Cire — Cosmetic Ingredient Risk Engine

CIRE API — Cosmetic Ingredient Risk Engine

Deterministic INCI safety scoring API for AI agents.
Designed for shopping agents, catalog filters, compliance pipelines, and K-beauty recommendation engines.

---

What is CIRE?

CIRE scores cosmetic ingredient lists (INCI format) for safety risk across 5 categories:

Category	Description
irritation_risk	Skin irritation potential
allergen_risk	Known allergens
pregnancy_risk	Ingredients unsafe during pregnancy
acne_risk	Comedogenic ingredients
interaction_risk	Dangerous ingredient combinations

Fully deterministic. Rule-based. No ML. Always returns valid JSON.

---

Quick Start

curl -X POST https://web-production-9cdb4.up.railway.app/v1/analyze \
  -H "X-API-Key: YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"inci": "Water, Retinol, Glycolic Acid"}'

Response:

{
  "risk_score": 52,
  "risk_level": "moderate",
  "confidence_score": 65,
  "category_results": { ... },
  "matches": { ... }
}

• risk_score: 0–100 (higher = safer)
• risk_level: low / moderate / high
• confidence_score: evidence strength (0–100)

---

Authentication

Pass your API key in the X-API-Key header.

X-API-Key: cire-xxxxxxxxxxxx

Get your API key → cire.ai

---

Pricing

Plan	Calls	Price
Starter	1,000	$9
Growth	10,000	$49
Scale	100,000	$299

Tier Behavior

• Free tier API key: can call /v1/analyze, /v1/credits, billing/account endpoints
• Pro tier API key: includes Free endpoints + /v1/synergy and /v1/optimal
• Calling Pro endpoints with a Free key returns:
  • 403 Forbidden
  • clear message: Pro tier required for this endpoint...

---

API Reference

Full interactive docs: /docs

Developer entrypoint:

• Developer Quickstart

Use-case articles:

• AI Shopping Agent Safety API
• K-Beauty Compliance INCI API
• CIRE Pro: Synergy + Optimal

POST /v1/analyze

Analyze an INCI string. Costs 1 credit.

POST /v1/batch

Batch analyze up to 100 INCI strings per request. Costs 1 credit per item.

Request:

{ "inci_list": ["Water, Glycerin", "Water, Retinol, Glycolic Acid"] }

POST /v1/synergy (Pro tier)

Deterministic pair scoring for beneficial and risky combinations.

Request:

{ "inci": "Water, Adapalene, Benzoyl Peroxide" }

Response includes:

• synergy_score
• conflict_score
• evidence_level
• evidence_refs

POST /v1/optimal (Pro tier)

Goal-based deterministic recommendation endpoint.

Request:

{ "inci": "Water, Niacinamide", "goal": "acne" }

Response includes:

• recommended_additions
• avoid_with_current
• evidence_level
• evidence_refs

GET /v1/credits

Check remaining credits and key tier.

Billing Automation (Paddle)

• /v1/webhook handles transaction.completed events.
• When custom_data.credits is present, credits are added automatically.
• When custom_data.tier is pro/enterprise, API key tier is upgraded automatically.
• Pro subscription checkout package ids supported: pro_monthly, pro_yearly (requires Paddle price IDs in env).

---

Use Cases

• Shopping agents — filter products by pregnancy safety, allergen risk
• Catalog managers — bulk scan thousands of SKUs
• Compliance pipelines — EU/K-beauty export regulation checks
• Recommendation engines — rank products by safety profile

---

Tech

• Pure Python, zero ML dependencies
• Deterministic outputs — same input always returns same output
• Evidence-based: EU restriction list, FDA advisories, CIR reports, peer-reviewed consensus
• 13/13 smoke tests passing

Pro Data Enrichment (idempotent)

To enrich Pro formulation tables without duplicating rows:

cd cire-api
python scripts/enrich_pro_seed.py

This script appends missing records for:

• evidence_refs
• ingredient_pairs
• formulation_goals

---

Built for the AI agent ecosystem. CIRE Certified.