Beginner Guide · for physicians

Use Claude with confidence — in the practice and at home.

No tech knowledge required. No generic AI phrases. This guide shows you visually, in the right order and with over 60 ready-to-use prompts, how to put Claude to work on real everyday medical problems from day one — from patient communication to vacation planning.

60+
Example prompts to copy
10
Topic areas, step by step
Work & private life kept separate
0
Prior knowledge required
DW
Dr. Butze, 52
General practitioner · Family practice, Berlin-Charlottenburg

"I have never used ChatGPT or Claude. But my daughter says it changes everything. I want to understand where it helps in the practice and in my private life — without a degree in prompt engineering."

Dr. Butze accompanies you through every section of this guide. Every example comes from his real practice or family life.

Learning path

Three clear stages — no overwhelm

This is how you build confidence step by step, without dealing with every feature at once on day one.

Day 1 — First wins

Create an account, understand the chat window, complete three simple tasks (patient education text, email, research). Pure experimentation.

Go to the getting-started chapter

Week 1 — Projects and routines

Set up your own "rooms" for practice and private life (Projects). Store templates and your own writing style once — Claude remembers.

Discover Projects

Month 1 — Your own assistants

Write system prompts, shape your own "staff members" — e.g. a patient-letter assistant, a travel planner, a literature scout.

Build system prompts
Quick Wins

Be impressed in 5 minutes

Three prompts that show the difference immediately. Just copy, paste into Claude, read the answer.

1. Patient education text in plain language

Chat prompt · 30 seconds
Explain to a 68-year-old female patient with no medical background what a colonoscopy is, how it works, how she needs to prepare, and what risks exist. Write no more than 250 words, very empathetically, at B1 language level, in complete sentences. Do not use any technical terms without an explanation in parentheses after them. Close with a list of three concrete questions she should ask me during the consultation.

2. A difficult reply to a Google Maps review

Chat prompt · 45 seconds
An anonymous patient left the following negative Google review: "Endless waiting time, cold practice, doctor didn't take even 5 minutes, simply unacceptable." Draft three versions of a professional, objective reply: 1. A short, de-escalating reply (max. 4 sentences). 2. A more detailed reply that explains our efforts to stay on schedule without attacking the patient. 3. A reply that invites them to continue the conversation offline. Take applicable professional conduct rules and medical confidentiality into account — never refer to specific patient data. Tone friendly but firm.

3. Plan a family trip — in one go

Chat prompt · 60 seconds
Plan a relaxed family vacation for me, my wife, and our two children (8 and 11 years old) from August 5 to 14 in South Tyrol. Budget: 4,500 euros including travel by car from Berlin. Create: - A route recommendation with two overnight stops - Three hotel or vacation rental suggestions in the mid-price range in the Merano/Bolzano region, family-friendly, with a pool - A relaxed daily plan: at most one "activity" per day, with three days deliberately left unplanned - A plan B for each activity in case of bad weather - A list of 5 restaurants that are also suitable for children At the end, list three things I probably won't think of (e.g. specific toll vignettes, ID documents, travel first-aid kit).
That is exactly the point.

You don't need to know how to prompt "correctly." You only need to say what you want, for whom, in what tone, at what length — and which details you typically overlook. Claude takes care of the rest.

Before you get started: one sentence on data protection.

Never enter real, identifiable patient data into Claude. Not the full name, not the date of birth, not the insurance number. This is non-negotiable — and it isn't necessary either. In every example, we show you how to describe cases so that Claude answers as helpfully as possible without you taking on compliance risks. → Complete rules & examples

Ready for the first step?

Start with the chapter "What is Claude anyway?" — five minutes of reading, and you have the foundation.

Basics · 5 minutes

What is Claude — and what is it not?

A comparison that actually fits: Claude is like a highly competent, fast, very well-read assistant — who, however, doesn't "practice" any profession and remembers nothing you told him yesterday (unless you tell him again).

