AI in Hero EMR

Run hands-free commands with the voice assistant

A round magic-wand button floats at the bottom-right of every Hero EMR screen; hover it and the label reads Voice assistant · F1, while the button’s full tooltip lists every action the wand offers your account. Press F1 anywhere (or click the wand) and speak. The assistant reads two things — your words and the patient whose chart is open — and turns them into a draft: every command ends in a pre-filled modal or composer for your review, never a direct action.

Every shortcut in this manual — F1 included — is a default. Remap any of them from Settings → Hotkeys (Ctrl+Shift+?), and the app’s own tooltips and flyout hints update to whatever you chose.

1. Start and stop recording. Press F1 or click the wand. The button turns red with a stop square, a waveform animates, and a dark status bar expands to the left; press F1 or click again to stop, or let it auto-stop at 60 seconds. First use asks for browser microphone permission, and if a note dictation is already running you’ll see Stop the active note dictation before starting Command Hero. — the two share one recorder.
2. Watch the status bar. It shows Transcribing..., then Processing command..., then your transcribed words in quotes with a summary of what the assistant did underneath. Results clear themselves after about five seconds of inactivity.
3. Confirm the patient. With a patient chart or patient tabs open, the command applies to that patient. Otherwise the assistant opens the patient search modal and shows Please select a patient to continue... before doing anything.

Ask clinical questions with Knowledge Hero

Ask a clinical question — first-line treatment, contraindications, diagnostic criteria — and Knowledge Hero answers it from your practice’s guideline library. The status bar turns emerald with Searching clinical guidelines..., then a panel slides in from the right titled Knowledge Hero, subtitled Clinical Decision Support. Ask it by voice with F1 or by typing the same question into the F2 command box — both run through one interpreter, and a clinical question routes here either way. (The F3 assistant is a different thing: it answers questions about the software, not clinical ones.)

1. Read the answer and its sources. The panel shows Your Question, the answer, and a numbered References list; header badges count the sources used (“N guidelines referenced”). Afterwards the status bar reports Found answer using N guideline(s) or Answer generated (no specific guidelines matched) — weigh an unmatched answer accordingly.
2. Copy or close. Copy Answer (it flips to Copied!) puts the answer on your clipboard; Done or Esc closes the panel.
3. Check whose chart the answer used. With a chart open, every clinical question uses that patient’s chart: their allergies, current medicines, latest weight and height (with its date) and active problems. The panel says so at the top — Answer for {name}, {age and sex} with Uses … listing the facts used — and a weight-based dose shows the charted weight and the arithmetic, or says there is no weight on file. Without a chart open it reads No patient chart was used. If you have moved to another patient’s chart since asking, a red banner says You are viewing {name}. with Go to {name}’s chart, and the answer can’t be copied into the wrong chart. Each part of the chart is read only if your account may read it.
4. Ask for the chart’s own lists. “What allergies does this patient have?”, or the same for medicines, problems or weight, opens a From the chart panel with the chart’s own list rather than an AI answer. For anything else in the chart (labs, notes, letters), Hero says it can’t search inside the chart yet and offers one click to open that part of the chart, such as Open Labs.
5. Curate the guideline library. Open the Settings gear and choose Knowledge Hero Guidelines. On the Guidelines list you can enable or disable individual guidelines, add and edit custom ones, Revert to Original, or Delete Custom Guideline. The Settings tab holds one toggle — Enable previsit workup generation based on previsit interview? — which makes completed previsit interviews generate Knowledge Hero guidance automatically (see Questionnaires & paperwork).

Ask how to do something in Hero EMR (F3)

Press F3 anywhere to toggle a compact chat panel above the wand button, titled Hero EMR Assistant with the subtitle How do I…? · F3. Use it for how-do-I questions about Hero EMR itself — setting up a workflow, where a feature lives, how a setting behaves. Its answers are drawn from this manual, the official Hero EMR provider manual, and each one ends with links back to the pages it used. It is not a clinical reference, and it does not read patient data.

