Intent triage
Patient question, appointment request, records request, payer notice, referral, vendor or spam — routed accordingly.
04 · Communication
AI Email Agent
A practice inbox is four inboxes wearing a trench coat: patients, payers, vendors and colleagues. The Email Agent separates them, drafts what it can, and files the rest where it belongs.
Agent at a glance
- StatusIn development
- CategoryCommunication
- Deployment2–4 weeks, shadow mode first
- Review modelHuman approval per action, configurable
- Headline metric70%+ Auto-triaged
- EvidenceFull action log, exportable any time
70%+
Inbox auto-triaged
Sorted, categorised and queued without a human reading it first.
< 1 hr
Patient reply draft
During business hours, waiting for staff approval.
0
Clinical advice sent
Anything clinical becomes a task, never an email reply.
What it does
Everything in this agent, in writing.
Reply drafting
Answers built from your policies, presented to staff as an editable draft with one-click send.
Records & forms routing
Release-of-information requests identified, logged and routed to the compliance queue with the clock started.
Payer notice handling
Denials, audit letters, credentialing requests and fee schedule changes escalated with the deadline extracted.
Referral intake
Referrals parsed, patient matched or created, and the appointment offer queued.
Attachment filing
PDFs and images classified and filed to the correct patient chart through the integration layer.
Deployment
Live in weeks, not quarters.
01
Inbox analysis
We categorise 90 days of mail to find what actually arrives and what it costs you to handle.
02
Routing rules
Queues, owners and SLAs defined with your team.
03
Draft-only mode
The agent drafts; staff send. Accuracy measured on the drafts.
04
Autonomy by category
Automatic sends for routine acknowledgements, drafts for everything else, permanent human review for clinical and financial content.
Guardrails
What it will never do.
These limits hold regardless of configuration. They exist so that you can put an agent in front of patients without putting your licence next to it.
- Never sends clinical content, results or financial detail without explicit staff approval
- Cannot be prompted into disclosing information to an unverified sender
- Release-of-information requests always route to a human compliance owner
- Attachment handling restricted to approved file types with malware scanning
- Drafts always labelled as drafts until a named staff member sends them
Human in the loop, by your choice
Every action can require approval before it commits. Start with all of email agent queued to your staff or to ours, watch the accuracy numbers, then widen autonomy workflow by workflow — or leave it queued forever. Both are supported.
Escalations land with certified people
Our review queue is staffed by the same certified coders, billers and credentialing specialists who run our managed services. Not a support desk, and not a chatbot watching a chatbot.
Full evidence trail
Input, decision, action, confidence, model version, reviewer and timestamp — retained and exportable as CSV or JSON, including after you leave us.
What we report against
Operational targets for this agent. They are internal commitments measured on your own data, reported monthly — including the misses.
70%+
Auto-triaged
< 1 hr
Draft turnaround
90%+
Routing accuracy
0
Clinical replies sent
Works with your systems
Your EMR, your numbers, your data.
- Microsoft 365
- Google Workspace
- Any IMAP mailbox
- Athena
- eClinicalWorks
- NextGen
We integrate through FHIR and HL7 where the vendor supports it, and through secure automation or structured document intake where it does not. No API required, no vendor commission taken.
FAQ
Common questions.
Does it read our whole mailbox?
It reads only the mailbox or shared inbox you designate, under a service account with least-privilege access. Access is logged and revocable at any time.
Will it email patients on its own?
Only for categories you explicitly approve, and only for non-clinical content. Everything else waits for a human to press send.
How does it handle a records request?
It identifies it as a release-of-information request, checks the form is valid, starts the deadline clock, and routes it to your compliance owner. It never sends records itself.
Can it work with our shared inbox?
Yes. Shared mailboxes are supported, with per-user attribution so the audit trail still shows which human approved what.
Related agents
Usually deployed together.
05
In development
AI Document & Referral Agent
Reads inbound faxes, scans, portal messages and referrals, identifies the document type, matches the patient, files it in the right chart and turns orders into tasks.
- Faxes and referrals classified and filed to the chart
- Patient matched by name, DOB and MRN — with exceptions queued
- Orders, labs and abnormal findings routed to a clinician
06
In development
AI Scribe
Listens to the visit and drafts the note in your format — SOAP, APSO or your own template — with suggested codes, orders and follow-up. The provider reviews, edits and signs.
- Ambient capture in the room or on telehealth
- Drafted in your template, in your voice
- Provider reviews and signs — the note is never auto-filed
07
In development
AI Coding & Charge Agent
Turns the signed note into codes and charges, checks them against payer rules and your fee schedule, and queues anything ambiguous for a certified human coder.
- ICD-10, CPT and HCPCS suggested from the signed note
- Payer-specific edits applied before the claim is built
- Certified coder reviews before release — always
See it run on your own workflow.
A 30-minute demo using your scenarios, your payer mix and your systems. We will show you the guardrails and the audit log, not just the happy path.
A 30-minute call with an operations lead. No obligation, no sales sequence, and we will say so if you do not need us.