MikroScore API

MikroScore is a structured knowledge system. All data is available via machine-readable JSON endpoints so AI agents, chatbots, and other tools can work directly with our supplement reviews.

Quick Start

# All ingredients with evidence ratings

curl https://mikroscore.com/data/wirkstoffe.json

# All products with transparency scores

curl https://mikroscore.com/data/products.json

# Knowledge graph (nodes + edges)

curl https://mikroscore.com/data/graph.json

Endpoints

GET /data/wirkstoffe.json

All 119 ingredient dossiers with evidence level, EFSA status, safety rating, and linked studies.

Example response ↓

{
  "ingredient_id": "magnesium",
  "name": "Magnesium",
  "category": "metabolic",
  "evidenceLevel": 4,
  "efsaClaimsAllowed": true,
  "safetyRating": "hoch",
  "typicalDoseMg": 300,
  "aliases": ["Mg", "Magnesiumcitrat"],
  "keyStudies": [{
    "title": "The effect of magnesium supplementation on...",
    "pmid": "23853635",
    "year": 2013,
    "finding": "Significant blood pressure reduction..."
  }],
  "updatedAt": "2026-07-18"
}

Use for: Ingredient recommendations, evidence checks, dosage info, EFSA status

GET /data/products.json

All 293 product reviews with transparency score, price per day, verdict, and availability.

Example response ↓

{
  "product_id": "thorne-magnesium-bisglycinate",
  "name": "Thorne Magnesium Bisglycinate",
  "brand": "Thorne",
  "ingredientId": "magnesium",
  "pricePerDay": 0.44,
  "doseMg": 200,
  "form": "capsule",
  "transparencyScore": 8.4,
  "verdict": "empfehlenswert",
  "availableInDE": false,
  "url": "https://mikroscore.com/produkte/thorne-magnesium-bisglycinate"
}

Use for: Product comparisons, price analysis, availability checks, verdict-based recommendations

GET /data/graph.json (DE) /data/graph.en.json (EN)

The full knowledge graph: 928 entities, 1953 relations. Contains nodes (ingredients, mechanisms, symptoms, studies, regulatory) and edges (relationships with type and confidence).

Example response ↓

{
  "nodes": [{
    "id": "magnesium",
    "type": "ingredient",
    "label": "Magnesium",
    "evidenceLevel": 4,
    "efsaApproved": true,
    "updatedAt": "2026-07-18"
  }],
  "edges": [{
    "source": "magnesium",
    "target": "blutdruck",
    "relation": "wird_eingesetzt_fuer",
    "confidence": 0.9,
    "relationSource": "expert-review"
  }, {
    "source": "magnesium",
    "target": "studie-23853635",
    "relation": "basiert_auf_studie",
    "confidence": 0.9,
    "relationSource": "pmid:23853635",
    "evidenceLevel": "human_rct"
  }]
}

Use for: Knowledge graph exploration, mechanism reasoning, multi-hop queries

GET /data/claims.json

All 19 claims checks with verdicts (supported / misleading / unsupported / unclear).

Use for: Fact checking, marketing claim validation, content moderation

Data Model

Entity Types

Type	Count	Description
ingredient	119	Ingredients with evidence ratings
product	293	Products with transparency scores
mechanism	24	Mechanisms of action (e.g. NMDA antagonist)
symptom	51	Symptoms and conditions
study	438	Studies with PMID and metadata
regulatory	2	Regulatory classifications (EFSA)
contraindication	~15	Contraindications

Relation Types

Relation	Count	Meaning
wird_eingesetzt_fuer	556	Ingredient → Symptom/Indication
basiert_auf_studie	413	Ingredient → Study (with PMID)
enthaelt	363	Product → Ingredient
benoetigt_biomarker_check	144	Ingredient → Relevant biomarker
wirkt_ueber	143	Ingredient → Mechanism
hat_regulatorischen_status	122	Ingredient → EFSA status
hat_interaktion_mit	68	Ingredient ↔ Ingredient (interaction)
kontraindiziert_bei	68	Ingredient → Contraindication
hat_nebenwirkung	45	Ingredient → Side effect

Agent Examples

"Which magnesium product is highest rated?"

# 1. Fetch all products for magnesium
products = GET /data/products.json
magnesium_products = filter(products, ingredientId == "magnesium")

# 2. Sort by transparency score
best = sort(magnesium_products, by=transparencyScore, desc)[0]
# → Thorne Magnesium Bisglycinate (8.4)

"Are there studies linking ashwagandha to stress reduction?"

# 1. Fetch knowledge graph
graph = GET /data/graph.json

# 2. Find ashwagandha → wird_eingesetzt_fuer → stress
# 3. Find ashwagandha → basiert_auf_studie → studie-PMIDs
# 4. For each study: check study_type and evidence_quality

"Which supplements have EFSA-approved health claims?"

ingredients = GET /data/wirkstoffe.json
efsa_approved = filter(ingredients, efsaClaimsAllowed == true)
# → List of all EFSA-approved ingredients

Technical Notes

• All endpoints return application/json
• Caching: Cache-Control: public, max-age=3600 (1 hour)
• No authentication required — all data is public
• No rate limiting, but please use fairly (static CDN, no database)
• CORS: open (static files on Vercel CDN)
• Every entity has updatedAt for freshness checks
• Every edge has relationSource (provenance: pmid / efsa-register / expert-review / product-yaml)
• Study edges additionally have evidenceLevel: meta_analysis / human_rct / human_observational / expert_review
• Studies include study_type, evidence_quality, n, coi (where available)
• Source code: github.com/YonaPaproth/longevity-check

Note: The API data is provided for informational and structured processing purposes. It does not replace professional medical advice. All information without warranty. Questions about usage: Contact.