1. Open it. Press F3, or hover the magic wand and click How do I…? — useful on a Mac, where F3 is often claimed by Mission Control. The assistant is available to every role — admins, physicians, and staff; answers hide dollar amounts for users without billing access. Dismiss it with Esc or the Close (Esc) button.
2. Pick an example or ask your own. The empty state says where answers come from — Ask how to do something in Hero EMR. Answers come from the official Hero EMR manual, with links to learn more. — above clickable starters such as How do I set up a cash-pay only price for one visit type?, How do I connect Zoom for a telehealth visit?, and How do I build paperwork a patient signs before their visit?
3. Send and read. Type into Ask a question… and press Enter (Shift+Enter for a new line). The panel shows Looking it up in the Hero EMR manual… while it works, then answers with headings, bullets, and numbered steps, closing with a compact Learn more row of links you can click straight through to the manual page. Any errors render inline in the thread.
4. Start a fresh question when the subject changes. The pencil button in the header (New chat) clears the transcript. Each question is answered on its own — the assistant does not carry the previous answer forward, so a follow-up like “and then?” will not land. Ask the whole question every time.

Generate note content with AI in the Note Writer

Two AI features live inside the note editor. AI Note Generation answers questions about the open patient while you write, via the .hero dot-phrase; Smart Encounter Completion extracts structured data from your note when you save. Both read the open patient’s chart and your note text — and both end in content you review before it counts.

1. Generate with .hero. Type .hero in the editor followed by a clinical prompt. The dropdown reads Type or select a question about this patient... and offers quick examples such as Summarize patient’s cardiovascular history, List all previous surgeries, and Summarize recent lab trends; your favorites and recent prompts are remembered. .hero is patient-bound — it reads the open chart: the chart’s own lists of allergies, current medicines, latest weight and problems, as well as the visit history. It lists every item, says “none” only where the chart does, and states the weight and its date for a weight-based dose. If the lists couldn’t be read, the note writer says Chart lists not read. Knowledge Hero (F2) uses the same chart facts when a chart is open.
2. Know which dot-phrases use AI. Among the data dot-phrases, only .imagesum is AI-powered — it inserts an AI imaging summary (search prompt Search imaging summaries). The others (.lab, .imaging, .questionnaire, .media) insert chart data directly, with no model in between.
3. Let dictation caption your skin photos. .derm, .lastderm and .dphotos place the pictures inside a Skin photos Smart Block — one row per photo with its thumbnail, body site and date, and an empty caption. Dictate the exam as you normally would (“left forearm, close-up: five millimeter dome-shaped brown papule, well defined”) and the ambient note generator writes each photo’s caption from your own words, matching photos by body site and by wide shot versus close-up, while the Physical Examination section is still written exactly as your template asks. AI-written captions carry a sparkle badge; a caption you type yourself is never overwritten. The block’s microphone dictates straight into it, and because the photos live in a Smart Block they survive a regeneration of the note. Insert .skinphotos in a note template to have the patient’s questionnaire photos pulled in automatically when the note opens.
4. Correct the AI photo interpretation by dictating. When the photos came from a questionnaire that ran the photo analysis, the block also carries a Photo interpretation section — category, morphology, colors, border, symmetry, size, notable features, differential, urgency, red flags and summary — pre-filled from the AI as a draft. Your dictated exam overrides it line by line: if the draft says brown papule, dermal nevus and you say “four millimeter pink macule, benign lentigo”, morphology becomes macule, colors pink, category and differential benign lentigo, and the lines you did not address stay as drafted, each change marked with the sparkle. Read the corrected lines and click Mark as reviewed to sign them onto the photos (the same signature the questionnaire panel shows); Reset to AI draft restores the analysis, and Sign again appears when a line changed after signing. Lines you edit by hand are never refilled.
5. Keep the built-in guide handy. The toolbar Info button (Note Writer info), or F8 from inside a note, opens the Note Writer Guide — tabs Shortcuts, AI & Data, Dictation, Collab, and Workflow — the in-app reference for everything in this section. (The Shortcuts tab also covers F2/Shift+F2 template-field navigation — while a note has focus the note owns F2, so it moves between fill fields rather than opening the assistant’s command box.)
6. Save, then review the extraction. Saving the note triggers Smart Encounter Completion — as the guide puts it, “When you save, AI automatically extracts structured data from your note”: Diagnoses & ICD-10, Medications, Billing codes, Follow-up plan, and After-visit summary. A processing modal shows Processing..., then AI extraction complete and Encounter completion data has been refreshed. Everything lands in the Encounter Completion checklist for review — see Auto-completion.

Dictate notes and capture telehealth audio

The round microphone button in the note header and toolbar records the visit. When you stop, the AI reads the transcript and the note template you chose (see Note templates) and generates a full note that streams into the editor — structured, templated prose, not your words verbatim. For telehealth, an audio-settings panel lets the same recording capture the Zoom side of the conversation. When you want your words verbatim instead of a generated note — a paragraph typed exactly as you say it — use raw dictation (Ctrl+D), covered in Magic wand quick actions.

