Trust, privacy and safety

Clear boundaries build more trust than inflated claims.

Vitsmida is being prepared for real school evaluation, so the trust story has to be plain: what data and roles exist, what safety boundaries are visible, and what has not been approved yet.

Privacy posture

Signed-out learner progress stays local-first in the browser. Signed-in flows use the current Auth.js, Prisma and Postgres seams for account, progress and class data, but the platform should not be treated as a completed enterprise compliance product.

Learner safety

Learner contact and moderation surfaces are designed around bounded classroom and organisation contexts. Safety-sensitive actions should stay explicit, reviewable and recoverable.

Role visibility

Learners, teachers, organisation admins and platform admins see different surfaces. Current role-aware navigation and teacher analytics guards have evidence, but wider permission expansion still needs fresh validation before stronger claims.

Support boundaries

Help and support surfaces explain what users can do next. Vitsmida should not imply 24/7 operational support, production service levels or formal incident commitments before those decisions exist.

Evidence boundary

This page is intentionally conservative. It does not describe Vitsmida as production-ready, broadly validated, formally audited or approved for rollout.

Representative validation remains no-go / wait.
Pilot is not approved.
Production is not approved.
Broad rollout confidence is not supported.
VIT-126 remains accepted residual risk only, not completed representative final-gate testing.
Vitsmida has zero completed coded representative sessions unless later evidence proves otherwise.

Evaluation without hidden risk

Buyers and evaluators can review the learner entry path and help materials now, but those paths are not a substitute for representative validation. Any demo, school pilot or production rollout should keep the same boundary visible until new evidence and Jira decisions change it.