DW
Dr. Butze asks
Family doctor · Berlin

"'It's a language model' isn't enough for me. I want to know: What can I realistically expect, where are the real limits, and which terms do I need to know?"

The fitting analogy

Imagine Claude as a young, very smart assistant in a library. Over the past years, this person has read millions of texts — studies, books, journal articles, forums, manuals. He can summarize, draft, structure, compare, and translate at lightning speed.

But: this person has never been a patient himself, has never run a practice, is not licensed to practice medicine — and his library reflects a certain point in time. He never leaves the library unless you send him out (that's "web search"). And at the end of the day he forgets the conversation when you leave the room — unless you have set up a permanent desk for him (those are "Projects" and "Memory").

What this means for the practice

Claude is ideal for everything you read, check, and cross-verify with clinical experience yourself. He is not ideal as the sole medical decision-making authority. You remain the doctor — Claude is the fast, friendly staff member.

What Claude can do — and what it can't

Claude is very good at…

  • Drafting texts (patient letter, email, blog article, patient education text)
  • Simplifying complex content — to any desired language level
  • Structuring (lists, comparison tables, SOAP notes, pros/cons)
  • Translating between languages — including medical texts
  • Summarizing long documents (guidelines, PDFs, studies)
  • Brainstorming and collecting differential diagnosis ideas
  • Building code, tables, small web apps, and visualizations
  • Describing and interpreting images (e.g. ECG sketch, photo of a report)

Claude is poor or dangerous at…

  • Independent clinical decisions (never without your review)
  • "Current" information without active web search
  • Guarantees — hallucinations happen, always cross-check
  • Real patient data — never enter anything identifiable
  • Legally binding statements without medical / legal review
  • Image generation (Claude does not generate images, videos, or audio)
  • Making phone calls, booking appointments itself (except with connectors)
  • Remembering old conversations if you don't enable that

How does it work at a high level?

You enter a prompt — that's simply your question or task. Claude processes this text and responds. You can follow up, correct, go deeper. That's called a conversation.

You type or speak

A question, a task, a piece of text — in plain German or English. Long inputs are fine. Attachments (PDF, images, Word) too.

Claude reads and processes everything in context

The "context" is Claude's short-term memory within the current chat. Very large, but limited. As long as the conversation is running, Claude can access everything that came before.

Claude responds — and you iterate

The first answer is rarely perfect. Just say what you want to change: "Shorter." "At B1 level." "With three examples." "In a different tone." That's the lever.

Optional: add tools

Web search, file upload, connection to Google Drive, calendar, Notion, and more (so-called "connectors"). With these, Claude can also do things instead of just answering.

Mini glossary — the eight words that keep coming up

Prompt
Your input to Claude. A question, an instruction, an assignment. The more precise (role, tone, length, target audience, format), the better the result.
Model (Sonnet, Opus, Haiku)
The "version" of Claude that answers. Opus is the strongest and slowest, Sonnet a good all-rounder, Haiku the fastest for light tasks. For doctors, the default setting is enough — you don't have to actively choose.
Context window
The "short-term memory" within a chat. Very large — you can paste an 80-page PDF and Claude understands it as a whole. But: switch to another chat, and this context is gone.
Hallucination
When Claude confidently invents something — a study, a dosage, a date, a source. Occurs less often than with earlier models, but it happens. Consequence: Always cross-check medically relevant facts. Better too often than too rarely.
System Prompt
The "job description" you give Claude once — role, tone, rules, preferences. Applies to all subsequent chats in a Project. More on this.
Project
A dedicated "workspace" with its own system prompt and its own files. E.g. Practice patient communication or Family travel planning. More on this.
Artifact
A "real document" or a small application that Claude builds to the right of the chat: a Word letter, an Excel spreadsheet, an interactive checklist, a small visualization. More on this.
Connector / MCP
A connection from Claude to an external service (Google Drive, Gmail, calendar, Notion, …). MCP is just the technical name for this standard. You enable it once in the settings — and Claude can then, for example, add an appointment to your calendar without you switching apps.

