PTG_APIs

Official Python Client SDK for PTG Services:

Cognixx AI (cognix_ai / CognixxAI) with automatic free model fallback (model="free" → COGNIX/COGNIX-coder)

Cognixx Cloud (CognixxCloud)

PCB (PCBClient - Paras Box Cloud)

Infinity AI (InfinityAI)

PTGWeather (100% Free real-time live weather without API keys)

PTGTime (Real-time network time & timezone sync)

📦 Installation

Install directly via pip:

pip install PTG_APIs

Or install locally in editable mode for development:

pip install -e .

🚀 Quickstart Examples

1. Direct Import of Cognixx AI

Access the default free endpoints, model credentials, and run inference immediately:

from PTG_APIs import cognix_ai

Check free default credentials & endpoint

print("Free API Key :", cognix_ai.free_api) print("Free Model :", cognix_ai.free_model) print("Free Endpoint:", cognix_ai.free_url)

Run prompt (model="free" automatically routes to COGNIX/COGNIX-coder)

response = cognix_ai.generate("Write a quick Python script to calculate factorials", model="free") print(response)

2. Live Weather & Time (No API Key Required)

from PTG_APIs import PTGWeather, PTGTime

Real-time weather lookup

weather = PTGWeather.get_current("Haridwar") print("Temperature:", weather.get("temp_c"), "°C") print("Condition :", weather.get("condition")) print("Humidity :", weather.get("humidity"))

Real-time network time lookup

current_time = PTGTime.now("Asia/Kolkata") print("Live Time:", current_time.get("datetime"))

3. Paras Box Cloud (PCB)

from PTG_APIs import PCBClient

pcb = PCBClient()

Check free endpoint details

print("PCB Key :", pcb.free_api) print("PCB Endpoint:", pcb.free_url)

List remote files

files = pcb.list_files(folder_path="/") print(files)

4. Master Unified Client

Interact with all services through a single instance:

from PTG_APIs import PTG

ptg = PTG()

Query weather

print(ptg.weather.get_current("Delhi"))

Query time

print(ptg.time.now())

Query Cognixx AI

reply = ptg.cognix_ai.generate("Explain quantum computing in two sentences", model="free") print(reply)

🛠️ Publishing to PyPI

To build and publish this package:

1. Install build & upload tools

pip install --upgrade build twine

2. Build the wheel and source distribution

python -m build

3. Upload to PyPI

twine upload dist/*

📄 License

This project is licensed under the MIT License.