EnrixaAI · Features

Marketplace-ready listings in minutes, not weeks.

Drop a brand name, upload a PDF catalog, or share existing spec sheets — EnrixaAI extracts, normalises and publishes complete product listings to Shopify, WooCommerce, Amazon, Noon and more.

Start free → 50 free listings Watch 90-second demo
Input · Whatever you have

Beauty of Joseon — Relief Sun rice probiotics

URL beautyofjoseon.com/products/relief-sun
PDF supplier-spec-EN-AR.pdf 2.3 MB
IMG 6 product photos mixed sizes
CSV existing-skus-noon.csv 847 rows
REG Saudi-FDA-clearance.pdf
EnrixaAI Pipeline
Output · Marketplace-ready PDP

Shopify draft · ready to publish

Beauty of Joseon
Relief Sun: Rice + Probiotics SPF 50+ PA++++
SAR 89.00
SPF 50+ Vegan EN+AR
SchemaBeauty
Variants50ml · 100ml
EAN (GS1)8809689861306 ✓
MarketplaceShopify · Noon · Amazon SA
Quality scorecard 9.1 / 10
100K+
Listings generated across tenants
10+
Marketplace integrations
16
Quality criteria scored per listing
CCAvenue Tashkeel SOUQROOT Maktabati Royal Canin PHAR/MEX
THREE WAYS IN

One pipeline. Three ways to feed it.

Whether you have a brand name, a folder of supplier PDFs, or a 60-page manufacturer catalog — EnrixaAI runs the same enrichment pipeline. The output is identical, the input is up to you.

Mode A · Discover

Give us the brand and product name. We search the open web.

When you're listing competitor or distributor catalogs.

Multi-backend search chain queries 20+ sources, ranks them, and synthesises a complete listing. Manufacturer pages first, then top retailers, review sites, and regulatory PDFs.

  • Tavily, Perplexity, Gemini, Anthropic Search, DeepSeek V4-Flash + Pro, Kimi K2
  • Cross-source merge picks the best signal per field
  • Manufacturer-first ranking with retailer fallback
  • Complete listing in 60–120 seconds
Output preview
// from "Daewoo DAF-AT3000" + "air fryer"
title: "Daewoo 7L Digital Air Fryer DAF-AT3000"
capacity: "7L"
power: "1700W"
programs: 12
sources: [mfr, currys, argos, …23]
elapsed: 87.2s
Mode B · Enrich

You have the data. We structure it.

When manufacturers send you spec sheets and photos.

Drop product photos, spec sheets, supplier PDFs and CSV exports. Vision LLM reads the marketing copy; OCR fallback handles scanned Arabic and English. Schema gaps fill from training-data knowledge — never invented.

  • Vision-first pipeline with EasyOCR fallback (AR + EN)
  • Multi-doc merge — keeps your richer values, fills the gaps
  • Same structured output as Discover, sourced from your data
  • Per-field source attribution for audit
Output preview
// from supplier-spec.pdf + 6 photos
title: "Royal Canin Persian Adult — 4kg"
analysis: { protein: 30%, fat: 30% }
images: 6 framed (1500×1500)
filled_from: supplier-spec.pdf
filled_from_kb: ["feeding_chart"]
Mode C · Bulk Import

One PDF catalog → fifty listings.

When you get a 60-page distributor catalog on Monday.

Drop a multi-SKU manufacturer catalog. PyMuPDF + EasyOCR enumerate every SKU, extract per-SKU images, and generate one listing per row. Sparse rows web-augment from Tavily; per-SKU retry on failures.

  • Auto-detects rows, pages and product blocks
  • Per-SKU image extraction with OCR fallback
  • Web-augment for sparse rows (Tavily)
  • Bulk gallery with per-SKU status chips
Output preview · 12 SKUs from 60-page PDF
✓
✓
✓
✓
✓
!
✓
✓
✓
✗
✓
✓
10 ready 1 review 1 failed 3m 42s
SMART FRAME · IMAGE STANDARDISATION

Six chaotic photos in. Six on-brand listings out.

Trustana-parity feature. Detect the product, fit it to canvas, apply margins, and batch process — all locally, free per image, manual trigger.

Before After

Detect → fit → margin → batch.

Trustana doesn't have a monopoly on white-background standardisation. Drop in 200 photos, get 200 PDP-ready images on a uniform 1500×1500 canvas — centred, padded, on-brand.

detect() → fit() → margin() → batch()
  • Customisable canvas — 1500×1500 default, any size you want
  • Margins — % of canvas, applied uniformly all four sides
  • Background — solid colour, match-original, or transparent PNG
  • Alignment — centre or bottom-anchored (great for shoes, bottles)
  • Output format — JPEG, PNG, WebP — your choice per build
What it doesn't do. Skips when the product would be enlarged beyond its source resolution — protects you from blurry upscales. Those land in the "review" chip and need a higher-res source.
QUALITY SCORECARD · 16 CRITERIA

Every listing graded before you publish.

