The OEKO-TEX 100 audit is not won during the audit — it is won in the live file of every batch.

The problem

The OEKO-TEX file exists — just not before the audit.

In a spinning mill, dyehouse or garment plant working under OEKO-TEX Standard 100, the traceability of a certified SKU lives scattered: OEKO-TEX certificates from yarn and dye lot suppliers as PDFs in mismatched folders, laboratory test reports in the quality manager's inbox, technical data sheets for auxiliaries (softener, stain repellent, sizing) in a spreadsheet, dyehouse batch sheets in the MES, metal trim batches in the ERP. When the annual renewal or an update to the substance Annex comes around, someone puts the file together by hand the week before — and every error is a nonconformity.

The three realities almost nobody manages to cross-reference on the floor in time:

1. What goes into this SKU in this batch — which yarn, which dye lot, which auxiliaries, which trim. Spread across five systems.
2. What each supplier certifies — and from when until when it is valid. Spread across PDFs.
3. What the OEKO-TEX Annex requires today — and what changed since the last audit. Spread across the Institute's bulletins.

The classic system (chaining it by hand the week before) works 99% of the time. The remaining 1% is the nonconformity the auditor finds that could have been spotted six months earlier — and the delayed renewal, with the loss of market access that comes with it.

How it fits the IRIS system

iLEAN Writer does not replace the quality manager — it keeps the file permanently live for them.

The problem with the OEKO-TEX file is not a lack of knowledge: the quality manager knows exactly what to cross-reference. The problem is document volume and speed: keeping dozens or hundreds of certified SKUs up to date against suppliers, tests and bulletins. iLEAN acts as the putty between the five systems and the PDFs where traceability lives on islands, without asking you to change ERP, MES or laboratory.

Connect captures everything that comes in (certificates, tests, delivery notes, bulletins). Writer cross-references by SKU/batch and keeps the file live. The person signs every renewal — the AI does not.

The iLEAN pieces applied to the OEKO-TEX 100 file:

• iLEAN Writer (Agent) — a specialized agent from the iLEAN Agents family; it lives in Central and works 24/7. It maintains the SKU → batch → components → certificates → tests → Annex in force graph. When the audit arrives, the file already exists. When the Annex is updated, the agent identifies the affected SKUs and proposes the new tests to run.
• Connect — captures everything: suppliers' OEKO-TEX certificates (PDFs arriving by email), accredited laboratory test reports (PDFs), technical data sheets for dye lots and auxiliaries, dyehouse batch sheets from the MES, trim delivery notes, bulletins from the OEKO-TEX Institute. It all comes in and becomes cross-referenceable data.
• iLEAN Edge — optional, for when you need to capture a physical batch at the warehouse inbound with a QR or a supplier label the system had never seen. Local reading, asking nothing of the old MES.

See the full IRIS architecture →

Before and after

OEKO-TEX file prepared by hand vs. a live file with iLEAN Writer

Aspect	Manual pre-audit preparation	With iLEAN Writer
State of the file	Exists, chopped up across PDFs and spreadsheets	Live and unified in Central
Annual renewal	Weeks of office work	Generated on the spot, with signature
Annex update	Manual review, SKU by SKU	Agent identifies affected SKUs and proposes tests
Expired supplier certificate	Discovered during audit week	Early warning to the purchasing manager
Pending tests	Arranged with the laboratory at the last minute	Planned with room to spare, predictable budget
Traceability to SKU/batch	Rebuilt if the auditor asks for it	Permanent, with a photo of the original document

Impact estimate

Impact estimate for your plant — to be validated with your numbers.

The block below is an estimate to be validated with the specific data of your plant. We put it forward so the committee has an order of magnitude; we refine it during the diagnostic.

• Mid-sized textile plant with 30-200 SKUs certified to OEKO-TEX Standard 100, supplying fashion, home, sportswear or technical markets.
• Writer pilot (agent + Connect capturing suppliers and tests + human validation workflow). First value expected within a few weeks: the first OEKO-TEX file per SKU packaged on its own.
• Indicative payback between 4 and 9 months, depending on the number of certified SKUs, the labor hours currently spent preparing the annual audit and the documented frequency of Annex updates requiring additional tests.
• Reduction in annual audit preparation time in the order of ≥30% already in the pilot.
• The hard lever is twofold: recurring savings on pre-audit office work and a lower risk of nonconformity or delayed renewal, which hits the sales channel directly (customers who require the OEKO-TEX label in their specifications).

And the quality manager's reasonable doubt

“What if the agent makes up a test or links the wrong certificate?” — the agent only recontextualizes data that already exists. It extracts the limits from the PDF of the accredited laboratory's test report, associates the supplier certificate with the specific batch via the delivery note, adds up the dye lots per batch from the MES. It does not generate tests and it does not invent batches. Hallucination is a problem of free generation, not of anchored tasks: the best models brought the error below 1.5% on tasks anchored to the source ^[1]. And even so, the person signs every renewal — the AI does not.

^[1] OpenAI paper “Why Language Models Hallucinate”, 2025 — on the reliability of AI in tasks anchored to the source.