Your procedures. Not a generic safety course.

Answers cited back to your own material, with the instructor one tap away when the material does not cover it

On our go/no-go benchmark every answer retrieved the right passage, carried a valid citation and stayed inside the material, and every out-of-scope question routed the learner to the instructor rather than a guess — against 54.2% for handing the model the whole document.

Every procedure step and rubric criterion carries a reference back to the document it came from

A criterion with no source reference is refused before it can be saved. Not a warning — a rejection.

Competency is scored by rules against an approved rubric, not by a language model’s opinion

Forty-one automated checks, including a deliberate break to prove the checks can fail. The model records what it observed; the rubric decides the mark.

Instructors attach their own documents to a course, review them as a draft, and publish when ready

Fourteen checks driven through a real browser, including that a student cannot see a draft and that publishing opens the document itself.

See which learners are struggling and which concepts they are stuck on

Twenty-nine automated checks on the cohort report, plus five more driven through a real browser.

Each institution’s data is isolated at the database level, not by application code

Ten checks run as a real signed-in user, so the database’s own access rules are genuinely exercised rather than assumed.