NutriPet Methodology

How ingredient grades and hazard alerts work

For eligible packaged-food labels, NutriPet calculates a versioned A-F ingredient-profile grade from recognized ingredient bands. It is a historical-policy heuristic, not a product-safety, nutritional-adequacy, or medical rating.

• Every listed ingredient must resolve to a supported catalog record; otherwise the product is Not rated instead of receiving a guessed letter.
• Declared percentages are used only when every assessed ingredient has one; otherwise label order supplies the deterministic weighting.
• Current hazard, recall, allergy, and sensitivity alerts remain separate, outrank the grade, and cannot be averaged away by other ingredients.
• AI features such as Smart Scan and Scout may help read labels, but they never create or override grade or hazard authority.
• The current ingredient-band snapshot is source-bound and AI-reviewed; it is not clinically or veterinary validated.
• None of this is veterinary advice - for medical or emergency questions, contact your veterinarian.

Evidence Standards

• Factual claims need a public page we can point to.
• When safety context matters, we lean on regulatory and veterinary sources.
• Public claims are limited to details that NutriPet can verify.

Unknown and Unverified Fields

If we cannot verify a detail, it stays labeled as unverified or not publicly disclosed as of 2026-02-26. That keeps the page honest and makes the gaps visible.

Update Cadence

• Public pages carry the date their facts were last verified; timestamps and access dates are refreshed whenever the underlying evidence changes.
• Citations, machine-readable facts, and the sitemap are re-checked on an ongoing basis rather than on a fixed public schedule.
• If a claim can no longer be verified, it is corrected or relabeled as unverified rather than left standing.