Skip to main content

Python quickstart

Capture AI API calls from any Python application using the built-in MITM proxy—no code changes required.

1. Download and launch the desktop app​

Download the LupisLabs desktop app for macOS or Windows.

The MITM proxy starts automatically when you launch the app on port 9010.

2. Open a proxy-configured terminal​

In the LupisLabs desktop app, click "Open Terminal" (or use the keyboard shortcut). This opens a new terminal window with:

  • ✅ Proxy environment variables automatically set (HTTP_PROXY, HTTPS_PROXY)
  • ✅ SSL certificates automatically trusted (no manual installation needed)
  • ✅ Works with Python, Node.js, curl, and other HTTP clients

You'll see a confirmation message like this:

================================================
  Lupis Labs - Proxy Environment Configured
================================================

  HTTP_PROXY:  http://127.0.0.1:9010
  HTTPS_PROXY: http://127.0.0.1:9010
  CA Certificate: /path/to/ca.pem
  NODE_EXTRA_CA_CERTS: Set (Node.js)
  REQUESTS_CA_BUNDLE: Set (Python)
  SSL_CERT_FILE: Set (Python/curl)

  All requests (Node.js, Python) will use the proxy.

================================================

3. Run your Python application​

In the proxy-configured terminal, run your Python application normally—no code changes needed:

# Run your script
python app.py

# Or start your server
uvicorn main:app --reload

# Or run in a notebook
jupyter notebook

Any AI SDK calls (OpenAI, Anthropic, etc.) will be automatically captured:

# Your existing code works as-is
from openai import OpenAI

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

4. View captured requests in the app​

Switch to the Proxy Monitor tab in the LupisLabs desktop app. You'll see:

  • 📋 Complete request and response payloads
  • 💰 Token usage and cost analytics
  • ⏱️ Response times and status codes
  • 💾 Save requests for replay later