1. Record, stop, and watch it generate. Click the mic (Start Recording / Stop Recording). The status runs Live → Transcribing → Generating note… → Done as the note streams in. The mic and the F1 assistant share one recorder, so they can’t run at once.
2. Silence pauses the recording; too little speech writes no note. If five minutes of recording bring almost no speech, the recording pauses: “We haven’t heard speech for 5 minutes, so recording is paused. Everything recorded so far is saved.” Choose Resume recording or Stop recording. Hero writes a note only from at least 30 heard words (filler such as a lone “thank you” doesn’t count); otherwise Not enough speech to write a note says how many words were heard and that your note was not changed, and the recording stays saved (Try again from saved recording).
3. Capture the patient on video visits. The chevron next to the mic (Show audio settings) opens the Audio source panel: a Patient audio (video visits) checkbox, a Microphone picker, and a Patient audio picker (Select a device). Tips under them suggest a device for your computer (for example the laptop microphone when you use speakers, or ZoomAudioDevice in the Zoom app). Rather than guessing, press Check my setup in the same panel and let Hero find the settings that work — see Check your dictation setup. Connecting and running Zoom visits is covered in Run telehealth visits.
4. Check the template. Dictation starts with the visit type’s note template, otherwise your default template. While you record, the Note Template picker in the recording overlay lists your My templates, and More templates… reaches the rest of the library — see Note templates. Transcripts are not labeled by speaker; there is no speaker-diarization setting.
5. Navigate away safely. Leaving the note mid-dictation shows a compact dock at the bottom-right with the patient’s name, a Live/Processing/Done badge, and Complete record / Return to note actions. Drag the dock by its top strip anywhere on screen, or minimize it into the top toolbar as one chip (recording dot, levels, status, patient) — the chip still shows every warning, and a click opens it again. Double-click the strip to send it back to its corner; this browser remembers where you put it. The browser also warns before you close the tab while a dictation is in progress.
6. A recording is never lost silently. If something stops a recording (for example you were signed out in another tab), Hero keeps what it recorded and says Your recording was stopped. with Open note. If you sign a visit whose recording was stopped and saved but never added to the note, Hero asks first — Your recording for {patient} isn’t in the note yet. — with Use the recording (opens it for the note with its template) or Discard recording and sign.

Smart Block fields that fill themselves from the chart

Some Smart Block fields read their value straight from the chart instead of waiting for you to type or dictate it. Each such field is paired with a system dot phrase: .spravato’s PHQ-9 today pulls .phq9today, its screening baseline pulls .spravatobaseline, and its sitting baseline blood pressure, pulse, respiratory rate and SpO2 pull the first reading of this visit’s vitals. Typed in a note, the same phrase inserts a readable line; pulled into a field, it supplies the number (or the blood pressure) alone.

1. Pulls fill empty fields only. A pulled field fills while it is empty: when the block opens on a note you can edit, and again when you come back to the window (at most once a minute), so a PHQ-9 the patient finishes on the kiosk after you inserted the block arrives without a click. A pull never replaces a value you typed, dictated, or carried forward from the last visit — in .spravato the last visit’s values come in first and pulls fill what is still empty. Signed, locked and read-only notes, previews and the Smart Block builder never pull.
2. A small link mark shows what came from the chart. A pulled value carries a muted chain-link mark beside the field’s label (or inside the score or vitals cell) and a very faint tint, for as long as the value is still the chart’s. Hover or tab to the mark to see where it came from, for example From the chart: PHQ-9 completed 10/28/2026 (.phq9today). The mark is only on screen: the note text prints the value like any other.
3. Edit, dictate over, or clear it. Change a pulled value and it becomes yours (the mark goes away). Dictation that states a different value replaces it and shows the AI sparkle; dictation that says nothing about it leaves it alone. Clear a pulled field and it stays empty: that field is not pulled again in that block.
4. The same access as typing the phrase. A pull shows exactly what typing its phrase in that note would show, under the same access to the patient. When the chart has nothing (no PHQ-9 for this visit, no vitals recorded yet) the field simply stays empty.
5. Pair a field with a phrase in the builder. In the Smart Block builder, a field’s Pull from chart select lists the phrases that fit the field type (scores fill number and text fields; a blood pressure fills text fields), with a separate choice for the first and the last reading of the visit, for example This visit’s BP (first reading). Its help line reads Fills the field when it is empty. Clinicians see a small link mark; editing the value makes it theirs. Practice and personal blocks can both use pulls.

