{
  "id": 13218204,
  "title": "Built TrailQR Raksha that forces people to go out and check QR before paying",
  "url": "https://urgent.news/2026/10/09/built-trailqr-raksha-that-forces-people-to-go-out-and-check-qr-before",
  "topic": "tech",
  "section": "Tech",
  "published": "2026-10-09T21:27:17.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/apparely/built-trailqr-raksha-that-forces-people-to-go-out-and-check-qr-before-paying-25h3"
  },
  "original_language": "en",
  "account": "QR codes are ubiquitous in India, from small shops to larger establishments. However, scanning a QR code alone is not sufficient for verifying its safety. A 2D QR matrix is not easily understandable by humans, unlike hyperlinks. Users cannot inspect the payee address, domain, or pre-filled transaction amounts before proceeding with a payment. To address this issue, the team created TrailQR Raksha, an offline-capable QR security guard designed for street environments across West Bengal. TrailQR Raksha operates based on deterministic rules, ensuring that users can safely scan QR codes without compromising their private information. The project leverages Google Gemma 4, an open-source AI model, for explaining the scan results in both English and Bengali. Additionally, Snowflake CoCo is used for identifying scam patterns within neighborhoods and is published as an open-standard Agent Skill. By utilizing offline-capable components and keeping sensitive data on the user's device, TrailQR Raksha ensures that no sensitive information is transmitted outside the user's phone. The project encourages users to physically interact with QR codes, avoiding the reliance on remote servers for security verification. This approach empowers consumers, especially those in less technologically advanced areas, to make informed decisions before making payments. The team behind TrailQR Raksha consists of friends who are committed to deploying practical solutions. The AI involved, Google Gemma 4, provides explanations for each scan in both English and Bengali, enhancing accessibility for a wider audience. Snowflake CoCo's role in tracking neighborhood scam patterns adds an additional layer of security to the tool. It is important to note that the AI never decides whether a QR code is safe; instead, the predefined rules determine the outcome. The tool's primary goal is to educate users and ensure they understand the risks associated with scanning QR codes, rather than providing automated assurances. TrailQR Raksha is now live at qr-raksha.vercel.app, allowing users to test its functionality and make informed decisions about QR code safety.",
  "summary": "This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass What I Built Almost the entire India runs on QR codes. From small shops to big ones, all have got one, attached somewhere or other. But just scanning the QR code was never enough. For example: A 2D QR matrix is human-unreadable. Unlike a hyperlink, victims cannot inspect the payee address, domain, or…",
  "key_points": [],
  "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."
}