Meet Soil Scout. The regenerative AI grown from real soil science.

Most AI tools are trained on whatever was on the open internet — which means the median answer about "how to fix soil" is dominated by synthetic NPK marketing copy, lawn-care chemical brands, and broad-strokes industrial farming. The default frame is conventional.

Soil Scout starts from a different library. Gabe Brown on cover cropping. Will Harris on integrating animals. The Rodale Institute on regenerative outcomes. Joel Salatin on pasture-based systems. The Weston A. Price Foundation on ancestral nutrition. Decades of soil-biology research that conventional ag has spent a generation ignoring.

The result: Soil Scout knows the difference between "USDA Organic" (which regulates inputs but not soil-mineral density) and "regenerative" (which rebuilds the soil itself). It recommends real-ingredient amendments — compost, cover crops, biological inoculants — before reaching for any product. And when a product makes sense, the recommendation is grounded in regen principles, not commission rates.

Every answer cites its sources. You can verify the reasoning. You can keep going deeper. You can disagree and push back — Soil Scout is built to think with you, not to perform for you.

Yes — Soil Scout is built on a large language model. We don't pretend otherwise. The difference is what we do with it: every answer is constrained by a curated regenerative knowledge base, instructed to cite its sources, and prevented from making recommendations outside that framework. When the vault doesn't cover something, Soil Scout says "I don't know" — not the generic web answer.

We update the underlying knowledge regularly as new research lands, new authorities publish, and new vendor partnerships pass the Good Soil Standard. We don't sell your conversations, and we don't train external models on what you ask. Your questions stay between you and the Scout.