Which model should I choose?

For 95% of all tasks in the family practice and in private life, Claude's default setting is the right one. You don't need to change anything. If you want the details:

Sonnet

The fast all-rounder. Ideal for emails, patient education texts, research, brainstorming, daily planning. Default in Claude Pro.

Opus

The strongest model. Worth it for long, complex tasks — e.g. when you want a 60-page guideline summarized or a complex letter drafted.

Haiku

The fastest, lightest model. For very short, quick requests — such as "What's that medication called again with active ingredient X?".

Simple recommendation

Just leave it on the default setting. Only switch if you notice that a task really needs long answers or particularly difficult reasoning — then pick Opus.

Feature · 8 minutes

The chat — your constant sparring partner.

The chat window is the fastest way to get something done with Claude. You open a new tab, write your question, get an answer, refine. Here you'll learn what separates a good prompt from a bad one — and collect ten templates that fit straight into your daily routine.

When is the chat the right tool?

Chat is right for…

  • One-off, quick requests
  • Quick research and "what if" questions
  • Spontaneous patient communication
  • Ad-hoc translations or explanations
  • Summaries of long texts
  • Brainstorming and differential diagnosis ideas

For recurring tasks, better to…

  • Create a Project (see chapter)
  • Define your own system prompt
  • Upload important files permanently
  • That way Claude doesn't need to be re-briefed every time

The anatomy of a good prompt

A good prompt implicitly answers five questions: who (role), what (task), for whom (target audience), how (tone & format), what to watch out for (constraints). You don't have to phrase it in this order — but the more of this is in the prompt, the better the answer.

Bad → better → very good: the same task in three stages

❌ Stage 1 — too vague
Write something about the flu vaccination for my patients.
✓ Stage 2 — solid standard
Write a short notice about the flu vaccination in my family practice. Target audience: patients aged 60 and over. Tone friendly but firm, no pressure. Length approx. 150 words.
★ Stage 3 — very good
Role: You are an experienced copywriter in a family practice. Task: Write a notice for our practice about this year's flu vaccination. Target audience: Patients aged 60 and over, many with high blood pressure or diabetes. Not very internet-savvy, they like clear, short sentences. Tone and style: Friendly, firm, not preachy. Empathetic. We respect that the decision rests with the patient. Format: - Headline, max. 6 words - Three short paragraphs, 2–3 sentences each - A short bullet list "What you should know" (3 points) - Closing: a simple invitation to make an appointment at the front desk Constraints: - No medical guarantees - Font-size friendly for a DIN A4 printout - No technical terms without a brief explanation At the end, give me two alternative headlines.
The pattern to take away

Role · Task · Target audience · Tone · Format · Constraints. If you keep these six building blocks in mind, you'll automatically write better prompts — without tricks and without a prompt engineering course.

Ten ready-made prompts — straight from Dr. Butze's daily routine

You can copy each of these prompts as-is, adapt it to your situation, and paste it into Claude.

Practice Patient communication Beginner

1. Explain a test result in plain terms

"Ms. M. took her lab report home but doesn't dare to ask. I want to explain to her in one sentence that her thyroid values are borderline."

Prompt
A female patient (early 50s, no medical background) has the following thyroid values: - TSH 4.8 mIU/l (reference 0.4–4.0) - fT4 normal - fT3 normal Write three sentences I can read to her in the consultation room in plain language. Goal: she understands that subclinical hypothyroidism may be present, that we see nothing dramatic, and that we want to re-check in three months. Empathetic, without downplaying. Close with a single question that allows her to bring up symptoms like fatigue or lack of energy herself.
Practice Research Beginner

2. Get a quick overview of a current guideline

"I don't have the DEGAM guideline on irritable bowel syndrome at hand and need the key points for a consultation in 20 minutes."

