🚀 MCP Prompt Enhancer Server
Transform simple prompts into detailed, professional specifications using AI.
This MCP (Model Context Protocol) server enhances your prompts by adding context, clarity, structure, and professional requirements. Perfect for getting better results from Claude, ChatGPT, and other AI assistants.
✨ What Does This Do?
The Prompt Enhancer takes your basic prompts and transforms them into detailed, professional specifications:
Input: "Create a login form"
Enhanced Output:
# Login Form Implementation
Create a secure, accessible login form component designed for a modern web application.
## Technology Stack & Implementation:
- React functional component using Hooks
- CSS Modules for styling with encapsulation
- TypeScript for type safety
## Core Functionality & Props:
- `children` (ReactNode): Content rendered inside the button
- `onClick` (function): Handle click events
- `disabled` (boolean): Control button state
- `variant` ('primary' | 'secondary' | 'danger'): Visual styles
- `size` ('small' | 'medium' | 'large'): Button dimensions
## Accessibility Requirements:
- ARIA labels and roles
- Keyboard navigation support
- Focus indicators
- Screen reader compatibility
## Security Considerations:
- Input validation and sanitization
- CSRF protection
- Rate limiting
- Secure password handling
🎯 Quick Start
Step 1: Get Your Gemini API Key
- Visit Google AI Studio
- Click "Create API Key"
- Copy your API key (starts with
AIza...)
Add this to your Claude Desktop configuration file:
📍 Config File Locations:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"prompt-enhancer": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://prompt-enhancer-mcp.sebrproduction.workers.dev",
"--header",
"X-Gemini-API-Key: ${GEMINI_API_KEY}",
"--header",
"X-Gemini-Model: ${GEMINI_MODEL}"
],
"env": {
"GEMINI_API_KEY": "your-api-key-here",
"GEMINI_MODEL": "gemini-2.5-flash-preview-05-20"
}
}
}
}
🔑 Replace your-api-key-here with your actual Gemini API key!
Step 3: Restart Claude Desktop
Close and reopen Claude Desktop to load the new MCP server.
Step 4: Test It!
In Claude, try asking:
"Use the prompt enhancer to improve this request: Create a todo app"
Claude will enhance your prompt and then use the enhanced version to give you a much more detailed and professional response!
🔧 Configuration Options
Option 1: With Your Own API Key (Recommended)
Pros:
- ✅ Your own Google AI quota and billing
- ✅ No usage limits from shared server
- ✅ Full control over API costs
- ✅ Better privacy and security
{
"mcpServers": {
"prompt-enhancer": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://prompt-enhancer-mcp.sebrproduction.workers.dev",
"--header",
"X-Gemini-API-Key: ${GEMINI_API_KEY}",
"--header",
"X-Gemini-Model: ${GEMINI_MODEL}"
],
"env": {
"GEMINI_API_KEY": "AIzaSyB79j7v1BfmPybZ5oM1he3XzQ2nqf_8rCw",
"GEMINI_MODEL": "gemini-2.5-flash-preview-05-20"
}
}
}
}
Option 2: Using Shared Server API Key
Pros:
- ✅ No API key setup required
- ✅ Instant setup
Cons:
- ⚠️ Shared quota limits
- ⚠️ May have usage restrictions
{
"mcpServers": {
"prompt-enhancer": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://prompt-enhancer-mcp.sebrproduction.workers.dev"
]
}
}
}
🤖 Available Models
You can choose different Gemini models by changing the GEMINI_MODEL value:
| Model | Description | Speed | Quality |
|---|
gemini-2.5-flash-preview-05-20 | Latest, fastest (Default) | ⚡⚡⚡ | ⭐⭐⭐⭐ |
gemini-1.5-pro | More detailed, slower | ⚡⚡ | ⭐⭐⭐⭐⭐ |
gemini-1.5-flash | Balanced option | ⚡⚡⚡ | ⭐⭐⭐ |
📖 How to Use
Basic Usage in Claude
- Simple Enhancement:
"Enhance this prompt: Build a calculator app"
- Specific Enhancement:
"Use the prompt enhancer to improve this request for a React component: Create a button"
- Multiple Enhancements:
"Enhance these prompts: 1) Create a navbar 2) Build a user profile page"
What Gets Enhanced
The AI enhances prompts by adding:
- 📋 Technical Specifications - Frameworks, languages, architecture
- 🎨 Design Requirements - UI/UX considerations, accessibility
- 🔒 Security Guidelines - Best practices, validation requirements
- 📊 Performance Criteria - Optimization and scalability needs
- 🧪 Testing Requirements - Unit tests, integration tests
- 📚 Documentation Needs - Comments, README, API docs
🧪 Testing Your Setup
Method 1: Check Server Status
curl https://prompt-enhancer-mcp.sebrproduction.workers.dev/health
Expected Response:
{
"status": "healthy",
"configuration": {
"apiKeyConfigured": true,
"apiKeySource": "client-config"
}
}
Method 2: Test Enhancement Directly
curl -X POST https://prompt-enhancer-mcp.sebrproduction.workers.dev/enhance \
-H "Content-Type: application/json" \
-H "X-Gemini-API-Key: your-api-key" \
-d '{"prompt": "Create a login form"}'
Method 3: MCP Inspector
npx @modelcontextprotocol/inspector@latest
Then connect to: https://prompt-enhancer-mcp.sebrproduction.workers.dev
Gemini API Pricing (Pay-per-use)
- Gemini 2.5 Flash: ~$0.00001 per request (very affordable)
- Free Tier: Generous limits for development and testing
- Billing: Appears on your Google Cloud account
Typical Usage Costs
- Light use (10 enhancements/day): ~$0.01/month
- Heavy use (100 enhancements/day): ~$0.10/month
- Enterprise use (1000 enhancements/day): ~$1.00/month
Costs are estimates and may vary based on prompt length and model used.