The phrases a field can pull. Each also works typed in any note:

Pick your note templates: My templates and the Library

A note template is the structure the AI writes into when you stop a dictation — its sections, wording, blanks, and choices. Templates, visit-type defaults, and your note-writer settings all live in the SOAP Notes tab: open the top-toolbar Settings gear and choose SOAP Notes. It opens as a closeable tab in the EMR workspace, like Billing or Questionnaires, with three sections in its left rail — Templates, Visit types, and Settings. The Templates section keeps two lists:

1. Switch between the two lists. The switch at the top of the template list flips between My templates and Library, each with a count of the templates it holds.
2. Add, remove, and reorder. To add a template, click Add on its Library row, or open it and turn on In My templates in the details panel. To remove one, hover its row in My templates and click ✕, or click Added on its Library row. Drag rows in My templates to reorder them; the dictation pickers list them in the same order.
3. Create a template with New. AI wizard (marked Recommended) drafts a template after you answer a few questions or share a sample note. Start from a copy copies a Standard template or one of yours, and Blank template lets you write the structure yourself.
4. Edit and save. The editor shows the name and description at the top, then the template body — type *** for a free-text blank and {{Option A|Option B}} for a choice (in the note, F2 jumps between them). Chart phrases such as @age@ or .lastvisit are filled from the chart, never by the AI (Chart phrases in AI templates). The collapsible AI instructions (Advanced) panel at the bottom holds extra instructions for the AI. Save (Ctrl/Cmd+S) and Discard sit in the header. Standard templates are read-only; click Make a copy to adjust one. If you switch to another EMR tab before saving, your edits are still there when you come back, with the banner Your unsaved changes were kept while you were away.
5. Use the details panel. Beside the editor: Default template with Make default; the In My templates switch; Dictation Defaults, which sets the Default Context Mode and the visit types that start dictation with this template (Auto-Select for Visit Types); Test template, which runs the template on a sample or pasted transcript (save first — the button reads Save to test while you have unsaved edits); Make a copy; and, for your own templates only, Delete template.
6. Choose a template while dictating. Dictation starts with the visit type’s note template, otherwise your default template. To use another, open the Note Template picker in the recording overlay, or in the Retranscribe from saved recording?, Start a new recording?, or Recover interrupted recording? dialog. It lists My templates, then More templates…, which opens Choose a template — a searchable list of the whole library. The template you pick there is used for that note only; tick Also add it to My templates to keep it on your list for next time.

Share a template with the practice

The Share with the practice switch on a template you wrote (on by default for new templates) puts it in everyone’s Library under Practice. Colleagues can add it to My templates, make it their default, link it to visit types, write notes with it and copy it, but only you or a template administrator can edit, rename or delete it; a colleague’s shared template opens read-only with Shared with the practice by <name>. and Make a copy. Turning the switch off asks Stop sharing this template? — colleagues who added it lose it.

Delete a custom template

1. Open the template in Settings → SOAP Notes → Templates. Find it under Only you or Practice in the Library.
2. Click Delete template in the details panel, then Delete. Delete this template? says where it disappears from: your library and My templates, or, for a shared one, the practice Library and everyone’s My templates. Notes already written with it are not affected.

Standard templates and colleagues’ shared templates have no Delete template button. To stop seeing a template on your list, take it off My templates instead.

Check your dictation setup: does Hero hear you and the patient?

Check my setup tests your microphone, speakers or headphones, and video-call audio in one to three minutes, then saves the audio settings that worked so ambient dictation hears both you and the patient. It is available to every clinician; there is nothing for an administrator to turn on.

1. Open the check. In a note, click the chevron next to the microphone (Show audio settings) and press Check my setup. An amber line above the record button — Check your audio before this video visit or You haven’t checked this setup yet. — opens it too, and can be dismissed for the day. You can also run it from Settings gear → Profile Settings → Dictation with Check my setup. Stop any recording first.
2. Set up the way you see patients, then press Start check. Headphones on or off as usual, and Meet or Zoom using your usual speaker. Run it when you are not on a call. If the browser asks to use the microphone, choose Allow.
3. Answer and read aloud when asked. Tell Hero where you hear the patient on video visits (Computer speakers, Headphones, or Pick a device), stay quiet, choose I’m ready, and read the sentence out loud. Hero then plays a short computer-generated test voice; stay quiet while it plays. It may ask you to read again (“Round 2. Hero tests your settings one at a time… This is normal”) — you are not stuck in a loop.
4. Read a sentence from the patient’s seat. Leave the computer where it is, sit where patients sit (or ask someone in the room), and choose I’m in the patient’s seat. Skip leaves the in-person patient side unchecked.
5. Read the result and press Done. Dictation is ready. Hero can hear you and the patient. means you are set. Otherwise the result says what to fix, for example moving the microphone closer to the patient’s chair, using computer speakers for video visits, or setting up headphone audio (Set up headphone audio). Check again reruns it. Hero saved the audio settings that worked. confirms the new settings.

