ayuai 1.0.0

A Python AI library integrating ChatGPT, Gemini & smart code suggestions.

ayuai

A Python AI library integrating ChatGPT, Google Gemini, and a smart code suggestion engine.

Installation (locally for testing)

1. Create and activate virtualenv:

   python -m venv venv
   source venv/bin/activate   # macOS / Linux
   venv\Scripts\activate    # Windows PowerShell
   

2. Install from local directory:

   pip install .
   

Usage

Chat (OpenAI / Gemini)

from ayuai import ChatAI

# For OpenAI:
ai = ChatAI(provider="openai", api_key="YOUR_OPENAI_API_KEY", model="gpt-4o-mini")
print(ai.ask("Hello AyuAI!"))

# For Gemini (use API KEY or HTTP key as required by Google)
ai2 = ChatAI(provider="gemini", api_key="YOUR_GOOGLE_API_KEY", model="gemini-pro")
print(ai2.ask("Hello from Gemini!"))

Code Suggestions

from ayuai import CodeSuggestions

cs = CodeSuggestions()
print(cs.suggest("print('Hello world')"))

Environment variables recommended

• AYUAI_OPENAI_KEY - Your OpenAI API key (if you don't pass it directly)
• AYUAI_GOOGLE_KEY - Your Google Generative AI key (for Gemini)

Publishing to PyPI (summary)

1. Make sure you have an account on https://pypi.org/.
2. Build distribution:

   python -m pip install --upgrade build twine
   python -m build
   

3. Upload:

   python -m twine upload dist/*
   

4. After a successful upload, users can pip install ayuai (if the name is available).

VS Code extension (simple demo)

A minimal VS Code extension scaffold is included under vscode-extension/ that runs a quick Python script using this package to demonstrate suggestions. See that folder for usage.

Security & Notes

• This project contains network calls to 3rd-party APIs; never store secrets in source control.

• Replace placeholder email and copyright where appropriate.

  Continuous Integration

This project includes GitHub Actions workflows:

• .github/workflows/ci.yml: runs tests on push and PRs to main.
• .github/workflows/publish.yml: builds and publishes to PyPI when you push a tag like v1.0.0.

Publishing notes

Create a PyPI API token and add it as a repository secret PYPI_API_TOKEN in GitHub to enable the publish workflow.

Tagging example:

git tag v1.0.0
git push origin v1.0.0

The publish workflow will run and upload the built distribution to PyPI using the token.