Description standards

ISAD(G)

General International Standard Archival Description

The Archiver's metadata fields map to ISAD(G)'s 26 elements of description across its seven areas: identity, context, content and structure, conditions of access, allied materials, notes, and description control. Hierarchical arrangement (fonds, series, file, item) is preserved throughout.

Maintained by the International Council on Archives (ICA).

EAD3

Encoded Archival Description, Version 3

The Archiver exports valid EAD3 XML finding aids with hierarchical <dsc> structure, controlled access terms, digital object links, and maintenance history recording AI provenance. EAD3 Version 1.1.1 is the current standard.

Maintained by the Society of American Archivists (SAA) in partnership with the Library of Congress.

See EAD3 export example →

Dublin Core

Dublin Core Metadata Element Set

Item-level Dublin Core XML exports include all 15 core elements plus Dublin Core Terms extensions for provenance, spatial coverage, and temporal range. Suitable for OAI-PMH harvesting and cross-domain interoperability.

Maintained by the Dublin Core Metadata Initiative (DCMI).

See Dublin Core export example →

Preservation and packaging standards

BagIt

RFC 8493 — File Packaging Format

BagIt packages include SHA-256 hash manifests for payload integrity verification, bag-info.txt with source organisation and provenance metadata, and a data directory containing original files alongside metadata sidecar files.

Specified by IETF RFC 8493. Widely adopted in digital preservation and library communities.

See BagIt export example →

PREMIS

Preservation Metadata: Implementation Strategies

Every AI action and archivist review is recorded as a PREMIS event with timestamps, agent identifiers, and outcome information. This creates a complete provenance chain from initial upload through AI processing to human review and export.

Maintained through a Library of Congress-sponsored maintenance activity and editorial committee.

See PREMIS provenance example →

ISAD(G) + Fonds Hierarchy

Multi-level Description

The Archiver supports full hierarchical arrangement: fonds, sub-fonds, series, sub-series, file, and item levels. Reference codes follow archival conventions and can be auto-generated, manually assigned, or imported from CSV.

Multi-level description principles defined by ISAD(G) Rule 2.4.

Works with your existing systems

The Archiver is designed to accelerate cataloguing and produce standards-ready outputs that plug into your existing system — not replace it.

AtoM

Access to Memory

Export CSV files formatted for direct import into AtoM, preserving hierarchical relationships, ISAD(G) fields, access conditions, and creator/provenance information. Supports AtoM's standard CSV import template.

See AtoM CSV example →

ArchivesSpace

Archives Information Management

Export CSV files mapped to ArchivesSpace's resource and archival object schema. Fields include component IDs, hierarchical levels, date expressions, extents, scope notes, and access restrictions.

See ArchivesSpace CSV example →

Archivematica

Digital Preservation System

Export BagIt packages with transfer metadata CSV files compatible with Archivematica's ingest workflow. File paths match the BagIt data directory structure and metadata columns use Dublin Core element names.

See Archivematica export example →

Know what the AI did and why

Every AI-generated field includes a confidence score and reasoning. Every archivist edit is logged. The full provenance chain is recorded in PREMIS events and available in your exports.

Per-field confidence AI reasoning Audit trail PREMIS provenance Human review required
How we use AI →