Profile Settings → Dictation lists the setups you have checked on this computer, each with an In person and a Video visits result, Check this setup and Forget. They are saved in this browser only, so check again on another computer, and after Hero tells you your devices, settings or browser changed.

Recordings and transcripts: read, recover, or rewrite a note from a recording

An ambient dictation recording and its transcript are kept, encrypted, on the computer and in the browser where you recorded it, for up to 10 days, including after you close the browser. They are there so Hero can finish, recover, or rewrite the note — not as a copy of the visit. The signed note is the record. Hero keeps no copy of the recording or transcript on its servers, so neither is visible on a different computer.

Read or copy the transcript

1. Open the note on the computer you recorded on. Click the history icon next to the microphone (Retranscribe or view transcript).
2. Choose View transcript. The Transcript window shows what the recording heard, before it was written into the note, as running text.
3. Press Copy to put it on the clipboard, or select part of it. Close clears it from the screen.

The window says “This transcript is kept only on this computer, encrypted, for 10 days after the recording, so it won’t be visible on a different computer.” If part of the recording was never transcribed (for example during a connection problem), it says the transcript is incomplete; use Retranscribe to fill it in. Viewing the transcript doesn’t change the note.

Rewrite a finished note with another template

1. Open the note on the computer you recorded on. Click the history icon next to the microphone (Retranscribe or view transcript) and choose Retranscribe.
2. Pick the template in Retranscribe from saved recording? Choose another template from the Note Template picker (or More templates…).
3. Decide what happens to the current text and press Retranscribe. Replace this text keeps the old text in Version History; Keep this text and add the new note after it adds the new note below.

Recover an interrupted recording

When a recording was cut off — the tab closed, the browser crashed, your session expired — open the same note on the same computer and look for the history icon next to the microphone, labelled Recover recording or view transcript, and choose Recover recording. The Recover interrupted recording? dialog reads “Hero found a recording that was interrupted before it was saved. It can write a new note from it.”; press Recover recording. If the page was closed or reloaded, Hero does not prompt you when you come back, so open the note and check for the icon. (While the page stays open, a Your recording was stopped. notice with Open note takes you there.) There is nothing to recover when:

• you are on a different computer or browser than the one you recorded on;
• you clicked Logout before every part of the recording was transcribed — after you sign back in, Hero can still use a recording only if every part of it had already been transcribed. Hero warns first (“A recording hasn’t been added to a note yet.”), so add it to the note before you log out;
• another clinician signed in on that computer after you, which removes your saved recordings there; or
• more than 10 days have passed.

“Saved on this computer” messages

Use another ambient scribe (Heidi and others)

Hero has no connection to outside ambient scribes such as Heidi, and can’t import their recordings. If you use one, paste the finished text into the Hero note and review it before signing, or attach the document and sign the visit with it (Attached outside note). For Zoom visits, Hero can import the Zoom transcript (Run telehealth visits).

What the AI reads from earlier visits: Note Enrichment and Chart Context

When dictation writes a note it can draw on the chart as well as today’s recording. Two settings decide how much: Note Enrichment (which earlier visits) and Chart Context (which parts of the chart history).

1. Choose the Note Enrichment for the recording. Open Note Enrichment in the recording overlay. A template’s Default Context Mode (in its Dictation Defaults, see Note templates) sets where it starts.

   Option	What it adds
   None	Nothing from earlier visits. Ideal for new patients.
   Last Note	The last note. For a simple follow-up.
   Problem Focus	The history of the problems on the problem list.
   Exploratory	A dated brief of the last visit — diagnoses, medications and decisions, open items — plus notes written since. Newer records win, and today’s details always come from today.

