AI-powered demo app to help you use Weaviate, with a twist.
To install it, pip
install the package with:
pip install weaviate-agent-demo
Then, try out the different CLI commands set up in src/weaviate_agent_demo/cli.py
, or in pyproject.toml
.
e.g.:
- ask_llm
- ask_basic_ragbot
- ask_ragbot_with_reformulation
- ask_ragbot_with_tools
- safely_ask_ragbot_with_tools
- ask_weaviate_agent
-
Clone the repo:
git clone https://github.com/yourusername/weaviate-agent-demo.git cd weaviate-agent-demo
-
Install Poetry (if not already installed):
curl -sSL https://install.python-poetry.org | python3 -
-
Install the project dependencies with Poetry:
poetry install
-
Activate the virtual environment:
poetry shell
-
Run a Weaviate instance (e.g., with Docker):
-
Run the files in
src/weaviate_agent_demo/db_setup
to ingest data:python src/weaviate_agent_demo/db_setup/1_create_collection.py python src/weaviate_agent_demo/db_setup/2_import.py
-
You can now use the CLI commands, for example:
ask_llm
Then, when prompted - try queries like "How do I connect to Weaviate", or "How do I use filters with dates in hybrid search?"
There is also a Streamlit app available. Try it with:
streamlit run app.py