1,341 | 2,828 | +111% | 0 | 0 | — |
case-18 | pass→pass | 13,114 | 11,461 | -13% | 1 | 1 | 0% | 2,355 | 3,444 | +46% | 0 | 0 | — |
case-19 | pass→pass | 5,695 | 3,727 | -35% | 1 | 1 | 0% | 882 | 1,788 | +103% | 0 | 0 | — |
case-20 | pass→pass | 4,838 | 5,117 | +6% | 1 | 1 | 0% | 754 | 2,089 | +177% | 0 | 0 | — |
case-21 | pass→pass | 10,332 | 8,061 | -22% | 1 | 1 | 0% | 1,469 | 2,770 | +89% | 0 | 0 | — |
case-22 | pass→pass | 4,648 | 3,477 | -25% | 1 | 1 | 0% | 767 | 1,788 | +133% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of 0 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.