{
  "id": 13360091,
  "title": "TouchGrass AI: Local Gemma 2B Micro-Adventure Generator",
  "url": "https://urgent.news/2026/10/10/touchgrass-ai-local-gemma-2b-micro-adventure-generator",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-10T08:02:15.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/shadow16ua/touchgrass-ai-local-gemma-2b-micro-adventure-generator-4kbp"
  },
  "original_language": "en",
  "account": "TouchGrass AI is a local, offline-first web application that creates hyper-local micro-adventures in three simple steps. Users choose their current environment and available time, then the app generates a unique real-world observation quest. The tasks may include identifying architectural features, observing colors in nature, or completing walking challenges. The app locks the device, allowing users to engage with their surroundings without needing an internet connection. Developed using Google's open-weight Gemma2:2b model running locally via Ollama, the backend is built with .NET 10 Minimal API (C#), while the frontend uses vanilla HTML, CSS, and JavaScript for a responsive, nature-inspired dark UI. By enforcing a strict JSON schema and restricting fictional character creation, the app reliably produces contextual, real-world tasks. Open innovation was crucial for the project's success, as it enabled 100% offline capability, total privacy, and transparency to iterate on prompt constraints.",
  "summary": "This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass What I Built TouchGrass AI is a lightweight, offline-first web application that generates hyper-local, 3-step micro-adventures. The goal is simple: make the screen the shortest part of the experience. You select your current environment (e.g., a local park, city center, or quiet neighborhood) and the time you…",
  "key_points": [
    "TouchGrass AI is an offline-first web app for generating hyper-local micro-adventures.",
    "Users input environment and time to receive unique real-world observation quests.",
    "Developed with Gemma 2B model locally via Ollama, and .NET 10 Minimal API backend."
  ],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}