Prompt with web search enabled
Search the web for the current DEGAM and/or DGVS guideline on irritable bowel syndrome (2024 or newer). Summarize the following aspects for me as a family doctor in a practice, in no more than 250 words: 1. Diagnostic criteria (Rome IV) 2. Most important warning signs that speak against IBS 3. Main pillars of therapy (lifestyle, diet, medication) 4. What I can realistically explain to a patient in 10 minutes during a consultation At the end, provide the source links so I can look up the original passage if needed. Clearly flag anything older than 2023.
Practice Administration Beginner

3. An empathetic refusal to a pharma rep

"I received three appointment requests from pharmaceutical reps today. I want to decline all three politely but firmly."

Prompt
Draft three short email refusals to pharmaceutical representatives who want an appointment at my family practice. Tone: polite, firm, no pressure to justify. I don't explain anything at length, but I offer that materials can be sent by email. Version 1: Very short, 3 sentences. Version 2: Medium, noting that my calendar is full until further notice. Version 3: With the option that materials are welcome by email. No assessment of the product itself. No promises for the future. Business courtesy without sounding ingratiating.
Practice Differential diagnosis Intermediate

4. Differential diagnosis brainstorming

"I have an unclear case — a 38-year-old patient with recurrent headaches, moderate nausea, mild visual disturbances. I want to be sure I'm not missing an avenue."

Prompt
Please help me with systematic differential diagnosis brainstorming. Case (anonymized): Female patient in her early 30s, recurrent unilateral headaches for about 6 weeks, moderate intensity (NRS 5–6), often in the morning, moderate nausea without vomiting, intermittent short-lasting visual disturbances (no classic scotoma, more like slight blurring). No aura in the strict sense. General condition good, blood pressure within normal range. Family history unremarkable. No trauma. Structure the differential diagnosis by likelihood into three groups: 1. Very common in family practice with this presentation 2. Important to consider 3. Rare, but must not be missed For each diagnosis: one note on which history question or which examination would strengthen or rule out the suspicion. At the end, mark the two "red flags" that would definitely require neurological / imaging workup. Important: You are my sparring partner, not my decision-making authority. I verify everything.
Private Family Beginner

5. Birthday speech for the father-in-law

"My father-in-law is turning 75. I have to give a short speech — and can't find a way to start."

Prompt
I need a short, heartfelt speech for my father-in-law's 75th birthday. Here are a few notes about him: - His name is Hans, born in 1951 - Was a carpenter for 40 years with his own workshop, now retired - Loves his garden, repairs everything himself - Three grandchildren, I'm the son-in-law (for 18 years) - Famously a man of few words, warm-hearted, very reliable - Occasion: 50 guests, banquet hall, a somewhat more formal setting Write a speech of exactly 90 seconds (approx. 220 words). Tone: warm, quietly humorous, with a small, honest anecdote (you can invent a plausible one that fits his character — I'll adapt it). Three sections: 1) Welcome & why I'm speaking today, 2) one specific quality I admire in him, 3) toast. Avoid kitsch, Bible quotes, pathos. Make it authentic, not a "formal address."
Private Finances Beginner

6. Prepare for an insurance comparison

"I have three private health insurance offers on my desk. I want to understand them before I talk to the broker."

Prompt
I have three private health insurance offers to compare (PDFs attached). Create for me: 1. A compact comparison table with the most important benefit points: - Deductible - Hospital (single/double room, chief physician) - Outpatient (visual aids, alternative practitioners, preventive care) - Dental (preservation, replacement, orthodontic treatment) - Premium refund if no claims are made - Switching/cancellation terms 2. Three critical questions I should ask the broker to make differences visible. 3. Three points I should pay special attention to (overlooked cost traps, switching risks, age-related premium surcharges). Write in German, factually, without marketing language. No recommendation for or against any provider — I decide. At the end, add a short note, if relevant, on what only an insurance broker can answer with binding effect.
Private Learning Beginner

7. Revive a language

"I learned Spanish at university. I want to reactivate it on the side without an app — 10 minutes a day."