2. Choose the chart history the AI may read. Settings gear → SOAP Notes → Settings → Chart Context: Past medical history, Past surgical history, Social history, Family history, GYN / obstetric history, Psychiatric history (medication trials, hospitalizations and suicidality history) and Patient questionnaires. A part you turn off is left out entirely.

Pull past psychotropic medication trials into the note

A template can’t copy a list out of an earlier note by itself. The trials have to be on the chart’s Psychiatric History, and from there they reach the note three ways:

• In a dictated note: keep Psychiatric history on in Chart Context.
• As text: type .psychmedtrials (.medtrials) or .psychhistory, or put one in a visit type’s starter text (Specialty phrases). These need the Psychiatric History chart section turned on.
• Keep the chart’s list current automatically: an administrator can turn on AI chart history updates under Admin → Documentation → AI chart history (off by default) and tick Psychiatric history. “After a visit is signed, AI updates these history sections from the note. Each change is marked for review: approve, edit or revert it.” Psychiatric History must first be on in Admin → Specialties & chart sections.

To start from the whole previous note instead, use Copy forward.

Diagnoses added in Encounter Completion

Information flows from the note to Encounter Completion, not back: saving the note fills Diagnoses from your assessment and plan, but a template or dot phrase can’t read the diagnoses you add by hand in Encounter Completion into the note text. Write or dictate the diagnosis in the assessment, or insert it from the problem list with .diags (Pickers).

Psychotherapy add-on and the split med-management / psychotherapy note

No particular words or phrases switch on the psychotherapy add-on part of a note. What decides it is the note template and what actually happened in the visit. A split template such as Hero’s standard Psychiatry + Psychotherapy Split Note (or your copy of it) writes two notes from one recording: the clinical psychiatry note and, only when the visit had a real therapy segment, a separate psychotherapy process note. Ordinary templates always write one combined note.

1. Turn the split on (administrator, once). Under Admin → Documentation → Note types, set Split psychiatry and psychotherapy notes to Enabled, choose a Psychiatry note type and a different Psychotherapy note type, and press Save split settings.
2. Record with the split template. Pick it in the recording overlay’s Note Template picker, or make it the visit type’s note template (Note templates).
3. Say the psychotherapy time out loud. The AI writes the process note only for genuine psychotherapy: a distinct therapy segment with a recognizable modality and structured therapeutic work. Medication counseling, side-effect strategies, sleep or lifestyle advice, and general supportive conversation during a medication visit don’t count and go in the clinical note’s plan. The clinical note’s Psychotherapy Add-On section records that psychotherapy was done, the modality and the time, and the time only when it was stated. Hero does not work out the split from the recording’s length: in a 60-minute recording, say for example “we spent 40 minutes on psychotherapy”.
4. Review the add-on code in Charges. On save, Hero suggests the add-on code from the documented psychotherapy minutes only: 16–37 minutes 90833, 38–52 90836, 53 or more 90838; no stated time, or under 16 minutes, means no add-on. It goes with the visit’s E/M code and is never billed alone. Check it before you sign (Charges & E/M).

Chart facts in an AI template come from the chart, not the AI

When an AI note template contains chart phrases such as @age@, .diaglast, .appointmentdate, .appointmenttime, @chartmeds@, @chartvitals@ or .nextappt, Hero fills them in from the chart. The AI never writes them, so it can’t put a wrong age, a wrong visit, a wrong appointment time or a made-up diagnosis list into your note. This works with every AI note template you dictate with: the note writer’s recorder, ambient dictation and a Zoom transcript import. There is nothing to turn on.

1. Write your template as you do today. Put chart phrases where you want chart facts, instructions in {braces}, and *** where the AI should fill something in from the visit (Note templates).
2. Dictate the visit and pick the template.
3. Read the note. The chart phrases stay exactly where your template put them, filled from the chart. The AI may still use those facts in its own sentences (for example the patient’s age in the history), but it can’t change them.

Good to know

• Both forms, any capitals. @age@ and a bare .age both work, in any capitals (@Age@, .LastVisit).
• Options. @age written@ gives “42 years old” and @age yo@ gives “42yo”; plain @age@ gives just “42”. Write This is a @age written@ patient if you want the words in the sentence.
• If the AI leaves a fact out or writes its own version (a wrong age, a made-up date), Hero puts the chart’s value back on that line of your template. A sentence the AI wrote there of its own (for example “Stop sertraline today because of a rash” after “Medications:”) is kept, right after the chart’s value.
• Appointment date and time now fill in during ambient dictation too, from the visit’s booked appointment.
• Inside an instruction. A phrase inside braces (for example {The patient is a @age@ patient ...}) is filled when the AI uses it, and left out when the AI rewrites the instruction its own way.
• Your text phrases (for example a .lab or .imaging that inserts a fill-in form) are left as they are, as before.
• Visit-only phrases. Some phrases need the visit’s own record (see Visit phrases); the note editor fills them right after the note is written, in the same place.
• Privacy. Phrases that show sensitive history to a parent who can read a teen’s record raise the same privacy notice as when you insert them by hand.

