This is a simple pet project to demonstrate how LangGraph works with a combinations of different tools.
First it takes user input (it can be a simple description of the task or a detailed instruction with the structure of the article).
Then it prepares a structure of the article and uses AI to generate a list of questions to search in the internet to gather information.
After that it takes search results and applies them to the structure of the article.
Finally, it generates a text of the article.
Run the following command to install the project in console:
- Clone the repository:
git clone [email protected]:sasha370/ai_writer.git
- Change to the project directory:
cd ai_writer
- Install package dependencies:
# install plugin for python version management (if you don't have it) curl -sSL https://install.python-poetry.org | python3 - # install dependencies poetry install
- Copy the
.env.example
file to.env
and fill in the missing values:cp .env.example .env
You can input the text in the console or use the task.txt
file.
Be aware that AI ignores most markup formatting so you don't need to clean up or prepare your text.
Run the following command to run the project in console:
poetry run python main.py