Python SDK

The official CID222 Python SDK for integrating content safety into your Python applications.

Planned Features

• Async Support — Full async/await support with asyncio
• Type Hints — Complete type annotations for IDE support
• Streaming — Native generator-based streaming
• Auto-retry — Built-in retry logic with exponential backoff
• Pydantic Models — Request/response validation

API Preview

Here's a preview of what the SDK will look like:

from cid222 import CID222

client = CID222(api_key="your-api-key")

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

print(response.choices[0].message.content)

# Async streaming
async for chunk in client.chat.completions.create(
    model="gpt-4o",
    messages=[...],
    stream=True
):
    print(chunk.choices[0].delta.content, end="")

Current Alternative

While the SDK is in development, you can use the REST API directly with the requests library:

import json
import os
import requests

# /chat/completions always responds with Server-Sent Events
response = requests.post(
    'https://api.cid222.ai/chat/completions',
    headers={
        'Authorization': f'Bearer {os.environ["CID222_API_KEY"]}',
        'Content-Type': 'application/json',
    },
    json={
        'model': 'gpt-4o',
        'messages': [
            {'role': 'user', 'content': 'Hello, world!'}
        ]
    },
    stream=True,
)

answer = ''
for line in response.iter_lines():
    if not line:
        continue
    line = line.decode('utf-8')
    if not line.startswith('data: '):
        continue
    data = line[6:]
    if data == '[DONE]':
        break
    event = json.loads(data)
    if 'error' in event:  # input blocked (type "content_rejected")
        raise RuntimeError('Blocked: ' + event['error'])
    if event.get('type') == 'output_content_rejected':
        raise RuntimeError('Blocked: ' + str(event.get('reason', '')))
    if not event.get('type') and isinstance(event.get('content'), str):
        # Model text. The gateway buffers and filters the output, then
        # sends the full answer as one event (id "filtered-response",
        # finish_reason "stop").
        if event.get('finish_reason') == 'stop':
            answer = event['content']
        else:
            answer += event['content']

print(answer)

See the Integration Examples for more Python code samples.