A curated library of high-quality, professionally crafted prompts for common development tasks. Save time and improve your AI interactions with battle-tested prompt templates.

• 🎯 30+ Professional Prompts - Carefully crafted for common development tasks
• 🚀 Instant Access - Simple slash commands for quick execution
• 📖 Learn by Example - See prompt engineering best practices in action
• 🎨 Organized Categories - Easy to browse and discover prompts
• 🔧 Customizable - Fork and adapt prompts to your needs
• 💡 Educational - Learn prompt engineering principles

Install the extension using the Gemini CLI:

gemini extensions install yourusername/prompt-library-extension

Or install from a local directory:

git clone https://github.com/yourusername/prompt-library-extension.git
cd prompt-library-extension
gemini extensions link .

After installation, restart Gemini CLI and start using prompts:

# Review code for security issues
/code-review:security "your code here"

# Generate a README file
/docs:write-readme "describe your project"

# Create unit tests
/testing:generate-unit-tests "your function here"

# Explain a complex concept
/learning:explain-concept "recursion"

# Debug an error
/debugging:debug-error "paste your error here"

• /code-review:security - Deep security analysis
• /code-review:performance - Performance optimization
• /code-review:best-practices - Best practices review
• /code-review:refactor - Refactoring suggestions

• /docs:write-readme - Generate comprehensive README
• /docs:write-api-docs - Create API documentation
• /docs:write-changelog - Generate changelog
• /docs:write-contributing - Create contribution guidelines

• /testing:generate-unit-tests - Create unit tests
• /testing:generate-e2e-tests - Create end-to-end tests
• /testing:edge-cases - Identify edge cases
• /testing:coverage-analysis - Analyze test coverage

• /debugging:debug-error - Diagnose and fix errors
• /debugging:trace-issue - Root cause analysis
• /debugging:performance-profile - Performance profiling

• /architecture:design-api - Design RESTful APIs
• /architecture:design-database - Design database schemas
• /architecture:system-design - System architecture
• /architecture:design-patterns - Suggest design patterns

• /learning:explain-concept - Explain technical concepts
• /learning:eli5 - Explain like I'm 5
• /learning:compare-tech - Compare technologies
• /learning:roadmap - Create learning paths

• /writing:technical-blog - Write technical posts
• /writing:email - Draft professional emails
• /writing:presentation - Create presentation outlines

• /prompts:improve - Improve existing prompts
• /prompts:create-template - Create prompt templates
• /prompts:best-practices - Learn prompt tips

/code-review:security "
function loginUser(username, password) {
  const query = 'SELECT * FROM users WHERE username = ' + username;
  return db.execute(query);
}
"

Output: Comprehensive security analysis identifying SQL injection vulnerability, authentication issues, and providing secure code examples.

/testing:generate-unit-tests "
function calculateDiscount(price, discountPercent) {
  return price - (price * discountPercent / 100);
}
"

Output: Complete test suite with happy paths, edge cases, boundary values, and error conditions.

/learning:explain-concept "closures in JavaScript"

Output: Multi-level explanation from ELI5 to technical details, with examples, use cases, and best practices.

/architecture:design-api "A blog platform with posts, comments, users, and tags"

Output: Complete REST API design with endpoints, data models, authentication, and OpenAPI documentation.

• Pre-crafted prompts eliminate trial and error
• Instant access via simple commands
• Consistent, high-quality results

• Based on prompt engineering best practices
• Comprehensive, structured outputs
• Professional-grade results

• See examples of effective prompts
• Understand what makes prompts work
• Build your own custom prompts

• Quick access to common tasks
• Reduce context switching
• Focus on problem-solving, not prompt crafting

1. Create a new .toml file in the appropriate commands/ subdirectory:

mkdir -p commands/mycategory
touch commands/mycategory/myprompt.toml

2. Write your prompt:

prompt = """
# My Custom Prompt

Your prompt content here with {{args}} for user input.
"""

3. Restart Gemini CLI to load the new prompt

Fork the repository and edit any .toml file to customize prompts to your needs.

Based on the prompts in this library, here are key principles:

❌ "Review this code"
✅ "Perform a security analysis focusing on input validation, SQL injection, and XSS vulnerabilities"

Use clear sections, numbering, and formatting to guide the AI's response.

Specify the language, framework, use case, and constraints.

Ask for code examples, not just explanations.

Specify exactly how you want the response structured.

Make prompts reusable with {{args}} placeholders.

Contributions are welcome! Here's how you can help:

1. Add New Prompts - Share your best prompts
2. Improve Existing Prompts - Make them clearer and more effective
3. Report Issues - Found a bug or unclear prompt?
4. Suggest Categories - What prompt categories are missing?

1. Fork the repository
2. Create a feature branch: git checkout -b feature/new-prompt
3. Add your prompt in the appropriate category
4. Test it thoroughly
5. Submit a pull request with clear description

For more information about Gemini CLI extensions:

• Gemini CLI Documentation
• Extension Development Guide
• Model Context Protocol

MIT License - see LICENSE file for details.

• Inspired by the Gemini CLI community
• Built on prompt engineering best practices
• Thanks to all contributors

• 🐛 Report Issues
• 💬 Discussions
• 📧 Email: your.email@example.com

• Add more prompt categories
• Interactive prompt builder
• Prompt analytics (most used, success rate)
• Community prompt sharing
• Multi-language support
• Prompt versioning
• A/B testing for prompts

My name is Harish Garg. I am a solo developer. I write at my blog here.

I also maintain a MCP Resources Directory & a coding ai tools directory

Made with ❤️ for the Gemini CLI community

Star ⭐ this repo if you find it useful!