Shraga

Shraga is a modern AI application with a backend based on FastAPI and a frontend built with React.

Installation and Setup

Requirements

1. Python 3.11 or higher
2. Pip
3. Poetry (for dependency management)
4. Node.js and pnpm (for frontend)

Backend Setup

First, install Poetry if you don't have it already:

pipx install poetry

Clone the repository and install dependencies:

git clone <repository-url>
cd shraga
poetry install --no-root
poetry run pre-commit install

To activate the virtual environment:

poetry shell
which python  # Verify the correct Python interpreter is being used

Development Tools

Several command-line tools are included with this package to help with development and management tasks:

User Management

# Create a user with basic authentication (interactive mode)
poetry run create-basic-auth-user

# Create a user with command-line arguments
poetry run create-basic-auth-user <email> <password> <config_file>

# Example
poetry run create-basic-auth-user user@example.com mypassword123 config.demo.yaml

This tool will:

• Prompt you to enter a user email address (with validation)
• Securely collect and validate a password
• Ask for the configuration file to update
• Generate a bcrypt hash of the password
• Save the credentials to a text file for reference
• Update the specified configuration file with the new user

Running the Application

Backend

To run the backend server with hot-reloading enabled:

SHRAGA_FLOWS_PATH=flows CONFIG_PATH=config.demo.yaml uvicorn main:app --reload

Frontend

To run the frontend development server:

cd frontend
pnpm install  # Only needed first time or when dependencies change
pnpm run dev

By default, the frontend will be available at http://localhost:5000 and will connect to the backend API running on http://localhost:8000.

Configuration

Shraga uses YAML configuration files to manage its settings. The repository includes:

• config.demo.yaml: A template configuration file with documentation

You can specify which configuration file to use with the CONFIG_PATH environment variable.

Demo Flow

To run the demo flow without requiring an LLM, Elasticsearch, or Opensearch:

1. Use the demo configuration file:

   export CONFIG_PATH=config.demo.yaml

2. Run the backend:

   export SHRAGA_FLOWS_PATH=flows
   uvicorn main:app --reload

Project Structure

• /frontend: React web interface
• /flows: Flow definitions

Code Style

Pre-commit hooks enforce code style:

poetry run pre-commit install

License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.