This project is a reverse proxy built with Express.js to securely forward requests to the OpenAI API. It enables interaction with OpenAI models, such as gpt-4o-mini, while securely handling your API keys using environment variables.

• Secure API Key Handling: The OpenAI API key is stored securely in an environment variable, keeping it hidden from the client-side.
• Request Forwarding: It forwards requests from clients to OpenAI's /v1/chat/completions endpoint.
• CORS Support: CORS is enabled to allow communication between different origins.
• Error Handling: Detailed error messages are logged for debugging, and proper HTTP response codes are returned to the client.

• Node.js (version 14.x or later)
• npm (or yarn)
• Docker (optional, if you want to run it in a container)

git clone https://github.com/your-username/openai-reverse-proxy.git
cd openai-reverse-proxy

npm install

Create a .env file in the project root and add your OpenAI API key:

touch .env

Then, add the following content to the .env file:

OPENAI_API_KEY=sk-your-openai-api-key

Start the server locally:

npm start

By default, the server runs on http://localhost:3000.

This route forwards requests to OpenAI's chat/completions API.

You can send a request to the reverse proxy like this:

curl -X POST http://localhost:3000/api/openai/v1/chat/completions \
    -H "Content-Type: application/json" \
    -d '{
        "model": "gpt-4o-mini",
        "messages": [{"role": "user", "content": "Say hello to the world."}],
        "max_tokens": 10
    }'

{
  "id": "chatcmpl-...",
  "object": "chat.completion",
  "created": 1634766363,
  "model": "gpt-4o-mini",
  "choices": [
    {
      "index": 0,
      "message": {
        "role": "assistant",
        "content": "Hello, world!"
      },
      "finish_reason": "stop"
    }
  ]
}

You can also test the basic root route by accessing:

http://localhost:3000/

You should receive:

Welcome to the OpenAI Reverse Proxy!

If you prefer to run the reverse proxy in a Docker container:

docker build -t openai-reverse-proxy .

docker run -d -p 3000:3000 --env-file .env openai-reverse-proxy

This will run the proxy on http://localhost:3000.

If there's an error in communicating with the OpenAI API, the proxy will return a 500 Internal Server Error with details in the response body.

{
  "error": {
    "message": "You must provide a model parameter.",
    "type": "invalid_request_error",
    "code": null
  }
}

Feel free to open an issue or submit a pull request for any improvements or bug fixes.

This project is licensed under the MIT License. See the LICENSE file for details.