Code Previews

1. LLM Basics: here

2. Lang Basics: here

3. Prompt Engineering: here

4. Tool Agent: here

5. Output Parser: here

6. Lang Memory: here

Get Started

1. Install Miniconda (Optional)

Install Miniconda

To get started quickly, ensure you have Miniconda installed. Use the following command for your operating system to install Miniconda in one step:

For Linux or macOS:

curl -o Miniconda.sh https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh && bash Miniconda.sh -b -p $HOME/miniconda && rm Miniconda.sh && export PATH="$HOME/miniconda/bin:$PATH"

For Windows (PowerShell):

curl https://repo.anaconda.com/miniconda/Miniconda3-latest-Windows-x86_64.exe -o miniconda.exe
Start-Process -FilePath ".\miniconda.exe" -ArgumentList "/S" -Wait
del miniconda.exe

More information here.

Verify Installation:

Once installed, verify that conda is installed:

conda --version

Initialize Conda:

For Linux/macOS:

~/miniconda/bin/conda init

Restart your terminal to apply changes.

For Windows: Search for Anaconda Prompt in Windows search and open it.

2. Set Up the Python Environment

Using Conda (Recommended):

conda create -n marimo_env python=3.12 -y
conda activate marimo_env

Using Virtualenv:

python -m venv marimo_env
source marimo_env/bin/activate  # For Linux/macOS
harmful_detection_env\Scripts\activate     # For Windows

3. Install Dependencies

Install Git: Use the following guide to install Git: Git

Install Marimo:

pip install marimo

4. Start Marimo

Clone repository:

git clone https://github.com/SecureAIAutonomyLab/AgenticAISystems.git

Save OpenAI API Key:

• Create .env file

nano AgenticAISystems/.env

• Add the following line to the .env file:

OPENAI_API_KEY=<replace-with-your-key>

Start marimo server:

marimo edit --headless --no-token

Note: I highly recommend reading this guide about how to run marimo notebooks because they work a bit differently than regular notebooks.