TutorOS 1.1 · public demo

Privacy and data handling

TutorOS is presented with synthetic learner data so the complete evidence workflow can be evaluated without exposing a real student.

Use synthetic data only

The preloaded Maya scenario is fictional. Do not enter names, contact details, school records, health information, or any other personal data about a real student or minor into this public demo.

No TutorOS persistence

This version has no account, database, analytics tracker, or application storage. Edits remain in the current browser session and reset when the demo is reloaded.

Optional OpenAI generation

The credential-free sample, mastery decisions, trajectory, benchmark, next-session brief, Honesty Gate, and tutor sign-off run without a model call. Without a server API key, spark-marked actions return a highlighted local mock and send nothing to OpenAI. When live generation is configured, the tutoring context entered for that action is sent from the TutorOS server to OpenAI to produce the requested plan or report draft.

Human review remains required

Generated drafts are not automatically sent anywhere. The tutor can edit the wording, must pass the Honesty Gate, and must sign off the current evidence packet before the parent update can be copied.

Deployment status

v1.1.0 is publicly deployed for review. Operators should verify/api/health, the credential-free workflow, and this notice before using the URL in a final submission.