Quick Start

Dependencies

• Git
• Git LFS (https://git-lfs.com/)
• (Optional) pyenv (https://github.com/pyenv/pyenv#getting-pyenv)
• Python >=3.9.16 (pyenv install)
• poetry (pip install poetry)
• Python packages (poetry install)
• Dependencies for custom tool curl_rss_tool (+ html2text + sumy)
  • html2text (sudo apt install html2text)
  • nltk data (poetry run python -m nltk.downloader all)
• Ollama
  • llama3 model, or other models used

Env

Create a .env.local file, then update with your API keys.

cp .env.example .env.local

Development

Dual CLI/API framework

This project is set up as a dual CLI/API framework, allowing you to use the same core functionality through both a command-line interface and a REST API. It utilizes FastAPI for the API layer and Click for the CLI layer, with shared functions in the src/core directory.

API Examples:

1. Start a ollama server

ollama serve

2. Start the API server in separate terminal

poetry run start --reload

3. Access the API endpoints using a web browser, curl, or a REST client:

curl http://127.0.0.1:3000/hello
curl http://127.0.0.1:3000/hello/Doge

CLI Examples:

1. Start the ollama server in separate terminal

ollama serve

2. Run script from shell

# poetry run hello
Hello World!

# poetry run hello Doge
Hello Doge!

Testing

poetry run test

Deployment

Serverless

The project is set up to use serverless on AWS using [default] profile. To install serverless, go to https://www.serverless.com/framework/docs/getting-started#installation

# Deploy changes
serverless deploy

# View deployed endpoints and resources
serverless info

# Invoke deployed functions
serverless invoke