Frequently Asked Questions

GitHub Copilot API Gateway is a VS Code extension that exposes your authenticated GitHub Copilot session as a local HTTP API server. This allows you to use your Copilot subscription with any tool that supports OpenAI-compatible APIs—like LangChain, Cursor, Aider, and more—without paying for separate API keys.

Yes! The extension itself is completely free and open source. You just need an active GitHub Copilot subscription (Individual at $10/mo, Business, or Enterprise). If you're a student with the GitHub Student Developer Pack, you likely have free access to Copilot.

This extension uses your existing GitHub Copilot subscription. If your plan includes GPT-4 (which most do), then yes—you can access GPT-4 programmatically without paying extra for OpenAI API credits. However, you still need a valid Copilot subscription.

The available models depend on your Copilot subscription, but typically include:

• OpenAI: GPT-4o, GPT-4, GPT-3.5 Turbo
• Anthropic: Claude 3.5 Sonnet, Claude 3 Opus
• Google: Gemini Pro, Gemini Flash
• Meta: Llama 3.1, Llama 3.2

Use the /v1/models endpoint to see all available models for your subscription.

Yes. The extension runs entirely locally on your machine. No data is sent to any third-party server other than GitHub itself (which is where Copilot is hosted). You can enforce:

• IP Allowlisting to restrict access
• API Key authentication
• Full audit logging for compliance
• Rate limiting to prevent abuse

This extension uses GitHub Copilot's official VS Code extension API to authenticate and make requests. It doesn't extract or store tokens, doesn't make unauthorized requests, and operates within the normal usage patterns. However, you should review GitHub's current terms for your subscription type and use responsibly.

By default, no. The server binds to 127.0.0.1 (localhost only). To enable LAN access:

1. Change server.host to 0.0.0.0
2. Configure your firewall to allow the port
3. Use IP Allowlisting to restrict to trusted devices
4. Optionally enable API Key authentication

You don't need to extract or hunt for a static API key! This extension handles the secure authentication handshake with GitHub automatically through your VS Code session. Your local tools just need to point to http://127.0.0.1:3030 and use any placeholder API key.

This extension is a local LLM server, but powered by GitHub's cloud. It's a perfect alternative to running heavy local models if your hardware is limited, offering cloud-quality performance (GPT-4o, Claude 3.5) with local-like control and privacy.

Make sure you have:

• VS Code version 1.95 or higher
• GitHub Copilot extension installed
• GitHub Copilot Chat extension installed
• You're signed in with a valid Copilot subscription

Try restarting VS Code after installing the prerequisites.

Common issues:

• Copilot not authenticated: Ensure you're signed into GitHub Copilot in VS Code
• Wrong port: Check that your client is using the same port as configured (default 3030)
• API Key mismatch: If you've set server.apiKey, your client must send the correct Bearer token
• IP blocked: If you have an IP allowlist, make sure your client's IP is included

Check the VS Code Output panel (select "GitHub Copilot API Gateway") for detailed logs.

The extension itself doesn't impose hard rate limits beyond what you configure. However, GitHub Copilot has fair-use policies. If you're making extremely high-volume requests, you may experience throttling from GitHub's side. For typical development use cases, you shouldn't hit any limits.

This extension is designed for development, prototyping, and personal projects. For production deployments, consider:

• Official API services with SLAs
• Your organization's compliance requirements
• GitHub Copilot's terms of service

That said, many developers use it for internal tools, dev/staging environments, and private projects without issues.