Use case

Records your successors can still open.

Retention is a legal duty measured in decades; most PDFs are not built to survive one. Convert, prove, and seal the archive on your own infrastructure.

Repair-first PDF/A Independently validated Signed and timestamped No page meter

Most archives fail the day someone checks.

Between the retention mandate and the folder of PDFs sits a set of quiet failures that only surface years later, when nobody who caused them is around.

  • Ordinary PDFs rot. Fonts referenced but not embedded, device-dependent colour, encryption: files that render today and fail the decade test the standard exists for.
  • Declared is not conforming. Stamping PDF/A metadata onto a broken file produces an archive that fails its first independent validation; conversion has to repair, then prove.
  • Scans are pictures, not records. Without OCR, a digitized archive cannot be searched, cited, or audited at any useful speed.
  • Authenticity is an afterthought. A record that cannot prove who filed it and that it has not changed since is a claim, not evidence; that is what signatures, timestamps, and LTV exist for.
  • Per-page pricing fights the backlog. Metered cloud document services make the one project every records team must do, the full migration, the one they cannot afford.

Digitize, conform, prove, seal.

Four steps, two battle-tested engines, and an output an auditor can verify without trusting you.

01

Digitize

Scans and TIFFs become searchable PDFs; mixed batches split into records.

02

Conform

PdfAConverter repairs fonts, colour, and metadata, then rewrites as PDF/A-1b, 2b, or 3b.

03

Prove

PdfAValidator, a separate engine, re-parses the output and judges it rule by rule.

04

Seal

Signatures, RFC 3161 timestamps, and LTV make authenticity verifiable for decades.

One engine, two ways to deploy it.

The scenario decides the form: an archiving service every department calls, or archival built into your own records software.

Serve it

LM-Kit One: the archiving endpoint

OCR, PDF/A conversion, signing, timestamping, and LTV as REST calls any system can make: the scanner workflow, the DMS, the retention job. Authenticated, audited, and scaled beyond one machine when the backlog demands it.

Deploy LM-Kit One

Embed it

LM-Kit.NET: archival inside your product

Records software, DMS vendors, and line-of-business .NET applications ship conversion, validation, and signing as typed calls in their own process, offline included.

Embed LM-Kit.NET

What makes it defensible.

The properties an archivist and an auditor ask about first, stated plainly.

Two engines

The prover is not the producer

Conversion and validation share no verdict path, so a Compliant result means the record conforms, not that one tool agrees with itself. Benchmarked at 99.9% veraPDF-verified conformance.

Sealed

Authenticity that outlives certificates

Signatures with timestamps and LTV embed the material verification needs, so records verify long after the signing certificates expire.

Unmetered

The backlog is affordable

No per-page fees and no billing connection, unlike metered cloud document services; a million-page migration is a weekend of compute.

Both forms, working code.

Convert, prove, and seal: as REST calls to the server, or typed calls in your own process.

The archival chain as two endpoints on LM-Kit One.

terminal
# Convert to archival PDF/A (repair-first).
curl http://your-server:5189/lmkit/v1/pdf-to-pdfa \
  -H "Authorization: Bearer your-server-token" \
  -F "file=@record.pdf"

# Seal the conforming output.
curl http://your-server:5189/lmkit/v1/pdf-sign \
  -H "Authorization: Bearer your-server-token" \
  -F "file=@record.pdfa.pdf"

Frequently asked questions.

What makes this different from a cloud document service's containers?

Even on-premises containers from cloud document services typically require a metering connection to run, and fully offline tiers sit behind commitment plans. This pipeline runs on your hardware with no billing connection, no page meter, and no minimum volume.

How is conformance actually proven?

By a second, separate engine. PdfAValidator re-parses the converted file from scratch and judges it against the ISO 19005 rule catalog; it shares no verdict path with the converter. Across a 3,200+ file benchmark, 99.9% of converted documents also pass independent veraPDF validation.

Can it handle scans and mixed paper batches?

Yes. OCR gives scans a searchable text layer, multipage TIFFs convert straight to searchable PDF/A, and semantic splitting separates batch scans into individual records before conversion.

What keeps a sealed record verifiable in ten years?

Long-term validation. Signatures are timestamped through your TSA and the document security store embeds the certificates and revocation data verification needs, so a record proves its integrity long after the original signing certificates expire.

Private Document Intelligence

Archive the backlog without counting pages.