My Scripts

This repository contains small Python utilities for interacting with AI models. All scripts are located in the scripts/ directory.

Setup

Create the conda environment and install the required Python packages:

conda env create -f environment.yml
conda activate myscripts

Available Scripts

• claude_call.py – Query Anthropic's Claude model and save results to a DOCX and HTML file.
• groq_simple_chatbot.py – Simple terminal chatbot using the Groq API.
• read_mp3_summary.py – Transcribe an MP3 file, summarize it, and generate expert opinions using the Groq API.
• testing_local_model.py – Evaluate local models with Ollama and save the outputs.

See scripts/README.md for more details on each script.

LLM Experiments

The llm-experiments/ directory contains more advanced AI experiments:

• mitre-attack/ – Extract MITRE ATT&CK TTPs from cybersecurity scenarios
• iso-vector-db/ – RAG system with FAISS vector database
• security-awareness/ – Generate security awareness training content

See llm-experiments/README.md for details.

Monitoring App

The monitoring-app/ directory contains a system monitoring demo with FastAPI backend and React frontend. See its README for setup instructions.

Related

• Lyttek AI Journey – Documentation of our AI automation journey