PharmaDB

Free evidence-aware food & pharmacognosy compendium

For fruits, vegetables, spices, herbs, mushrooms, meat, and dairy — structured for provenance, safety, and clinical review.

Wellness information only No diagnosis No prescribing Local-first ready

Search the database

Browse ingredient monographs, nutrient profiles, safety flags, and source-grade evidence.

API status: checking…

  • 34data tables in `db/`
  • 2.1M+foods
  • 27.2M+nutrient rows
Important: this is an informational wellness platform, not a diagnostic, prescribing, or emergency service.

Catalogue domains

Core collections mapped from the database folder and the imported PharmaDB build packs.

Fruits

Polyphenols, fibre, acids, micronutrients, glycaemic context.

Vegetables

Carotenoids, glucosinolates, nitrate content, mineral profile.

Spices

Volatiles, phenolics, pungency compounds, usage constraints.

Herbs

Traditional use, caution notes, and compound summary.

Mushrooms

Bioactives, culinary safety, species-level provenance.

Meat

Protein quality, iron, B12, preparation-related safety.

Dairy

Protein, calcium, fermentation, allergy and tolerance context.

Recipes

Generated meal ideas with source-linked ingredient composition.

Local AI agent workspaces

Designed as guarded, assistant-grade workspaces — informational, not autonomous medical advice.

Agent 1 — Clinical nutrition

Doctor + diet planner + nutrition researcher + anatomy/biology/pharmacology reviewer.

  • Population-aware meal guidance
  • Interaction and contraindication checks
  • Evidence grading and citations

Agent 2 — Natural diet planner

Focuses on natural food strategies, ingredient pairing, and structured wellness planning.

  • Goal-based dietary plans
  • Ingredient alternatives
  • Local-only inference option

Agent 3 — Master chef lab

Researches and creates new meals and recipes, with a scheduled enrichment loop.

  • Recipe generation every 12h
  • Ingredient compatibility scoring
  • Recipe database expansion

Safety & governance

Evidence, privacy, and clinical review are first-class. Dangerous or uncertain outputs must be blocked or downgraded.

EvidenceSource snapshots + claim grades
SafetyAllergen, interaction, and population rules
PrivacyConsent and deletion workflows
OperationsIdempotent jobs, outbox, audit trail