taxonomy-creation

Taxonomy-Creation

Claude Code plugin for generating taxonomy / lookup tables and loading them into a database.

Use it when you're building a CMS, an analytics warehouse, an eval pipeline, or any data project that needs reference data — countries, currencies, languages, US states, timezones — or a custom domain taxonomy generated from a natural-language spec (e.g. "types of LLM evals", product categories, content tags).

What it does

• Standard taxonomies — ISO 3166 countries, ISO 4217 currencies, ISO 639 languages, US states, Canadian provinces, IANA timezones, HTTP status codes, MIME types. Offline-first, cached.
• Custom taxonomies from spec — describe what you want, get back a structured CSV/JSON with stable codes.
• Hierarchical taxonomies — parent/child trees as adjacency list, materialized path, or nested set.
• Load into a database — Postgres, SQLite, MySQL/MariaDB. Idempotent reload modes (replace / upsert / append) and DDL generation from the taxonomy file.
• Export to other formats — CSV, TSV, JSON, NDJSON, SQL seed files, YAML.
• Validate before load — uniqueness, completeness, code format, FK integrity for hierarchies, length budgets.

Skills

• generate-standard — pull a ready-made standard taxonomy (countries, currencies, etc.)
• generate-from-spec — generate a custom taxonomy from a natural-language or structured spec
• hierarchical-taxonomies — build parent/child taxonomy trees with the schema flavour you want
• validate-taxonomy — pre-load checks (uniqueness, completeness, format, FK integrity)
• load-to-postgres — DDL + load with replace / upsert / append modes
• load-to-sqlite — same, for SQLite (with STRICT table support)
• load-to-mysql — same, for MySQL/MariaDB (with utf8mb4 defaults)
• export-formats — export to CSV, TSV, JSON, NDJSON, SQL seed file, YAML
• onboarding — first-run walkthrough with a worked example

Typical pipeline

generate-standard           validate-taxonomy        load-to-postgres
       OR              →                       →     OR  load-to-sqlite
generate-from-spec                                    OR  load-to-mysql
                                                      OR  export-formats

Where data lives

Generated taxonomies, cached standard datasets, and load logs live under:

${CLAUDE_USER_DATA:-${XDG_DATA_HOME:-$HOME/.local/share}/claude-plugins}/taxonomy-creation/
├── data/      # generated taxonomies, ready to load
├── cache/     # cached standard datasets (ISO lists etc.)
├── state/     # loader logs, last-load metadata
└── config.json   # optional overrides

Database credentials are never stored. Use environment variables (PGPASSWORD, MYSQL_PWD) or a .pgpass file.

Installation

# Always-available (user scope):
claude plugins install taxonomy-creation@danielrosehill

# Project-scoped (run from inside the project repo):
claude plugins install taxonomy-creation@danielrosehill --scope project

License

MIT