Clinician-supervised drafting utility

CareScript Studio

Build a precise OpenEvidence prompt, paste the result back, then package the patient-facing education for your own review.

LOCAL • NO ACCOUNT
Keep it de-identified. Do not enter a patient name, date of birth, medical record number, address, phone number, or other identifying information. This file stores only the self-email address you choose, in this browser.
01 · Patient lens

Set the communication target

No identifiers
02 · Topic & focus

Choose one or combine carefully

0 selected

Granular focus

Choose a topic above to reveal focused subtopics.
03 · Clinical guardrails

Make the draft safer to review

Core rules locked
Red flags and escalationAsk for urgent symptoms and when to call the clinician, urgent care, or emergency services.
Always on
Separate patient copy from clinician notesPatient content first; references and uncertainties after a clear review divider.
Always on
Teach-back checksAdd 2–3 questions the patient can answer to confirm understanding.
Accessibility adaptationsOffer seated, balance-supported, low-vision, dexterity, and caregiver alternatives when relevant.
Medication comparison frameWhen medications are selected, distinguish indication, benefit, common/serious harms, monitoring, cost/access, and questions for the prescriber.
Experimental · Optional

Image directions for an LLM

Do not use generated clinical or exercise images with a patient without consulting the treating MD and checking every position, label, laterality cue, and safety detail.