snowleopard 0.1.1

snowleopard.ai client library

Snow Leopard SDK for Python

Python client library for Snow Leopard Playground APIs.

Installation

pip install snowleopard

Quick Start

from snowleopard import SnowLeopardPlaygroundClient

# Initialize the client (or AsyncSnowLeopardPlaygroundClient)
client = SnowLeopardPlaygroundClient(api_key="your-api-key")

# Query your data in natural language
response = client.retrieve(
    datafile_id="your-datafile-id",
    user_query="How many users signed up last month?"
)

Getting Started

1. Get your API key from https://auth.snowleopard.ai/account/api_keys

2. Upload your datafiles at https://try.snowleopard.ai

3. Set your API key via environment variable:

   export SNOWLEOPARD_API_KEY="your-api-key"
   

   Or pass it directly to the client:

   SnowLeopardPlaygroundClient(api_key="your-api-key")
   

Usage

Synchronous Client

from snowleopard import SnowLeopardPlaygroundClient

with SnowLeopardPlaygroundClient() as client:
    # Get data directly from a natural language query
    response = client.retrieve("datafile-id", "What's the total revenue?")
    print(response.data)
    
    # Stream natural language summary of live data
    for chunk in client.response("datafile-id", "Show me top 10 customers"):
        print(chunk)

Async Client

from snowleopard import AsyncSnowLeopardPlaygroundClient

async with AsyncSnowLeopardPlaygroundClient() as client:
    # Get complete results
    response = await client.retrieve("datafile-id", "What's the total revenue?")
    print(response.data)

    # Get streaming results
    async for chunk in client.response("datafile-id", "Show me top 10 customers"):
        print(chunk)

CLI

The SDK includes a command-line interface:

pip install snowleopard
snowy retrieve <datafile-id> "How many records are there?"
snowy response <datafile-id> "Summarize the data"

Contributing

For SDK developer docs and how to contribute, see CONTRIBUTING.md