🚨 Troubleshooting
Solution: Make sure you've:
- Added your API key to the
env section - Restarted Claude Desktop after config changes
- Used the correct header format in
args
Problem: "Invalid API key"
Solution:
- Verify your API key at Google AI Studio
- Make sure it starts with
AIza - Check for extra spaces or characters
Problem: "Model not found"
Solution:
- Try using
gemini-1.5-flash instead of the preview model - Check available models in Google AI Studio
Problem: Claude doesn't see the MCP server
Solutions:
- Restart Claude Desktop (most common fix)
- Check that your config file is valid JSON
- Make sure the file path is correct for your OS
- Try the simple configuration first (without API key)
Problem: "Request failed" or timeout errors
Solutions:
- Check your internet connection
- Verify the server is running: https://prompt-enhancer-mcp.sebrproduction.workers.dev/health
- Try a different Gemini model
- Check your API quota at Google AI Studio
🔧 Advanced Configuration
You can add additional headers if needed:
{
"args": [
"-y",
"mcp-remote",
"https://prompt-enhancer-mcp.sebrproduction.workers.dev",
"--header", "X-Gemini-API-Key: ${GEMINI_API_KEY}",
"--header", "X-Gemini-Model: ${GEMINI_MODEL}",
"--header", "X-Custom-Header: ${CUSTOM_VALUE}"
],
"env": {
"GEMINI_API_KEY": "your-key",
"GEMINI_MODEL": "gemini-2.5-flash-preview-05-20",
"CUSTOM_VALUE": "your-value"
}
}
Multiple Model Configurations
You can set up multiple instances with different models:
{
"mcpServers": {
"prompt-enhancer-fast": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://prompt-enhancer-mcp.sebrproduction.workers.dev", "--header", "X-Gemini-Model: gemini-2.5-flash-preview-05-20"],
"env": {"GEMINI_API_KEY": "your-key"}
},
"prompt-enhancer-detailed": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://prompt-enhancer-mcp.sebrproduction.workers.dev", "--header", "X-Gemini-Model: gemini-1.5-pro"],
"env": {"GEMINI_API_KEY": "your-key"}
}
}
}
🌟 Use Cases
For Developers
- Component Creation: Turn simple component requests into detailed specifications
- API Design: Transform basic API ideas into comprehensive endpoint documentation
- Architecture Planning: Enhance system design prompts with best practices
For Content Creators
- Blog Posts: Expand topic ideas into detailed outlines
- Documentation: Transform basic docs into comprehensive guides
- Marketing Copy: Enhance brief concepts into detailed marketing strategies
For Researchers
- Literature Reviews: Expand research topics into structured investigation plans
- Data Analysis: Transform basic questions into detailed analytical frameworks
- Report Writing: Enhance outline ideas into comprehensive report structures
🔒 Privacy & Security
- API Keys: Your API keys are sent directly to Google's servers, not stored by our service
- Prompts: When using your own API key, your prompts are only processed by Google's Gemini API
- Logs: The server logs basic request metadata but not prompt content
- Data: No user data is permanently stored or shared
🚀 Server Architecture
This MCP server runs on Cloudflare Workers with:
- ⚡ Global Edge Network - Sub-100ms response times worldwide
- 🔄 Auto-scaling - Handles traffic spikes automatically
- 🆓 Zero Cost - Free tier supports 100,000 requests/day
- 🔒 Built-in Security - HTTPS, DDoS protection, CORS support
- 📊 Real-time Monitoring - Health checks and performance metrics
🤝 Contributing
Found a bug or want to contribute?
- Issues: Report bugs or request features
- Pull Requests: Submit improvements
- Discussions: Share use cases and tips
📄 License
MIT License - Use this freely for personal and commercial projects.
🙋♂️ Support
Need help? Here's how to get support:
- Check this README - Most questions are answered here
- Test the server - Use the health check endpoint
- Try the simple config - Start without API keys to test basic functionality
- Check GitHub Issues - See if others have had similar problems
- Create an issue - If you're still stuck, create a detailed issue report
🎉 Ready to enhance your prompts? Follow the Quick Start guide above and start getting better AI responses today!