{
  "id": 12039127,
  "title": "GutTracker: A Food & Symptom Journal That Runs Fully Offline for My Sister",
  "url": "https://urgent.news/2026/10/05/guttracker-a-food-symptom-journal-that-runs-fully-offline-for-my",
  "topic": "tech",
  "section": "Tech",
  "published": "2026-10-05T00:41:42.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/diwakararya/guttracker-a-food-symptom-journal-that-runs-fully-offline-for-my-sister-b98"
  },
  "original_language": "en",
  "account": "GutTracker is an offline food and symptom journal created specifically for the author's sister who has been experiencing digestive issues for several weeks. The application allows users to input meals in a conversational manner and log their symptoms, which are then analyzed to identify patterns and correlations. The data processing happens locally on the user's machine, ensuring privacy and eliminating the need for cloud storage or AI services. The application runs on llama.cpp, a locally hosted open-weight model, and computes statistics using Java. It uses a LanguageGuard to filter out diagnostic or prescriptive language, ensuring that it only delivers factual insights without making any health recommendations. The demo version of GutTracker showcases the application's capabilities while adhering to the challenge's restrictions, running on Render's platform without a local model. The source code is available on GitHub, and the application is designed to be privacy-first, solely serving as a tool for personal observation and analysis of food and symptom patterns.",
  "summary": "This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend What I Built My sister has been having gut problems for the past few weeks. The usual drill: write down what you ate, write down how you felt, try to spot the pattern — and never quite manage to see it, because a notebook can't compute anything. So I built GutTracker , a food and symptom journal for her. She types a…",
  "key_points": [
    "GutTracker is an offline food and symptom journal for a sister with digestive issues.",
    "The application processes data locally on the user's machine, ensuring privacy.",
    "It uses llama.cpp and Java to analyze patterns without providing health recommendations."
  ],
  "editors_take": "This development gives individuals with digestive issues a private tool to track food and symptoms, allowing them to identify patterns without relying on cloud storage or AI services.",
  "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."
}