Heuristic-only. Runs in <10ms per listing. Zero LLM cost. Per-category bonus criteria for pharmacy, medical, beauty and pet food.

What gets scored — and why.

16 base criteria across completeness, consistency, regulatory and presentation. Each schema adds 2–4 category-specific bonuses on top.

Pharmacy
AAFCO-equivalents, contraindication coherence, dosage cross-check, regulatory clearance match.
Medical Devices
CE Mark presence, EN/IEC standards consistent with class, sterilisation method declared, MDR-2017/745 fields.
Beauty
Full INCI list, allergen disclosure, claim softening (cosmetic-allowed wording), FDA-cosmetic vs. drug check.
Pet Food
Guaranteed analysis present, AAFCO life-stage match, region-appropriate formulation, feeding chart parity.
Generic
Default schema for everything else — base 16 with no domain bonuses.
Quality Report
build_id: bld_pet_2c84e1 · royal_canin_persian_adult_4kg
heuristic · 8ms
7.4
Amber · review recommended
Score 7.4 / 10 · 11 of 16 criteria full

2 partial, 3 missing. Pet-food bonus: AAFCO life-stage matched (+0.4). Region check: NA recipe on a Gulf-targeted listing — flagged.

CriterionScore
Title — descriptor + brand + variant1.0 / 1.0
Description ≥ 80 words, no marketing fluff1.0 / 1.0
Bullet features (3–7)1.0 / 1.0
Primary image & gallery (≥ 3)1.0 / 1.0
EAN / GS1 prefix valid for target market0.5 / 1.0
Brand & model1.0 / 1.0
Variants normalised (size, colour, pack)1.0 / 1.0
Categorisation 3-level taxonomy1.0 / 1.0
Pricing & currency vs target market0.0 / 1.0
Schema-required fields (per category)0.6 / 1.0
Marketing claim safety (no medical claims)1.0 / 1.0
Translation completeness (target locales)0.0 / 1.0
+ Pet food: guaranteed analysis+0.4
+ Pet food: AAFCO life-stage match+0.4
+ Pet food: region-appropriate formulation−0.2
CATEGORY-AWARE SCHEMAS

Schemas that understand their regulations.

Five schemas ship today. Each comes with consistency checks unique to its category — not generic rules patched onto a generic shape.

Pharmacy & Healthcare

OTC, supplements, medical OTC. Built for SFDA, MoPH, CDSCO submissions.

Example check Catches contradicting pregnancy stances across contraindications and warnings — e.g. "safe in pregnancy" in one section while listed as a Category C drug in another.

Medical Devices

Class I–III devices, MDR-2017/745. EN/IEC standard validation per class.

Example check Validates declared EN / IEC standards against the CE class — flags Class IIa devices missing EN 60601-1 or with mismatched IEC 62366 usability claims.

Beauty

Skincare, cosmetics, fragrance. INCI-aware with cosmetic claim guardrails.

Example check Softens cosmetic claims that cross into drug territory. "Reduces wrinkles" → "visibly improves the appearance of fine lines". Logged with the original for review.

Pet Food

Dry, wet, treats, supplements. AAFCO + life-stage + region-aware.

Example check Flags region-mismatched formulations — e.g. a North-America recipe attached to a Gulf-targeted listing where the AAFCO equivalent and ingredient sourcing differ.

Generic

Default schema for everything else — electronics, home, fashion, books.

Example check Base 16-criterion scorecard with no domain bonuses. Recommended starting point when you don't see your category yet — we'll suggest a domain schema once you have 20+ SKUs.
MULTI-MARKETPLACE PUBLISHING

One click. Every storefront you sell on.

Round-trip image upload, EAN harvesting, schema-mapped metafields. Publish to draft, edit in your storefront's admin.

Shopify
WooCommerce
Magento
Amazon
Noon
Flipkart
Zid
Salla
Whole Foods
private label

One-click publish to draft — your storefront's admin still owns the source of truth. EnrixaAI ships round-trip image hosting, EAN harvesting against GS1, and schema-mapped metafields so nothing is lost in translation.

REGIONAL INTELLIGENCE · MENA + INDIA

Built for the markets our competitors gloss over.

Trustana, Plytix, Akeneo are tuned for North America and Europe. EnrixaAI starts there — and goes further.

MENA SA · UAE · QA India EG ID

Local rules, region-aware checks.

  • Arabic translation with RTL preservation

    Opt-in per build. Source order, list markers, and bidirectional punctuation kept intact — no broken brackets or stranded numerals.

  • EAN / GS1 prefix validation

    Flags Latin-America EANs (770) on Gulf-targeted listings. Validates against the GS1 country prefix table per target market.

  • Currency vs target market

    SAR for Saudi, AED for UAE, INR for India. Catches USD on listings where the storefront default is local.

  • Country-specific regulatory clearance

    Saudi FDA, Qatar MoPH, India CDSCO. Pulls clearance numbers from your uploaded PDFs and matches them to the listed product.