Prompt
You are my personal Spanish coach. My level: about B1, with clear gaps in the subjunctive and conditional forms. Today's session (10 minutes): 1. Greet me in Spanish in two sentences. 2. Ask me three short questions about my day — I'll answer in simple sentences. 3. Correct me kindly when I get something wrong. Don't just show the mistake, but briefly explain the rule. 4. At the end: three new vocabulary words that fit the conversation, in a small table (Spanish — English — example sentence). If I answer in English, that's fine. Then correct me kindly and ask the question again in Spanish. Be patient but consistent.
Practice Staff Intermediate

8. Draft a job posting for a medical assistant

"I need a new medical assistant and can't find one. Maybe it's also because of my old ad."

Prompt
I'm looking for a medical assistant (MFA) for my family practice in Berlin-Charlottenburg (two doctors, approx. 1,000 patient billing cases per quarter). We need someone full-time, ideally with experience in blood draws, ECG, minor wound care. What we offer: fair pay above the standard rate, four-day week possible, regular working hours (8 a.m.–5 p.m., no on-call duty), permanent contract, annual continuing education allowance, good team, state-of-the-art practice organization. Write a job posting that works in four channels: 1. A detailed version (300 words) for the practice website and Indeed 2. A short version (120 words) for an Instagram carousel 3. Three bullet points (20 words each) for LinkedIn 4. A WhatsApp text (max. 80 words) that I can send to acquaintances to forward Tone: respectful, honest, not sensational. A slight Berlin flavor in individual phrases is fine, but subtle. Never "go-getter mentality" or "rockstar MFA" or similar clichés. We're looking for people who stay long-term.
Private Family Beginner

9. Think through a parenting question

"My son (11) wants a smartphone. I'm unsure. I need clarity before we talk about it."

Prompt
My 11-year-old son wants a smartphone because "all his friends have one." I'm undecided. Help me think this through in a structured way: 1. What is the current state of research (as impartially as possible) on smartphones for 11-year-olds? Three arguments for, three against. Please nothing dogmatic. 2. What intermediate options exist? (e.g. flip phone without internet, borrowed device at fixed times, smartwatch …) 3. If I say yes, which five rules would education researchers typically recommend so it doesn't get out of control? (screen time, bedroom, app selection, social media, emergencies) 4. Three sentences I can say to my son if I say yes. Three sentences I can say if I say no. Both in a way that makes him feel taken seriously. Tone: calm, balanced, no scaremongering. No clichés like "screens are destroying the youth."
Practice Research Intermediate

10. Summarize a guideline PDF

"I downloaded a 60-page DGIM guideline. I need it as an update in 5 minutes."

Prompt (attach the PDF to the chat)
Attached is a guideline as a PDF. I'm a family doctor and need a practice briefing, not a scientific abstract. Please deliver it in exactly this structure: 1. **Core message in one sentence** — what has changed compared to the previous version? 2. **Top 5 recommendations for everyday family practice** — what do I concretely do differently starting Monday? 3. **Three risks / weaknesses of the guideline** — what is controversial, where does uncertainty remain? 4. **A "cheat sheet" table** with the most important dosages, values, or timings. 5. **Five typical questions patients will now ask me more often** — each with a 2-sentence answer building block. Stick strictly to the contents of the guideline. If something isn't in it, say so explicitly instead of guessing. Maximum length: 700 words.

Common beginner mistakes and how to fix them

"It produced nonsense." → You gave too little context
When Claude answers generically, role, target audience, or tone is usually missing. Add it afterwards. Example: "That was too general. Write it for my patient group aged 70 and over, in plain language, with three concrete tips."
"It was too long." → Set a word limit
Claude tends to answer at length. Say: "Maximum 150 words." Or: "In exactly three sentences." It works very reliably.
"It sounds like AI." → Say what it should not sound like
Negative examples help. "Don't sound like a PR agency. Avoid words like 'innovative', 'sustainable', 'forward-looking'. Write the way one person talks to another."
"The facts were wrong." → For medical facts, enable web search or prompt differently
For current guidelines, dosages, studies, always enable web search in Claude (icon in the input field) and ask for sources. When in doubt: "If you're not sure, say so. Don't invent anything."
"It assumed too much." → Stop it and ask
You can interrupt Claude during an answer ("Stop.") and re-brief. Or say right at the start: "Before you answer, ask me three clarifying questions."

