See OntoGraph through real decisions
Use cases are the human-facing layer of the OntoGraph documentation. Each one starts with a recognisable situation, explains the outcome in plain language, and links to the unchanged executable reference queries that prove the bounded behaviour.
Choose a use case when you want to know why the capability matters. Choose the executed proof-case library when you want to see human questions answered and evidence-checked by the runtime. Choose the reference-query catalogue when you want the exact canonical input, request, expected output, and verified runtime result.
Available now
| User goal | Benefit | Support |
|---|---|---|
| Review a change without making it live | Separate accepted reality from review-ready and promotion-ready possibilities | Supported preview |
The generated proof for that use case executes three human questions and checks the returned current-state and governance evidence before the page can be published.
How support labels work
- Supported preview — executable in the current Developer Preview for the exact scope described by the page.
- Partial — meaningful executable proof exists, but an intended product surface or workflow remains unavailable.
- Planned — the user need is documented, but current executable product proof is not yet claimed.
Support applies only to the scope stated on each page. It is not model certification, production-backend equivalence, or permission to approve or promote canonical state.
Good next additions
The section is designed to grow around user questions such as:
- Where did this claim come from, and what evidence supports it?
- What changed between two accepted versions?
- What is blocking this proposal from moving forward?
- Can we import this external model without treating it as accepted truth?
- What meaning might be lost if we move to another storage backend?
Each future page must add evidence metadata and pass the same documentation drift gate before it can be published.