  • Pet-food & pharmacy formulation alignment

    Region-appropriate recipes and dosages — flags an NA formulation marketed into Gulf or Indian channels.

BYO KEYS · INTEGRATIONS

Your keys. Your storage. Your bill.

No per-listing markup on AI. We're a tool — not a reseller of OpenAI tokens. Bring your own keys, or use our pooled platform plan.

Bring your own keys

Privacy and cost control, by default.

Per-tenant API keys mean your prompts and your data hit your AI provider directly — never our pooled account. Privacy guarantee + you keep your provider's volume discount.

Anthropic OpenAI Gemini DeepSeek Tavily Kimi Perplexity

On the Plus tier we run a super-admin platform pool — pay per credit, no key setup needed.

Built-in integrations

Storage, billing, storefronts.

Drop-in connectors for the boring infrastructure work — so you can spend the day on listings, not on wiring AWS S3 to your CMS.

  • Image hosting. Cloudflare R2 default. AWS S3, Backblaze B2, Wasabi and self-hosted MinIO all supported.
  • Billing & credits. Stripe with a per-tenant credit ledger. Refunds, retries and webhook reconciliation handled.
  • Storefront APIs. WordPress Application Passwords, Shopify Admin API, WooCommerce REST. OAuth for Magento and Salla.
  • Webhooks & events. Build-completed, scorecard-ready, publish-success — sub to whatever fires your downstream automation.
HONEST COMPARISON

EnrixaAI vs the alternatives.

Where we win, where we're at parity, and where the competition wins. We're not faking checkmarks.

Capability EnrixaAI Trustana DIY (sheets + writers)
Multi-mode pipeline (Discover + Enrich + Bulk PDF) ~
Category-aware consistency checks (5 schemas)
Bring-your-own API keys (no per-listing markup) N/A
16-criterion quality scorecard per listing
Per-SKU bulk PDF catalog extraction manual
Smart Frame image standardisation
Schema templates & field mapping
Ecosystem maturity & established customer base growing N/A

Trustana wins on ecosystem maturity. We're newer — and faster, more flexible, more transparent on cost.

PRICING SNAPSHOT

Try before you commit. 50 free listings.

Three plans. The Free tier is enough to evaluate Discover, Enrich and the scorecard on real catalogs.

Free
$0/mo

Run real builds on real catalogs. No credit card.

  • 50 listings / month
  • Discover & Enrich modes
  • Quality scorecard
  • Smart Frame (free per image)
  • 1 storefront connection
  • BYO API keys
Start free
Plus
$49/mo

For brands shipping ~500 listings/month. Platform pool included.

  • 500 listings / month
  • All 3 modes (Discover, Enrich, Bulk)
  • Platform AI pool — no own keys needed
  • 5 storefront connections
  • Bulk PDF import (50-page max)
  • Webhooks + Zapier
  • Email support
Start 14-day Plus trial
Pro
$249/mo

For agencies and brands with 5K+ SKU catalogs.

  • 5,000 listings / month
  • All Plus features
  • Unlimited storefronts
  • Unlimited bulk PDF size
  • Multi-tenant + sub-account roles
  • Priority support + Slack channel
Talk to sales

Full pricing breakdown, overage rates, and annual discount on the Pricing page →

Start free — 50 listings on us,
no credit card.

Run a real build on your real catalog. See the scorecard. Compare to your current process.

Start free → 50 listings
or book a demo if you have a 5K-SKU catalog.
Frequently Asked Questions

Questions we get the most.

Short answers below. Longer technical detail in the docs →

Three things. First, ChatGPT doesn't browse 20+ ranked sources or merge them; we do. Second, ChatGPT has no concept of a category-specific schema — it'll happily hallucinate INCI ingredients. Third, ChatGPT can't talk to Shopify Admin API, harvest GS1 EANs, or score the output. EnrixaAI is a pipeline; ChatGPT is one piece of it.
Manufacturer pages first, then top retailers, review sites and regulatory PDFs. In Discover Mode we run Tavily, Perplexity, and direct Anthropic / Gemini search. In Enrich Mode the source is whatever you upload — CSVs, PDFs, photos, supplier emails. Every field has source attribution so you can audit.
On the Free tier, yes — it keeps your data flowing through your own provider account. On Plus we offer a platform pool so you don't have to. Pro tenants can mix: a few critical accounts on BYO, the long tail on the pool.
Today: Shopify, WooCommerce, Magento, Amazon, Noon, Flipkart, Zid, Salla, and Whole Foods Market private-label. Each connector publishes to draft — your storefront's admin still owns the source of truth. See the full integration matrix →
Per-listing credits, billed monthly. Free is 50/mo, Plus is 500/mo at $49, Pro is 5K/mo at $249. No per-listing markup on AI — you pay for the orchestration, not the OpenAI tokens. Full breakdown on the pricing page.
No. Your inputs, outputs, scorecards and webhooks are tenant-isolated at the database row, image bucket prefix, and API-key level. We don't train on your data, full stop. Data Processing Addendum →