A black lab named Henley comes in for what the owner is calling a recheck on a skin issue. The chart from your practice management system shows the SOAP note from three weeks ago, the medication that was dispensed, and the next-recheck date. What the chart doesn't show is the side conversation you had at the end of that visit about Henley's owner traveling to Italy in May, the worry she had about whether the boarding kennel could give the meds correctly, and the half-promise you made to follow up by phone before her trip.
Most veterinarians have the same parallel system going. The PMS handles the medical record, billing, and inventory. It rarely handles the working memory of the practice — the conversational context, the inter-visit threads, the running list of "things I keep meaning to" that follow you out the clinic door at 6 PM.
This post is about replacing that scattered scratchpad with a notes workspace that the whole team can use, and an assistant that can read across the practice's accumulated context in seconds.
A scope note: this is a place for your own structured private notes — clinical thinking, operations, client communication threads, reference material. It is not a substitute for your medical record system. Where the boundary sits is a clinical and regulatory judgment you make as the veterinarian.
What a small-animal practice needs from a notes system
Before getting into the mechanics, the shape matters. Most practices need their working notes to do five things:
- Per-patient continuity — what came up last visit, what was deferred, what to bring up at the next exam
- Client communication threads — the texture of how this owner talks about their pet, their tolerance for cost, their willingness to medicate
- Cross-patient pattern recognition — every Schnauzer with chronic ear issues, every senior cat on the same renal protocol
- Operational memory — drug ordering, equipment service, the new-tech onboarding packet, the running list of warranty issues with the lab
- Reference material — formulary notes, dosing tables, conference handouts, journal articles you've been meaning to revisit
Plus it has to be usable on a busy clinic floor — a doctor or tech needs to open it on a phone or laptop, find the patient, add a note, and get back to the next exam. The same chart-adjacent shape shows up across regulated work — see AI Notes for Dentists and AI Notes for Healthcare.
Markdown pages, organized however your clinic thinks
In Docapybara, every patient gets a markdown page. The title is usually the patient's name and an identifier — Henley Garcia (Lab, 6yo NM). Inside the page, you keep the running thread: client conversations, your private clinical thinking, photos, the plan for next visit, the deferred items.
Page nesting goes as deep as you need. A common shape: Clients → Garcia, M. → Henley, with sub-pages for History, Photos, Recheck Threads. Or for high-volume practices: a flat Patients section with one page per active patient. Either works.
Plain markdown means the notes are searchable, copyable, and exportable. If a patient transfers to another practice and the owner asks for working records to share, you can copy the page out as text. If a specialist needs the conversational context that didn't make it into the SOAP note, you can paste in the relevant section of the page.
A live database of pending recheck and follow-up calls
A :::database::: directive embeds a live database directly inside any markdown page. The most useful place to put one in a veterinary practice: the daily-board page, or a "follow-ups owed" page, with columns for patient, owner, what's owed (call / recheck / lab review / refill), due date, status, and notes.
Six column types are available, which covers most of what you'd want to track. The table sits next to the written context — sort by due date, you see the week. Filter by status, you see what's overdue.
When the assistant adds a row — say, after a surgery you tell it "add a 10-day post-op call for Henley, due next Tuesday, with a note to ask about appetite" — the change is live across the page. When you ask "what calls are owed this week and which ones haven't been made?", it reads the database and tells you.
For drug ordering, a similar inline database tracks pending orders by supplier, drug, quantity, ETA, status. When the receptionist asks "are we still waiting on the Convenia order?", the answer is one query away.
The agent reads across your whole patient history
Capy, the assistant inside Docapybara, reads across your entire vault when you ask. The kinds of questions that become answerable in seconds:
- "What did Mrs. Garcia tell us last time about Henley's response to the new diet?" The agent finds Henley's page, locates the relevant conversation note, and quotes it back.
- "Pull every senior cat in the practice currently on a renal protocol and tell me which ones haven't had labs in the last 90 days." It scans patient pages, returns a list with the relevant detail.
- "Have we seen this presentation in any other Schnauzer in the last year? Pull the case notes and what worked." Cross-patient retrieval, in one query.
The clinical thinking is still yours. What changes is how much of the practice's working memory is actually accessible to you instead of fragmented across years of notes.
Recording the consult so the conversation survives
The most useful detail in a small-animal exam is often the conversation, not the physical exam findings. The owner mentions the limping only happens after long walks. They bring up a budget concern that means the imaging case will need to be staged. They say their other dog had a similar issue last year that they handled differently and they want to know why.
Docapybara records audio inside the workspace and transcribes with speaker labels — so when you replay the consult, you can see what was said by whom. For complex driveability cases (every senior with a multi-system presentation), pre-surgical consults, and second-opinion visits, having the full conversation captured beats relying on what you remember twenty minutes later.
The transcript drops onto the patient's page. Tomorrow when the recheck is on the schedule, the conversation is there as searchable text — not buried in your memory of a Tuesday afternoon.
Old PDFs of records, lab reports, and reference material
Most established practices have years of PDFs sitting on a network drive — referral letters from specialists, lab reports, scanned consult forms, conference handouts, drug-rep materials. They're useful when findable and frustrating when not.
Drop the PDFs into Docapybara and the conversion pipeline turns each one into searchable markdown. The agent can now read across them. So when you ask "what did the dermatologist's referral say about Henley's allergy workup?", the agent pulls the relevant excerpt as text. The original PDF is still one click away when you need to see the original.
This matters most for legacy data — the years of records and reference material accumulated before you had a working notes system. The institutional memory becomes searchable instead of staying locked in PDFs nobody opens.
Operational notes — drugs, equipment, training, the warranty drawer
A clinic runs on more than patients. Drug inventory, equipment service intervals (anesthesia machine, autoclave, dental unit, x-ray), the new-hire onboarding packet, the lab-warranty drawer, the after-hours emergency policy — all of that lives somewhere, and in most clinics it's a folder of Word docs nobody updates.
In Docapybara, those operational notes are nested markdown pages. A common layout:
Operations → Drugs → Controlled Log as an inline databaseOperations → Equipment → Anesthesia Machine Service Log (database) and SOP for Daily Check (written)Operations → Training → New DVM Onboarding, New Tech Onboarding, Receptionist SOPOperations → Suppliers with one page per supplier (account, contact, return policy notes)Operations → Lab → Pending Cases (database) and Issue Log for warranty/quality threads
When you onboard a new tech, you point them at one nested folder and they have everything — written process, live trackers, and the conventions the practice uses for patient pages. When you ask the assistant "what's the dental unit due for service next?", it reads the database row and tells you. When you ask "what's outstanding with IDEXX from last week's lab issues?", it finds the relevant note. The SOP layer specifically is covered in Standard Operating Procedures, Without the Wiki Maintenance Tax.
This is where the workspace stops being just a notes app and becomes the operational brain of a small clinic. One person — or one practice — one vault, every category of working memory in the same searchable place. The agent-acts-on-docs side is laid out in Claude Code for Documents.
Try Docapybara free
The fastest test: open Docapybara, create a page for tomorrow morning's first three appointments, paste in whatever working notes you have on each (the SOAP note, recent client communication, anything you'd want at hand), and ask the assistant for a three-line summary of what to remember at each visit. Five minutes of setup, and you'll know whether having the context next to you instead of buried in the PMS changes the visit.
Try Docapybara free — bring a few of your messiest patient threads, last week's lab PDFs, and one operational SOP that's been waiting for a rewrite. See how the workspace handles them.