Iteration: 70% in one minute → 95% in ten

Don't expect the first answer to be perfect. It rarely is. But: with 4–5 short follow-up comments you get extremely close to the goal. Example iteration:

Follow-up comments — just type them in
Follow-up comments that almost always work: — "Shorter. Half the length at most." — "More empathetic in tone, less factual." — "Fewer technical terms." — "Make it concrete. Example sentence, not theory." — "Write three versions." — "Write it the way a 75-year-old patient would understand it." — "Give me the last paragraph again in two versions — neutral and warmer." — "What did you assume that I might see differently?"
The biggest lever

It's not "the perfect prompt." It's the willingness to iterate in 3–4 steps. Once you internalize that, you have truly understood Claude.

Features · 8 minutes

Projects — your own rooms for recurring topics.

A chat is like a single conversation in an elevator. A Project is like your own consultation room: the same style, the same documents on the table, the same rules — every time you enter it. You don't have to explain everything each time.

DW
Dr. Butze, after three weeks of using Claude
Consultation hours · Wednesday 3:30 p.m.

"Every time, I write to Claude all over again: 'Answer in my practice's style, briefly, on friendly terms with patients, always with a note about follow-up consultation, no jargon.' That gets annoying after the tenth time. There must be a way to store this once."

There is. It's called a Project.

Concept

Chat vs. Project — the decisive difference.

Project

  • For recurring topics ("patient communication")
  • Its own instructions, stored once
  • Its own files (e.g. practice guidelines), Claude always has them "on the table"
  • All chats within it share the same guidelines
  • You open the Project — and start working immediately

Regular chat

  • One-off, isolated questions
  • You have to provide context every time
  • No "personal library"
  • Every chat starts from zero
  • Good for spontaneous things, bad for recurring ones

Important: within a Project you still have any number of individual chats. The Project is not "one single, everlasting conversation," but a frame in which all conversations share the same rules and the same background.

Structure

Three building blocks — that's all it takes.

Project instructions ("ground rules")

A text field in which you permanently tell Claude how to behave in this Project: role, tone, language, length, formats, taboos. This is not a one-off prompt — Claude automatically reads these instructions in every chat in the Project. A good project instruction text is 100–400 words long.

Knowledge base ("library")

Files that Claude has "on the table" in every chat of the Project: your practice guidelines, your own style guide, FAQs, templates, patient education texts, a vocabulary list. With Pro/Max, large files are automatically made searchable (RAG).

Chats ("the actual work")

Every conversation you have in the Project is visibly saved. You can jump back at any time. However, different chats don't share memories — each chat starts "fresh" with the background knowledge from building blocks 1 and 2.

An important clarification.

Project instructions are a special form of "system prompt," just project-specific. They apply in addition to your profile settings (which apply to everything). If both exist, they complement each other. The profile settings say "this is who I am," the project instructions say "in this room I work on X."

Instructions

Create a Project — in 4 clicks.

Left sidebar → "Projects"

In the left column, click the Projects menu item (or go directly to claude.ai/projects). You'll see either an empty overview (normal at the beginning) or your existing projects.

Top right → "+ New Project"

Enter a name ("Patient communication family practice") and optionally a description. Both are only visible to you and have no effect on Claude.

Enter the project instructions

On the Project's main page you'll see an input field "Instructions" or "Set custom instructions". This is where the text goes that we prepared in the templates below — copy & paste.

Fill the knowledge base (optional, but strongly recommended)

On the right side of the Project main page there is an area called "Project knowledge". Use the + icon to upload files (PDF, DOCX, TXT, CSV — up to 30 MB per file). Examples below.

You're in.