Auto-fill prescriptions from your note (Medication Hero)

Wherever you build a prescription — encounter completion orders (Prescribing) or the standalone order flow (Standalone orders) — the Edit Medication dialog has an orange Auto-fill button (wand icon, tooltip Autofill using clinical context from your SOAP note). It sends the medication, your SOAP note, and patient context to Medication Hero, shows Loading..., and drafts the whole regimen. The same engine powers the F1 prescribe command (Voice assistant) and order extraction from faxes and inbox messages (Faxed orders, Inbox AI).

1. Document the plan first, then click Auto-fill. Auto-fill quality depends on what’s in your note. Run it after the plan is written and the fields are derived from your documentation; run it before documenting and you get defaults, not chart-derived values.
2. Review every filled field. It fills action, dose and dose units, route, frequency, duration, dispense quantity and units, and a recommended NDC, and it can set the PRN state — ticking As needed (PRN) and filling the reason field (As needed for... (e.g. pain, nausea, anxiety)). The regenerated sig preview mirrors exactly what the pharmacy will receive — read it the way the pharmacist will.
3. Read the warning banners. A Medication Hero warning amber banner lists any safety flags from the analysis; if key values couldn’t be determined you’ll see Some key information could not be determined, please check medication dosage and duration carefully. Controlled substances also show a PDMP status box — see Safety checks.
4. Controlled drugs are drafted in the fields. For Schedule II–V medications Auto-fill and the assistant always fill the discrete, reviewable fields, never written directions. You can still choose Write the directions myself when the fields can’t say it (for example a different dose at noon); Auto-fill then leaves the directions you wrote as they are. Signing them uses EPCS.

Use AI in the Inbox: drafted replies & order extraction

Every Inbox reply box has the same AI assist: the model drafts in the box, and you edit and send. The click-by-click workflow lives in AI replies; this section maps where each control is, what the model reads, and what to check before sending.

1. Draft in the reply box. Type the facts and plan in shorthand and click Expand, or click Quick reply on an empty box for a draft from the conversation. The result is marked Expanded by AI or Drafted by AI, and Undo puts back what you had.
2. Edit, then send it yourself. Nothing is ever sent automatically. Check the recipient and channel, then press the send button. Stored suggestion chips, when a thread has them, can still insert a draft into the box for the same review.
3. Treat extracted orders as drafts. Extract orders parses your message into draft diagnoses, medications, labs, imaging, referrals, questionnaire orders, and a follow-up timeframe in collapsible panels with counts (for example 3 orders). It needs the thread linked to an encounter (This message is not linked to an encounter) and the message fully sent (Wait for the message to send before extracting orders); the drafts then go through the normal order review and safety checks — see Safety checks.

What the drafts will and won’t say

• Portal replies read like a chat, not a letter. No “Dear…” or sign-off unless your saved style adds one, and when the newest message in the thread is yours, the draft is a follow-up rather than an answer. An email reply may still read as a letter.
• No promises about prescriptions you didn’t make. A draft says a prescription was sent only if your words say so (“rx sent”, “called in”); “send it in” lets it say the prescription will be sent. For a controlled medicine, a draft never says a refill was approved or is ready to pick up unless you wrote it. A sentence that breaks these rules is removed or marked with a red line (Not allowed for a controlled medicine) for you to fix.

Reach Command Hero from the magic wand — hover, type, or click

The round magic-wand button at the bottom-right of every screen is more than the F1 voice trigger. Hover it (or tab into it) and a labelled menu rises above it — one row per action, each with its current shortcut printed on the right. Clicking the wand itself still starts and stops the voice assistant — the menu simply adds typed and click entry points to the same command engine, so you never have to speak to use it.

What the menu holds depends on your account — rows appear only when you have the matching access:

The commands follow your role, in the menu, the voice assistant and the F4 guide alike: ordering and prescribing need order access, clinical questions need note access, editing a note needs permission to edit visit drafts, messages need permission to send messages, and reminders need to-dos. Billing staff get no commands — the wand opens the Hero EMR help instead — and the check-in kiosk gets no assistant at all. A command your role can’t use says Hero commands aren’t available for your role.

