Standalone web application to compare responses from multiple Hydra.ai LLMs to the same question in a single, simple interface.

Fork of: mcp-rubber-duck - Original MCP Rubber Duck project

• Parallel Queries: Send the same question to N Regolo.ai models simultaneously
• Side-by-Side Display: View all responses in cards for easy comparison
• Model Filtering: Select models by type (chat, completion, embedding)
• Rate Limiting: 30 requests/hour per IP to prevent abuse
• API Key Persistence: Your key is saved in localStorage, no need to re-enter
• Dual Deployment: Run locally with npm or containerized with Docker

• Node.js 18+
• Docker (optional, for containerized deployment)
• Regolo.ai API key (get one at https://regolo.ai)

This is the simplest way to run the app locally.

# Clone the repository (if you haven't already)
git clone <your-fork-url>
cd regolo-rubberduck

# Install dependencies
npm install

# Compile the CSS
npm run build:css

# Start the server
npm start

The server will be available at http://localhost:3000

For containerized deployment.

# Build and start the container
docker compose up --build

# Or run in background
docker compose up -d --build

The server will be available at http://localhost:3000

For development with hot-reloading (if you need to modify code).

# Install dependencies
npm install

# Compile the CSS
npm run build:css

# Start in development mode
npm run dev

1. Get API Key: Visit https://hydra.ai and register to get your API key
2. Open the app: Navigate to http://localhost:3000 in your browser
3. Enter API Key: Type your key in the top-right field and click "Salva"
4. Select Models: Choose which models to query (up to 10)
5. Ask Question: Type your question in the textarea
6. Compare Responses: Click "Invia" and view all responses side-by-side

Create a .env file in the root directory:

PORT=3000
RATE_LIMIT_MAX=30
RATE_LIMIT_WINDOW_MS=3600000

Fetches the list of available models from Hydra.ai.

Headers:

• Authorization: Bearer <your-api-key>

Response:

{
  "models": [
    { "id": "model-1", "name": "Model 1", "mode": "chat" },
    { "id": "model-2", "name": "Model 2", "mode": "completion" }
  ]
}

Sends parallel queries to multiple models.

Headers:

• Content-Type: application/json

Body:

{
  "apiKey": "your-api-key",
  "models": ["model-1", "model-2"],
  "messages": [
    { "role": "user", "content": "What is the capital of France?" }
  ],
  "maxTokens": 1024
}

Response:

{
  "results": [
    {
      "model": "model-1",
      "response": "The capital of France is Paris.",
      "tokens": { "prompt": 10, "completion": 8, "total": 18 },
      "duration_ms": 1234,
      "error": null
    },
    {
      "model": "model-2",
      "response": "",
      "tokens": { "prompt": 0, "completion": 0, "total": 0 },
      "duration_ms": 0,
      "error": "API key non valida"
    }
  ]
}

The application implements rate limiting to prevent abuse:

• Limit: 30 requests per hour per IP address
• Window: Resets every hour
• Response: HTTP 429 with Retry-After header when exceeded

• Verify Node.js 18+ is installed: node --version
• Ensure npm install completed without errors
• Check that port 3000 is not already in use

• Verify the key is correct and copied entirely from https://regolo.ai
• Make sure you clicked "Salva" after entering the key
• Try removing and re-entering the key

• Check internet connection
• Verify API key is valid (see above)
• Open browser console (F12) to see any errors

• Wait for the timer to expire (1 hour from first request)
• Or restart the server to reset the in-memory counter

MIT

• Original Project: mcp-rubber-duck by nesquikm