Skip to content

nicholyx/ai-skills / .agents/skills/plugin-creator

plugin-creator

Verified

Comprehensive plugin development skill for Claude Code. Analyzes business requirements and creates complete plugins including MCP servers, skills, slash commands, or combinations thereof. Supports TypeScript, Python, and any other language. Provides architecture guidance, templates, and end-to-end implementation.

assets/templates/slash-command/README.md

View on GitHub
# Slash Command Template

Template for creating Claude Code slash commands.

## Quick Start

1. **Copy template**
   ```bash
   cp example-command.md your-command.md
   ```

2. **Customize the command**
   - Replace `command-name` with your command name
   - Update description
   - Write command prompt
   - Add examples

3. **Install command**
   ```bash
   cp your-command.md ~/.claude/commands/
   ```

4. **Test command**
   ```
   /your-command [arguments]
   ```

## Command Structure

```markdown
# command-name            ← Command name (used as /command-name)

Brief description        ← One-line description

## Prompt                 ← Required section

Command execution logic  ← What happens when command runs
```

## Naming Conventions

**Good command names:**
- `/deploy` - Clear, short action
- `/test:unit` - Namespaced for related commands
- `/db:migrate` - Category prefix for organization
- `/api:test` - Clear category and action

**Bad command names:**
- `/d` - Too short, unclear
- `/do-the-deployment-thing` - Too long
- `/testunit` - Missing namespace separator
- `/api_test` - Use colon, not underscore

## Using Arguments

### No Arguments
```markdown
## Prompt

Run all tests with full coverage.
```

### Optional Arguments
```markdown
## Prompt

Run tests {{args}}.

If args provided: use as test filter
Otherwise: run all tests
```

### Required Arguments
```markdown
## Prompt

Deploy to {{args}} environment.

If no environment specified:
  Display: "Usage: /deploy <environment>"
  Stop

Valid environments: dev, staging, prod
```

### Parsing Arguments
```markdown
## Prompt

Process arguments: {{args}}

Expected format: key=value pairs
Example: env=prod version=1.2.3

Steps:
1. Split {{args}} on spaces
2. Parse each as key=value
3. Extract environment and version
4. Proceed with deployment
```

## Integration Patterns

### With Skills
```markdown
## Prompt

Use the deployment-skill for deployment workflow.

Steps:
1. Load deployment skill
2. Reference deployment_checklist.md
3. Run deployment script
4. Verify with health check
```

### With MCP Tools
```markdown
## Prompt

Deploy using cloud MCP server.

Steps:
1. Use `build_image` tool
2. Use `push_to_registry` tool
3. Use `deploy_service` tool with {{args}} environment
4. Use `health_check` tool to verify
```

### Command Composition
```markdown
## Prompt

Run pre-deployment checks:

1. Execute /test command
2. Execute /lint command
3. Execute /build command

If all pass: proceed with deployment
Otherwise: stop and report failures
```

## Error Handling

```markdown
## Prompt

Validate inputs: {{args}}

Required: environment (dev/staging/prod)

If environment missing:
  Display: "Error: Environment required"
  Display: "Usage: /deploy <environment>"
  Stop

If environment invalid:
  Display: "Error: Invalid environment '{{args}}'"
  Display: "Valid options: dev, staging, prod"
  Stop

Proceed with deployment to {{args}}
```

## Best Practices

### Clear Feedback
```markdown
Good:
✓ Tests passed (45/45)
✓ Build complete
→ Deploying to production...
✓ Deployment successful

Bad:
Running...
Done.
```

### Progress Indication
```markdown
Deploying to staging...
1. ✓ Running tests
2. ✓ Building image
3. → Pushing to registry
4. ⏳ Deploying service
5. ⏳ Running health checks
```

### Helpful Errors
```markdown
Good:
Error: Tests failed (3 failures)
Run '/test:unit --verbose' to see details

Bad:
Error
```

### Confirmation for Dangerous Operations
```markdown
⚠️ PRODUCTION DEPLOYMENT

This will deploy to PRODUCTION environment.
Type 'CONFIRM' to proceed.

[Wait for user input]
```

## Examples

### Simple Command
```markdown
# format

Format code according to project standards.

## Prompt

Format code in current file:
1. Run appropriate formatter (prettier/black/etc)
2. Apply project style rules
3. Report changes made
```

### Parameterized Command
```markdown
# deploy

Deploy to specified environment.

## Prompt

Deploy to {{args}} environment.

Environments: dev, staging, prod

Steps:
1. Validate environment
2. Run pre-deploy checks
3. Build and deploy
4. Verify deployment
```

### Integrated Command
```markdown
# db:migrate

Run database migrations using database skill.

## Prompt

Run database migrations using database MCP server.

Steps:
1. Use `get_pending_migrations` tool
2. Display pending migrations
3. Ask for confirmation
4. Use `run_migration` tool for each
5. Report results
```

## Installation

Place command files in:
```
~/.claude/commands/your-command.md
```

Or for project-specific commands:
```
your-project/.claude/commands/your-command.md
```

## Testing Checklist

- [ ] Command name is clear and follows conventions
- [ ] Description accurately describes what command does
- [ ] Arguments are documented and validated
- [ ] Error messages are clear and helpful
- [ ] Integration with skills/MCP works correctly
- [ ] Examples demonstrate usage accurately
- [ ] Dangerous operations require confirmation

## Resources

- [Slash Command Guide](../../references/slash_command_guide.md)
- [Best Practices](../../references/best_practices.md)
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 471kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall