A convenient wrapper for the OpenAI API.
Project description
Botwrap
A convenient wrapper for the OpenAI API, designed to provide additional functionalities and simplify integration into various applications. This package includes a core team of assistants with unique personas and tools to handle a variety of tasks.
Table of Contents
Project Overview
The Botwrap project aims to simplify the use of OpenAI's API by providing a structured approach to manage and utilize different AI assistants. Each assistant has specific capabilities and tools to perform designated tasks effectively. The project is organized into several modules, including assistants, threads, files, vector stores, and more.
Installation
-
Clone the repository:
git clone https://github.com/BizPrincess/botwrap.git cd botwrap
-
Create a virtual environment and activate it:
python -m venv venv source venv/bin/activate # On Windows use `venv\Scripts\activate`
-
Install the required packages:
pip install -r requirements.txt
-
Install the package:
pip install .
Usage
Example Usage
-
Set up your environment variables: Create a
.env
file in the root directory of the project and add your OpenAI API key:OPENAI_API_KEY=your-openai-api-key
-
Run the main application:
python -m openaiwrapper.main
This will initialize the package and perform various interactions with the assistants defined in the project.
Configuration
The project uses environment variables for configuration. Create a .env
file in the root directory and add the following:
OPENAI_API_KEY=your-openai-api-key
LOG_FILE=application.log
API_BASE_URL=https://api.openai.com/v1
ENVIRONMENT=development # or 'production' based on your setup
The configuration can be customized in the openaiwrapper/config.py
file, which defines different settings for development and production environments.
Profiles
The project includes several predefined profiles for the core team of assistants. Each profile specifies the assistant's name, description, model, instructions, and available tools.
Core Team Profiles
- Brie: People Operations Specialist and Humanities Expert
- Britt: Business Strategist and Computer Science Enthusiast
- Derek: Innovative CEO and Engineering Economist
- Max: Strategy Analyst and Critical Thinker
- Nate: Software Engineer and Gaming Enthusiast
- Riley: AI Expert, Prompt Engineer & Project Manager
Profiles are stored as JSON files in the openaiwrapper/profiles
directory. For example, the profile for Brie is stored in coreteam_brie.json
.
Example Non-Coreteam Profile
- Example Non-Coreteam Assistant: General-purpose assistant for various tasks.
Testing
To run the tests, you can use the unittest
framework included with Python. The tests are located in the tests
directory.
-
Run all tests:
python -m unittest discover -s tests
-
Check the test coverage (optional): You can use
coverage.py
to check the test coverage:pip install coverage coverage run -m unittest discover -s tests coverage report
Contributing
Contributions are welcome! Please follow these steps to contribute:
- Fork the repository
- Create a new branch:
git checkout -b feature/YourFeature
- Make your changes and commit them:
git commit -m "Add some feature"
- Push to the branch:
git push origin feature/YourFeature
- Open a pull request
License
This project is licensed under the MIT License. See the LICENSE file for details.
Additional Notes
- Core Team Profiles: The project includes predefined profiles for the core team of assistants. Each profile specifies the assistant's name, description, model, instructions, and available tools.
- Logging: The project uses Python's logging module to log important information and debug messages.
- Mocking for Tests: The tests use
unittest.mock
to mock external dependencies and ensure the functions are tested in isolation. - Task Orchestrator: The
orchestrator
module contains logic for orchestrating conversations with multiple assistants. This includes initializing assistants, managing conversation threads, distributing tasks, and aggregating responses.
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.