Standards & data

Real data, not a PDF in disguise.

This is the page for whoever wants the detail: the quality manager, IT, the consultant. Trama doesn't generate a static sheet — it produces structured data in the standards the digital passport will require, and versions the schema so when the rules are final you regenerate, instead of redoing.

The standards we use

GS1 Digital Link

The URI behind the QR: the international standard that gives each garment a unique identity and connects the QR to its data. It's the same language as large retail.

JSON-LD

The passport in machine-readable form (gs1:WearableProduct). Search engines, marketplaces and authorities can read it without ambiguity.

ISO 3758 · Ginetex

Care symbols and normalized vocabularies: «wash at 30°» becomes standard data, not free text everyone writes their own way.

Versioned schema

When the textile delegated act is final, the schema updates and you regenerate the passports. Zero manual rework.

Registry-ready

The architecture is compatible with the EU DPP Central Registry, Europe's infrastructure going live in 2026.

Multilingual

The passport page shows in the visitor's language, without you translating every field by hand.

Anti-hallucination
Nothing enters the passport without your approval.

A passport is a document with legal weight: invented data isn't a detail, it's a risk. That's why every value the AI extracts carries two things — its source (the row, the page, the photo it came from) and a confidence level. The AI does the tedious work; you confirm the truth. It's the opposite of a black box.

AI extraction Verified

Composition

Silk 82% · Cotton 18%

source: row 14 · collection_fw25.xlsx

94
confidence

Origin

Made in Italy

source: spec sheet · p. 2

98
confidence

Care

30° · No tumble dry

source: label photo

88
confidence
Privacy and security, by design

No cookies, no IP stored

The public passport pages don't profile their visitors. Country comes from edge geo, the IP address is never stored.

Your data stays yours

Single-tenant in beta: each brand's data is isolated. We don't resell it, we don't use it to train third-party models.

Honest about words

We say «DPP-ready, aligned to GS1 and the ESPR draft», never «ESPR-compliant»: that's impossible until the act is final. We get you ready, without selling certainties that don't exist yet.

Want to go into the technical detail?

Send us a real spec sheet: we'll show you the JSON-LD and the passport it produces. No commitment.