Now click into a chat within this Project. Ask a question that fits the context — and watch how Claude automatically takes your guidelines and files into account without you having to repeat anything.

Templates

Five ready-made Project templates for doctors

Each template consists of a name, a project instruction (to copy), and a suggestion for which files belong in the knowledge base.

Work Project #1

🩺 Patient communication family practice

For all patient education texts, patient information, reply emails, and explanatory letters. Claude speaks here the way you would write — if you had unlimited time.

Project Instructions (copy)

Project Instructions
# Role You are my co-author for patient communication in a family practice in Berlin. # Target audience The recipients of my texts are adult patients without medical training. Average age 30–80 years. Mixed education levels. Many read quickly and on the side. # Style & tone - Clear, calm everyday language. No jargon without translation. - Respectful, but not submissive. - Informal address for younger patients only if I explicitly request it; otherwise always formal. - Active instead of passive ("We recommend that you…" instead of "It is recommended…"). - Sentences under 18 words. Better two short ones than one convoluted one. # Content rules - Every text ends with a clear next step for the patient. - With medical recommendations ALWAYS: "If in doubt, please contact the practice first." - Never mention specific dosages without my explicit instruction. - Never "make" diagnoses — explain based on my indication instead. - No promises of cure, no guarantees. # Format Standard length: 120–180 words, unless I explicitly request otherwise. Letters: salutation at the top, closing "Kind regards, your practice team Dr. Butze" — unless I provide a different name. # What you should always do - Ask if medical context is missing. - At the end, briefly flag where you are uncertain about the content and what I should double-check.

Knowledge base — what to upload?

  • Your existing patient education texts (as style examples — anonymize first)
  • A list of "forbidden terms" you never want to see in patient mail
  • Optional: a file with common patient questions and your standard answers
Work Project #2

📚 Continuing education & literature overview

Summarize studies, understand guidelines, compress new recommendations into two paragraphs. Here, Claude becomes your personal, critical editor.

Project Instructions (copy)

Project Instructions
# Role You are my scientific editor and study summarizer for family medicine continuing education. # My background I am a board-certified specialist in general medicine, I read English-language studies without difficulty, but with limited time I only want the core points relevant to my practice. # What I want from you 1. For every study / guideline / publication, a structure of: - "Research question in one sentence" - "Method in three sentences" - "Main result" - "What does this mean for a German family practice?" - "Which limitations do I see as a DOCTOR — not the authors" - "Concrete suggested change in my practice (or: 'No change needed')" 2. ALWAYS cite sources explicitly (author, year, journal). 3. If you add something you did NOT read in the study but supplemented from your own prior knowledge — say so explicitly. Mark it with "[Note from Claude:]". 4. Never present a recommendation as settled when it is controversial in the literature. # Language & format - German. - Technical language allowed, but precise. - Maximum 400 words per study summary. # Self-check before every answer - Did I make a claim I cannot support from the source? → flag or delete. - Would an experienced general practitioner say: "This is relevant to my daily work"? - Did I simplify without simplifying incorrectly?

Knowledge base — what to upload?

  • PDFs of current guidelines (DEGAM, AWMF) — where publicly available
  • A list of your clinical focus areas ("cardiovascular risk", "type 2 diabetes", "early dementia")
  • Your own notes from previous training courses as "vocabulary"
Work Project #3

📣 Practice external communication

Website, Google profile, newsletter, answering machine scripts, job postings, notices — you want to come across as a modern, attentive practice, not like 1998.

