The purpose of this project is to facilitate interaction with different AI agents. These agents can perform various tasks, automate workflows, and aid in the development process.
There are zero built in system prompts.
Step-by-step instructions on how to install and set up the project.
- Clone the repository:
$ git clone https://github.com/michaelirey/hatch.git
- Navigate to the project directory:
$ cd hatch
- Install the required packages:
$ pip install -r requirements.txt
- Make sure openai api key is set
$ export OPENAI_API_KEY=...
Instructions and examples on how to use the project.
$ python app.py
The project allows you to interact with various AI agents. For example, in app.py
, a HelloWorldAgent
is used to look up domain information:
See: https://platform.openai.com/docs/assistants/tools/function-calling
The thread_summary.py
script provides a summary of a specific thread. To use it, run the following command:
$ python thread_summary.py <thread_id>
This script uses the ThreadAnalyzer
and ReportPrinter
classes to analyze and print the thread summary.
You can import new assistants using the import_assistant.py
script. This script retrieves the assistant configuration from OpenAI and generates the corresponding Python class.
To import an assistant, run:
$ python import_assistant.py <assistant_id>
This will save the assistant configuration and generate a new Python class in the assistants
directory.
Instructions on how to run tests.
$ pytest
Information about the project's license. MIT License