Documentation

Ontoshire does two things. It’s a registry: publish an ontology from a GitHub repo — pushing a git tag is what publishes a version — and it’s served at a pinned, citeable URL. And it’s a validation authority: pin that version with the Ontoshire client and validate your own instance data against its SHACL shapes, locally, before that data goes anywhere.

Just want to try it? Publishing and sharing an ontology is the core workflow — start with Quickstart: publish & discover. Want to validate data against a schema too? Quickstart: validate data takes you from install to a passing (and failing) validation in a few minutes.

Three things to know#

Publishing — tag a release and Ontoshire ingests your ontology. Add SHACL shapes and Ontoshire serves them as versioned, resolvable artifacts. This is how you define what “valid” means for your schema.
Discovery & reuse — this is the original offering: search and browse published ontologies, follow their dependency graphs, and reuse what you find. Every public version gets a citeable, resolvable URL — and once you’ve published, other ontologists can owl:imports yours the same way.
The validation client — a command-line tool and Python SDK that validates your instance data against a pinned owner/repo@version and its shapes. It runs in your environment; your data never leaves it. See data contracts for why validating against a published version is different from validating against a file on disk.

Visibility#

Public ontologies are browsable by anyone and free to pull. Private ontologies (a Pro feature) are visible only to their owner — Ontoshire never reveals that another user’s private repository exists. Programmatic pulls always require a token; human browsing does not.

A GitHub organisation can share private ontologies across its members on the Org plan — Orgs & seats covers how membership and per-seat billing work. See pricing for what each plan costs.