AI agent for dietitians
Supplement and Medication Interaction Screening Agent
Every client's combinations are checked and any risk is passed to a prescriber before the client continues.
What it does
Clients often take supplements their prescriber does not know about, and some combinations, such as St John's wort with certain medicines, are risky. The agent reads the client's list of medicines and supplements, then checks each pair against trusted references and records the source. It sorts findings into serious, moderate and minor, and drafts a message for the client's prescriber or pharmacist describing the combination and asking for review. When the client changes their list or the prescriber responds, it rechecks. It never tells the client to stop or start anything. The professional approves every message. Edge case: a serious interaction is flagged to the professional at once for a same-day call to the prescriber.
How it works
Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.
Read the steps as a list
- New or changed client list
- Read the medicines and supplements
- Check each pair against the references
- Rank interactions by severity and note the source
- Draft a message to the prescriber or pharmacist
- Professional approves the message and what is told to the clientThe agent waits here for your OK.
- Send the message and log it
- Has the prescriber responded and does the list now match their advice?If not: Send a reminder, ask the professional to call and keep the item open. Back to step 3.
- Recheck the list after any change
- Interaction report
How it decides
It checks every pair against trusted references and ranks risk by severity.
- Serious interactions go to the professional at once
- Every finding needs a source
- Never advise stopping a medicine
- Recheck after any change
Make it yours
Every agent is a starting point. You choose these settings for your own situation.
- Reference sources
- Severity levels
- Who receives alerts
- Recheck timing
- Message format
What keeps you in control
It always asks you first
- Professional approves messages to prescribers
- Professional approves what is told to clients
Hard limits
- Never tell a client to stop or start a medicine
- Never rely on one source for a serious finding
- Never send without approval
It stops when
- Done: all findings are reviewed by a prescriber and recorded
- Stop: the reference is unavailable
Set it up
We guide you through the set-up, step by step
Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.
- One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
- The agent then walks you through connecting your own data, one source at a time
- A downloadable copy with the flow chart, the rules and the full guide