Python Sandbox

The Python Sandbox is ideal for custom scripts, Jupyter notebooks, or minimal agent frameworks that use the official OpenAI Python SDK.

By simply changing the base_url in the SDK, all your Python AI scripts benefit from Cencori's unified logging, model routing, and spend caps.

Quick Start

1. Deploy from the Marketplace

Navigate to Agents → Agent Marketplace in your project dashboard and click Deploy Agent on the Python Sandbox card.

2. Generate an API Key

Go to the agent's Configuration tab and click Generate Key.

3. Install the OpenAI SDK

pip install openai

4. Initialize the Client

Initialize the standard OpenAI client, but point it to Cencori:

from openai import OpenAI
 
client = OpenAI(
    # 1. Point the client to the Cencori Gateway
    base_url="https://cencori.com/api/v1",
    
    # 2. Use your Cencori Agent Key
    api_key="cake_YOUR_KEY_HERE",
)
 
response = client.chat.completions.create(
    # 3. Model selection happens in the Cencori Dashboard, 
    # but the SDK still requires this field string
    model="gpt-4o",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Write a python script to calculate fibonacci."},
    ]
)
 
print(response.choices[0].message.content)

Using Other SDKs

You can use the exact same approach for the Node.js/TypeScript SDK:

import OpenAI from "openai";
 
const openai = new OpenAI({
  baseURL: "https://cencori.com/api/v1",
  apiKey: "cake_YOUR_KEY_HERE",
});

Because Cencori perfectly implements the OpenAI proxy standard, any SDK or library built for OpenAI works out of the box.