A photo of a prescription in. A plain-language explanation out, with the disclaimer front and centre.
Patients leave a consultation holding a piece of paper they can't read. This recipe turns that photo into something they can. TinyVision reads the prescription image directly and pulls the medication names, dosages, frequencies, and durations. AI then explains them in plain language and adds general education per medication: what it's for, how it's usually taken, common precautions and side effects. It only restates what was extracted. It never diagnoses, never changes a dose, and never contradicts the prescription. Every email carries a prominent not-medical-advice disclaimer, and an unreadable photo is sent back for a retake rather than guessed at.
Free · no credit card · you connect your own apps
Installing provisions each one, wired together and ready to run.
Where a patient uploads a photo of their prescription. Publishing it auto-creates the Prescription Summaries response sheet, which the workflow extends with request_id, summary, review_flag, and status.
Fires on each upload. TinyVision extracts the medications, dosages, frequencies, and durations; AI turns that into a plain-language summary plus general medication education, or flags the upload as unclear; the result is saved to the row and emailed to the patient.
Delivers the plain-language summary and per-medication guidance with the boxed disclaimer, or asks for a clearer photo when the upload couldn't be read confidently.
You connect your own accounts. Swap any app that offers an alternative at install.
Email the plain-language summary or a request for fuller detail to the patient.
The patient submits the Prescription Upload form with their name, email, and a photo of their prescription. Publishing the form auto-creates its response sheet, so every submission and the photo itself are captured with no extra setup, which also gives your healthcare staff a searchable record to review.
A TinyVision node reads the uploaded photo directly and extracts the medication names, dosages, frequencies, and durations exactly as written. The patient never retypes anything, which matters most for the patients least able to read the handwriting in the first place.
A TinyGPT step turns the extracted details into a plain-language summary (medications, dosage schedule, duration, instructions) and adds general educational information for each medication: purpose, general usage, common precautions, common side effects, and correct-use tips. The boundary is strict and deliberate: it restates only what was extracted, and never diagnoses, changes a dose, or contradicts the prescription. If the photo couldn't be read confidently, it flags the upload rather than guessing.
The summary and a review flag are written straight back onto the same row in Prescription Summaries, with a status of summarized or needs-more-details. Healthcare staff can review the upload and the AI summary side by side, and search across every past request from one place.
When the photo was read clearly, the patient receives the plain-language summary and per-medication guidance with a prominent boxed disclaimer: this is educational information only, not medical advice; always follow your doctor; consult your pharmacist or prescribing doctor before changing anything. The disclaimer text is yours to set at install. Swap Gmail for Brevo.
If the photo can't be read confidently enough to identify a medication, dosage, and frequency, the patient is automatically asked to resubmit a clearer one. A guessed summary is worse than no summary, so the recipe never produces one.
Everything it creates is an ordinary form, table and workflow in your workspace. Edit any of it after install. Nothing is locked.
