I Built a VS Code Extension That Turns GitHub Copilot Into a Full OpenAI-Compatible API

Suhaib Bin Younis — Sun, 21 Dec 2025 08:22:49 +0000

The Problem

I was paying for too many AI services.

GitHub Copilot: $10/month
OpenAI API: $50-100/month
ChatGPT Plus: $20/month

One day I realized something obvious: Copilot uses GPT-4. I'm already paying for access to one of the best language models. I just couldn't use it outside of VS Code's autocomplete.

So I fixed that.

What I Built

GitHub Copilot API Gateway — a VS Code extension that exposes your Copilot subscription as a local HTTP server.

It implements three API formats:

OpenAI (/v1/chat/completions)
Anthropic (/v1/messages)
Google Gemini (/v1beta/models/:model:generateContent)

Any tool that works with these APIs now works with your Copilot subscription.

Why This Matters

For Individual Developers

You no longer need to:

Pay for OpenAI API credits to test LangChain
Set up Ollama and download 40GB models
Configure local inference with LM Studio
Manage API keys across different services

Just point your code at http://127.0.0.1:3030/v1 and it works.

For Teams and Enterprises

The math is simple:

Without This	With This
$10 Copilot + $50-200 API costs	$10 Copilot
Per developer, per month	Per developer, per month

For a team of 50, that's potentially $2,500-10,000/month in savings.

How It Works

Installation

# Install from VS Code Marketplace
ext install suhaibbinyounis.github-copilot-api-vscode

Or search "GitHub Copilot API Gateway" in VS Code.

Start the Server

Open the Copilot API sidebar
Click "Start Server"
Server runs at http://127.0.0.1:3030

Use It

from openai import OpenAI

client = OpenAI(
    base_url="http://127.0.0.1:3030/v1",
    api_key="not-needed"
)

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)

That's it. Your existing OpenAI code works unchanged.

Framework Integrations

LangChain

from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    base_url="http://127.0.0.1:3030/v1",
    api_key="not-needed"
)

response = llm.invoke("Explain recursion")

LlamaIndex

from llama_index.llms.openai import OpenAI

llm = OpenAI(
    api_base="http://127.0.0.1:3030/v1",
    api_key="not-needed"
)

AutoGPT / CrewAI / Any Agent

export OPENAI_API_BASE=http://127.0.0.1:3030/v1
export OPENAI_API_KEY=not-needed

# Now run your agent
python agent.py

Built-in AI Apps

The extension also includes purpose-built AI applications:

Playwright Test Generator — Describe tests in English, get complete projects
Code Review Assistant — AI feedback on your diffs
Commit Message Generator — Semantic commits from staged changes
Documentation Generator — Auto-docs for any codebase

These run directly in VS Code. No external tools needed.

Security Features

This isn't a toy. It includes production-ready security:

IP allowlisting
Bearer token authentication
Rate limiting
Request payload limits
Automatic data redaction in logs

Configure via VS Code settings:

{
  "githubCopilotApi.server.apiKey": "your-secret",
  "githubCopilotApi.server.ipAllowlist": ["192.168.1.0/24"],
  "githubCopilotApi.server.rateLimitPerMinute": 60
}

API Documentation

Full Swagger UI included at /docs:

Source Code

Everything is open source:

suhaibbinyounis / github-copilot-api-vscode

Unlock GitHub Copilot as a local API Gateway. Use Copilot with Cursor, LangChain, and any OpenAI-compatible tool.

GitHub Copilot API Gateway

Use GitHub Copilot, Gemini, Ollama & any VS Code AI — like any other API.
One extension. Zero API keys. Auto-discovers every model. Works with LangChain, Cursor, Aider, and 50+ tools

⚡ Quick Start

# 1. Install from VS Code Marketplace (search "GitHub Copilot API Gateway")
# 2. Click "Start Server" in the sidebar
# 3. Done. Your local API is ready.

Use it like OpenAI:

from openai import OpenAI

client = OpenAI(
    base_url="http://127.0.0.1:3030/v1",
    api_key="anything"  # No real API key needed
)

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Hello!"}]
)

🔒 Your data stays local. This extension runs entirely on your machine — no third-party servers, no telemetry, no data collection. The full source code is available on GitHub for you to…

View on GitHub

Try It

ext install suhaibbinyounis.github-copilot-api-vscode

Or visit the VS Code Marketplace.

If this is useful, consider starring the repo. Questions or feedback? Drop a comment below.

Introducing notes.suhaib.in — A Simple Space to Learn in Public

Suhaib Bin Younis — Sun, 20 Jul 2025 04:46:40 +0000

I built notes.suhaib.in as a kind of digital scratchpad — a place to log ideas, small experiments, debugging wins, and things I’m still wrapping my head around. It’s not a blog. It’s not polished. It’s just my stream of working thoughts as I build and learn.

The idea came from a simple frustration: we often wait too long to share what we’re learning because it’s not "done." But dev work is rarely ever done — we’re always evolving, pivoting, refactoring. So I figured... why not make the note-taking part public too?

The content you'll find there is raw — some of it’s half-baked, and some of it’s just breadcrumbs for future posts. I jot stuff like:

Quick tech notes (things I forget and have to Google again)
Setup guides I wish I’d had earlier
Snippets from AI/ML experiments that didn’t make it into full articles

Stuff I’m testing with tools like Bun, LangChain, Supabase, etc.
It’s built for speed. I wanted a frictionless way to publish without worrying about structure or SEO. Just write and post. That’s it.
My hope is that it helps other devs — even if it’s just one line that solves your issue or sparks an idea.

Check it out if you’re into that kind of vibe: notes.suhaib.in. And if you’ve got a similar space or want to build one, let me know — I’d love to see it.

DEV Community: Suhaib Bin Younis