1. Type a command — F2. Press F2 (or click the Type a command row) and a small Type a command for Hero box opens just above the wand. Whatever you type runs through the same pipeline as the voice assistant — “order a CBC”, “prescribe amoxicillin 500 mg”, “message John Smith that his labs look fine” — and ends in the matching drafted order, prescription, or message for your review. Enter sends, Esc closes, and the dark status bar to the left of the wand shows Processing… and the result exactly as it does for voice. The footnote says it plainly: Same commands as the voice assistant · Enter to send · Esc to close.

   

   • In a note, F2 first jumps to the next *** blank; with none left it opens the box. Esc empties and closes the box and puts you back in the note.
   • Orders say what you typed. “stat”, “ASAP” or “urgent” sets that order’s priority (for labs that aren’t told STAT electronically, the row says Call the lab.); “A1c and lipids” becomes two rows, “chest x-ray and CBC” an imaging row and a lab row, up to five at once. An imaging study is matched to your practice’s imaging list with the views, contrast and side you said; a row Hero can’t match says why and can’t be sent. Typed while this patient’s draft visit note is open, the orders are added, unsent, to that visit’s Orders to review there.
   • Hero only starts medicines. “Stop simvastatin” opens no order window: Simvastatin was not stopped — Hero only starts medicines. Stop it in Medications.
   • Charts and your day. “Open Sofia Garcia’s chart” opens it, or a pick list when several match; “What is on my schedule today” lists your visits for the practice’s day, in practice time (Your schedule). A follow-up such as “and for this patient?” is read with your last question.
   • Reminders with a day but no time are due at 8:00 AM practice time, and the confirmation names the weekday and time zone.
   • A command Hero can’t place stays in the box with one plain line, so you can reword it.
2. Raw dictation — Ctrl+D. The microphone row records your exact words anywhere in the app, cleans them up with AI — punctuation, false starts, run-on sentences — and inserts the result at your cursor, in whatever field or note you were typing in when you started. A small Raw dictation panel tracks the run: Recording…, then Transcribing… and Improving transcription…. If the text has nowhere unambiguous to land — no field was focused, or the cleanup failed — the panel keeps it in an editable box with Copy text rather than dropping it somewhere you didn’t put your cursor. A dictation runs up to ten minutes (voice commands stop at 60 seconds), and it shares the one recorder with F1 and note dictation, so only one of the three can run at a time.
3. Create orders — Ctrl+O. The Create orders row (or Ctrl+O) opens the standard order-entry dialog, Create Orders, for medications, labs, imaging, screeners, and forms. With a patient chart open, that patient is already selected — the dialog carries their name in its title; with no patient in context it opens the patient search first, exactly like a voice-driven order. Staff see it only if they may enter lab or imaging orders, and their orders wait for the physician they choose to sign them.

   

4. Quick note — Ctrl+N (physicians only). The Quick note row (or Ctrl+N) opens the Quick note dialog: a rich-text editor with a Bold, Italic, Underline, and bulleted/numbered-list toolbar. Its subtitle spells out where the note goes — … files to the chart under a new notes-only encounter. As with orders, no patient in context opens the patient search first. Quick note is available to physicians only.

   

   

5. File it, and it is on the chart. Sign & file to chart creates the notes-only encounter, signs the note, and confirms with a green Quick note signed and filed to the chart. banner — filing is signing, with no separate step.

   

What is the AI pre-visit chat?

The AI pre-visit chat is a short interview the patient completes in the patient portal before a visit: an AI asks the patient questions in a chat, the patient types answers or taps quick responses, and Hero saves a pre-visit interview summary for the care team.

1. Turn it on per visit type. In the visit type editor (Build the visit catalog), under Options, tick Requires Pre-visit Tasks, then AI Pre-visit Chat. The visit type card then shows AI pre-visit chat: Yes.
2. The patient completes it. It appears among the patient’s tasks for that appointment in the portal. An unfinished pre-visit interview blocks patient self check-in and telehealth join (Questionnaires & paperwork).
3. Use the summary. The pre-visit interview summary feeds the visit preview’s summary of what happened since the last visit. If Enable previsit workup generation based on previsit interview? is on in Knowledge Hero Guidelines, a completed interview also generates Knowledge Hero guidance (Knowledge Hero).