frostbite-offline-gardener

🌿 FrostBite: Offline Open-Weight Garden Scout

Hacktoberfest 2026 — Week 1: Touch Grass Challenge
A lightweight, zero-cloud CLI tool that pairs local frost benchmark data with an on-device open SLM (smollm2:1.7b) to give you a 5-step physical outdoor gardening card in under 3 seconds.

---

📖 Overview

Most generative AI tools encourage endless conversational screen time. FrostBite flips that script:

• Runs 100% offline on your machine — no internet required.
• Accepts your hardiness zone or region and looks up historical frost benchmark dates locally.
• Prompts a locally running lightweight SLM with strict constraints to output concise, actionable chores.
• Renders an ASCII index-card format directly in your terminal and saves garden_todo.txt so you can fold the slip, pocket a pencil, close your laptop, and head outdoors.

---

🛠️ Tech Stack

• Language: Python 3.10+
• Inference Engine: Ollama (local loopback at http://localhost:11434)
• Model: smollm2:1.7b (Hugging Face / Ollama) — ultra-compact 1.7B parameter open-weight model designed for fast, local CPU/RAM inference
• Dependencies: requests (for local REST communication)

---

🚀 Quickstart

1. Prerequisites

Install Ollama and ensure it is running locally.

2. Pull the Open-Weight Model

Run the following in your terminal to fetch the lightweight model (~1 GB):

```bash ollama run smollm2:1.7b