SkeinRank is an open-source toolkit for explainable terminology normalization and attribute enrichment for technical documents, incidents, and search indexes.

It helps teams map internal jargon, aliases, and domain-specific terms to canonical technical attributes that can be used for search, filtering, reranking, analytics, and governance.

• Normalizes aliases such as k8s, kube, and kuber into canonical entities like kubernetes.
• Extracts typed attributes from technical text, such as TOOL, DB, COMPONENT, ERROR, VERSION, and ENVIRONMENT.
• Groups canonical values by slots, so search systems can distinguish tools, databases, components, errors, versions, and environments.
• Produces explainable passport/debug output showing what was matched, accepted, filtered, or normalized.
• Enriches JSONL datasets and Elasticsearch indexes with compact search-friendly metadata.
• Supports custom terminology profiles, profile validation, and snapshot-based runtime configuration.
• Provides a governance foundation for managing profiles, canonical terms, aliases, and snapshots.

• skeinrank — main monorepo with core, server, Elasticsearch provider, and governance package.

SkeinRank is currently focused on:

• terminology normalization
• technical search enrichment
• Elasticsearch enrichment
• profile validation
• governance workflows
• snapshot-based runtime architecture

SkeinRank is in early active development.