Summarises the most recent Use Cases run as a compact
matrix — the same questions, the same verdicts, no second pass over the model. Run your use cases,
then come back here.
What happens to your model. It is sent to this server and on to Google's Gemini API
(Vertex AI) to be checked, and it is not stored there — no database, nothing written to disk.
Server logs record sizes, timings and verdicts, never your model or your questions.
Your work — the model, the questions and the last run — is kept in this browser only, so a
refresh does not lose it. It never reaches the server as saved state.
A pre-prod check on one model. Load model… takes a Graphdraft, Neo4j, Neptune,
Gremlin, RDF/OWL, GraphQL or JSON-LD schema and converts it to the declared YAML above — editable, and
exactly what gets checked. Runs on the schema alone, no database needed. Loading needs no credentials;
validation calls Gemini on the server.
List your use cases — one per line, or Load questions… from a catalogue — and click Run use cases.
Run your use cases first — this view summarises that run.