Survival LLM
A free AI assistant and survival library (1.5 million offline articles) that runs on your own computer, with no internet. Shows the source of every answer. Mac and Windows.
Dated 7- and 15-day hazard outlooks for Bangladesh's 64 districts: the artifact behind every number, an open validation scorecard, and the limits stated.
OtherTypeScript · JavaScriptGitHub @myself-aasOnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
No updates detected yet. GitHub releases will appear here.
Owners of public repositories under a personal GitHub account can sign in to verify ownership and manage the introduction themselves. No administrator approval is needed.
Sign in, generate a verification file, and commit it to this repository’s default branch within 30 minutes. Files on other branches or forks do not count. Then check the file below to manage the project without administrator approval. Keep the file to retain your access.
If needed, you will go to sign-in. After signing in, return here and press your chosen verification button again.
Run /nomorevibe verify in your project folder to verify ownership and manage the introduction yourself.
Verify in one minuteAsk about other verification options
Multi Hazard Classification with severity indexing AI
Translation pending
Taken from the repository README
Loading the full README. The saved excerpt is shown below.
Multi-hazard early warning for Bangladesh agriculture. HazardNet publishes dated
7- and 15-day hazard outlooks for all 64 districts of Bangladesh — eight hazard classes,
a severity value and a confidence score per district and forecast date, in two honestly
labelled severity tracks.
This repository and the website publish results and outputs only:Read the full README ↗Multi-hazard early warning for Bangladesh agriculture. HazardNet publishes dated 7- and 15-day hazard outlooks for all 64 districts of Bangladesh — eight hazard classes, a severity value and a confidence score per district and forecast date, in two honestly labelled severity tracks.
Live site: https://www.hazardnet.live
This repository and the website publish results and outputs only:
The following are research-private and are deliberately not published here or on the
site (see docs/PUBLICATION_POLICY.md):
Trained model files in Models/ are retained unadvertised; they are not part of the
published surface, and nothing in this repository offers them for download.
| Path | Purpose |
|---|---|
frontend/ | The public web application (results surfaces, maps, blog, auth) |
api/ | Serverless API: stored forecast reads, alerts, chat assistant, weather widget |
backend/ | Self-host Express backend (same result-serving surfaces) |
Models/ | Trained artifacts (unadvertised — not part of the public surface) |
docs/ | Publication policy, ops runbooks, attribution |
scripts/ | Site build, result publishing and QA tooling only |
# 1. Install and run the frontend (Vite dev server on :3000)
cd frontend && npm install && npm run dev
# 2. (Optional) the self-host API beside it. The Vite dev server proxies
# /api, /metrics and /health to :3001, so the API must listen there.
cd .. && npm install && npm run dev:api
# One process, one port instead: `npm run build` then `npm start` — the backend
# serves the built frontend and the API together on :3000.
Build for production: npm run build at the repository root (or cd frontend && npm run build).
Source code: MIT. If you use HazardNet outputs in academic work or an
operational product, please cite the repository (CITATION.cff).
HazardNet is a Master's thesis project, Department of Agrometeorology, Bangladesh Agricultural University. It is not an official warning service: official warnings come from the Bangladesh Meteorological Department and the Flood Forecasting and Warning Centre. For emergencies, call 999.