A school IEP helper must avoid unauthorized medical diagnosis wording. Which prompt rule fits?

Forbidding medical diagnosis language outside an approved glossary is the right prompt rule for school IEP helpers that must not practice medicine. That constraint works because teachers and case managers need consistent, policy-approved vocabulary for supports and observations, not freestyle clinical labels that imply unauthorized diagnoses. Binding Claude to the glossary reduces compliance risk, parent confusion, and documentation that overstates what school staff are licensed to assert.

Limiting medical diagnosis language to terms cited in published clinical journals for IEP documentation fails because 'cited in a journal' is not the same gate as 'approved for this district's IEP glossary'—plenty of accurate clinical terms are still diagnostic labels school staff aren't licensed to apply. Flagging suspected diagnosis language for educator review after draft completion in the helper workflow fails because it allows unauthorized diagnostic wording into the draft first and relies on educator review instead of blocking it at the prompt with the approved glossary. Restricting vocabulary to plain language and common terms only for staff understanding fails because plain-language framing doesn't stop an informal diagnosis from being stated—a helper can say 'it sounds like he has ADHD' in perfectly plain words; the rule that actually prevents unauthorized diagnosis is the glossary boundary, not sentence simplicity.

Exam caveat: glossary limits do not replace human clinical judgment when a true evaluation is warranted; the helper should escalate, not diagnose. Operational check: maintain a district-approved IEP glossary, reject or rewrite outputs containing out-of-glossary diagnostic terms, train staff on escalation paths, and audit a sample of helper transcripts each term for unauthorized medical wording.