See it working
Try the demo
The live app runs on preparedness data. Ask it anything, see which source backed the answer, and switch between collections. That is the same interface you get for your own data.
Open the demoCurated preparedness search
Ask a preparedness question. Get an answer with the source attached.
The hosted demo searches a fixed collection instead of the open web, shows where each answer came from, and keeps the path open to run the same experience on your own data later.
Example search
We lost power after a storm. What should we do first?
Start with safety, water, refrigeration, lighting, and backup communication. Then check the strongest nearby guidance in your selected collection.
Why it is different
Not just a chatbot
Every answer comes from the collection you can inspect. If an answer looks wrong, you can read the source yourself instead of trusting a model to be right.
See how trust worksWhy it matters
Built for offline use
A preparedness tool that only works with internet access is not enough. The same engine can run locally, on a Raspberry Pi, or fully offline with your own models and your own collection.
See deployment optionsBuilt for more than one dataset
The preparedness collection is the example, not the limit. The same engine has been used for building codes, flu preparedness protocols, and humanitarian reference archives. If you have a website, PDFs, transcripts, or offline archives that people need to search, this is built to adapt to that. It is designed for curated knowledge sites and offline archives, and works with Kiwix ZIM archives as a straightforward way to bootstrap a local collection.
How it opens up
Start with the demo if you want to feel the product. Go to About if you want the trust model. Explore Collections if you want to see what is searchable today. Use the technical docs if you want the platform model, ingestion path, and GitHub handoff.
Start with a useful, trustworthy search experience, then follow the same pattern into your own data.