Scripture study that shows its work.
Every answer is grounded in text the tool actually retrieved — never in what a model remembers. Every claim is marked for what it is. Every citation is live: click it and the verse is there, in context, in the original language.
Three kinds of claim, and the difference matters
- The text asserts it
- It is on the page. Anyone reading the passage would see it.
- It was inferred
- It follows from the text, but the text does not say it. You can accept the premises and reject the conclusion.
- It is open
- The text underdetermines it. Reasonable readers land differently, and nothing in the passage settles it.
Most study tools blur these together, usually by accident — a smoothing phrase, a confident summary. That is how a reader ends up holding an interpretation and believing it is the text.
Sources ranked, not averaged
The Hebrew and Greek outrank the translations. The translations outrank the lexicons. Where they disagree, the answer says which one won and why — instead of quietly splitting the difference.
What it will not do
- Answer from memory when it cannot retrieve the text.
- Decide which tradition is right. It reports what each one rests on.
- Close a question the text leaves open.