🤖 Copilot Gateway

Use GitHub Copilot with any OpenAI-compatible client

A local proxy server that exposes OpenAI-compatible API endpoints, forwarding requests to the GitHub Copilot Chat API. Use GitHub Copilot with any client that supports the OpenAI API — Open WebUI, Chatbox, BoltAI, Elephas, Cline, Aider, and more.

Getting Started · Configuration · Client Setup · API Reference

---

✨ Features

• 🔄 OpenAI-compatible API — drop-in replacement for api.openai.com
• 🌊 Streaming support — real-time SSE streaming, just like OpenAI
• 🔐 Automatic auth — GitHub OAuth Device Flow, no tokens to manage
• 🧠 40+ models — GPT-4o, Claude, Gemini, o3-mini, and more via Copilot
• ⚡ Fast & lightweight — async Python with FastAPI + uvicorn
• 🔌 Zero config — works out of the box, just run and go

Prerequisites

• Python 3.11+
• An active GitHub Copilot subscription

Note: The GitHub CLI (gh) is not required. The server uses its own OAuth Device Flow.

Getting Started

Installation

git clone https://github.com/trsdn/github_copilot_openai_api_wrapper.git
cd github_copilot_openai_api_wrapper

python3 -m venv .venv
source .venv/bin/activate
pip install -e .

First Run

copilot-wrapper

On first launch, the server will guide you through GitHub authentication:

1. A one-time code is displayed (e.g. ABCD-1234)
2. Your browser opens to https://github.com/login/device
3. Enter the code and authorize
4. Done! Your token is saved to ~/.config/copilot-wrapper/github_token

The server is now running at http://127.0.0.1:8080 🚀

Configuration

Configure via environment variables or a .env file:

cp .env.example .env

Variable	Default	Description
HOST	127.0.0.1	Server bind address
PORT	8080	Server port
LOG_LEVEL	info	Log level (debug, info, warning, error)
GITHUB_TOKEN	—	Optional: skip Device Flow by providing a token directly

Client Setup

Point your OpenAI-compatible client at the local server:

Setting	Value
API Base URL	http://127.0.0.1:8080/v1
API Key	anything (e.g. x) — not validated
Model	gpt-4o, claude-3.5-sonnet, o3-mini, etc.

💡 Tip: Some clients (like Elephas) automatically append /v1 to the base URL. In that case, use http://127.0.0.1:8080 without the /v1 suffix.

Tested Clients

Client	Status	Notes
Open WebUI	✅	Set base URL to http://127.0.0.1:8080/v1
Chatbox	✅	Works out of the box
Elephas	✅	Use http://127.0.0.1:8080 (no /v1)
BoltAI	✅	Works out of the box
Aider	✅	Set --openai-api-base
Cline	✅	Use OpenAI-compatible provider
curl	✅	See examples below

API Reference

POST /v1/chat/completions

OpenAI-compatible chat completions. Supports both streaming and non-streaming.

# Non-streaming
curl http://localhost:8080/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

# Streaming (SSE)
curl http://localhost:8080/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o",
    "messages": [{"role": "user", "content": "Hello!"}],
    "stream": true
  }'

Supported parameters: model, messages, temperature, top_p, max_tokens, stream, stop, n

GET /v1/models

List all available models from GitHub Copilot.

curl http://localhost:8080/v1/models

GET /health

Health check endpoint.

curl http://localhost:8080/health
# → {"status": "ok"}

Architecture

┌──────────────────┐     ┌─────────────────────┐     ┌──────────────────────────┐
│   Chat Client    │────▶│   Copilot Gateway    │────▶│  api.githubcopilot.com   │
│  (any OpenAI-    │◀────│  localhost:8080      │◀────│  GitHub Copilot API      │
│   compatible)    │     │  FastAPI + uvicorn   │     │                          │
└──────────────────┘     └─────────────────────┘     └──────────────────────────┘
                              │
                              ▼
                         OAuth Device Flow
                         (first run only)

Project Structure

src/
├── main.py            # FastAPI app, endpoints, startup
├── config.py          # Configuration (pydantic-settings)
├── auth.py            # GitHub OAuth Device Flow + Copilot token management
├── copilot_client.py  # Async HTTP client for Copilot API
├── models.py          # Pydantic request/response models
└── middleware.py       # CORS, logging, error handling

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

Disclaimer

This project is for educational and personal use. It relies on internal GitHub Copilot API endpoints that are not officially documented. Use at your own risk and ensure compliance with GitHub's Terms of Service.

License

MIT