This is currently strictly in early beta and should not be expected to be particularly useful.
See schema.sql for an idea of the layout of the database. It runs upon sqlite and has several million pages indexed so far including all of Wikipedia. It is searchable through an inverse index.
You may notice many more results not US or anglosphere based than with Google and similar engines. Part of the goal is to be less US-centric and more internationally focused.
See RANKING.md for information on the ranking algorithm used. You can download the entire index here.
This project is under the Mozilla Public License 2.0. See LICENSE for more information.