Project Instructions (copy)

Project Instructions
# Role You are my marketing and copywriting partner for the external communication of the family practice Dr. Butze in Berlin-Charlottenburg. # The practice in one sentence A modern family practice that takes its time — professionally solid, human, no esotericism, no aggressive treatment promises. # Brand voice - Warm, but professional. - Direct Berlin tone — polite, but not stiff. - No advertising clichés ("We are your health professionals"). - No exaggerations, no superlatives. - Short sentences. Active verbs. # What I need, depending on the request - Website copy (homepage, services, team, directions) - Google My Business descriptions - Newsletter mailings - Answering machine announcements - Notices for the practice - Job postings for medical assistants and trainees - Social media posts (LinkedIn, Instagram) # Content taboos - No case examples involving patients. - No promises of cure. - No comparisons with other practices. - Always keep the German Act on Advertising in the Health Care System (HWG) in mind — ask if you're unsure whether a statement is borderline promotional. # Format - Keep it short. Website block: maximum 80 words. - For longer texts: offer an outline first, I give the green light.

Knowledge base — what to upload?

  • Your current website as a PDF or screenshots
  • A list of practice services, office hours, team
  • Examples of texts from other practices that you like (URL as a note)
  • A short "brand DNA" (10–15 keywords: "time, listening, honest, Berlin, family doctor, local, no esoteric nonsense…")
Private Project #4

🌍 Family life & travel

Vacation planning, weekend activities, birthday planning, family calendar, school communication. The Project area where you're not a doctor but a father or partner.

Project Instructions (copy)

Project Instructions
# Role You help me organize family life — travel, weekends, birthdays, school, play ideas. # Family - Me (52, family doctor, heavily committed at work) - My partner (49, architect) - Son (11, likes Lego, soccer, enjoys cooking with us) - Daughter (8, likes horses, books, doesn't like overly full days) - Two cats (relevant for vacation planning) # Values - We do NOT want to schedule every day down to the minute. - Quality time > program. - Realistic budgets. Mid-range rather than luxury. - We like good food. - We hike, cycle, read a lot, go to museums. # What I want from you - Concrete suggestions with reasoning, not platitudes. - For each suggestion: realistic costs, effort, what can go wrong. - A plan B for bad weather is mandatory. - If you're not sure (e.g. whether a restaurant is still open, whether a hotel is available in season), say so clearly — I'll call then. - Ask if something is missing (dates, budget, trip duration). # Format - Clearly structured (lists, tables where useful). - Never longer than necessary.

Knowledge base — what to upload?

  • A list of the most important family dates in the year (school vacations, birthdays, holidays)
  • A short note with favorite places ("We love South Tyrol, the Baltic Sea, the Black Forest, Italy…")
  • The family's allergies / intolerances
  • A few example photos with daily plans that worked well in the past
Private Project #5

🏠 Authorities, insurance, contracts

Letters to the health insurance fund, tax advisor, insurance company. Understanding contracts. Responding to outrageous demands factually but firmly. The area that drains energy every Sunday evening.

Project Instructions (copy)

Project Instructions
# Role You are my sober, precise co-author for letters to authorities, insurance companies, landlords, tax advisors, banks — all parties I need to write to in a legally clean and polite, but firm manner. # Style - Factual, clear, polite. - Firm without becoming emotional. - Never threatening, never submissive. - Active verbs. Concrete references (date, file number, contract clause). # Structure of a typical letter 1. Brief reference (date, file number, what it's about) 2. The facts in 2–3 sentences 3. What I concretely want (state a deadline if relevant — usually 14 days) 4. Polite closing # What you ALWAYS do - Point out factual errors if I make them in my own description. - Ask me whether I'm aware that I'm threatening legal action — if the letter is heading in that direction. - Remind me that YOU ARE NOT A LAWYER and cannot give legally binding advice when things get tricky. - Suggest seeking legal advice when money >500 € or open disputes are involved. # What you NEVER do - Threatening gestures, all-caps outrage, exclamation mark inflation. - Make claims I did not state in my request. # Format - Standard length: 150–250 words - Structured like a DIN 5008 business letter

Knowledge base — what to upload?

  • Your current letterhead / address
  • A list of the file and contract numbers you regularly deal with
  • A "hall of fame" of your own letters that once worked very well
Lessons learned

Five rules of thumb for good Projects

1. Better 5 specific ones than 1 big one

One Project per clearly delimited context. "Practice communication" and "family" don't belong in the same Project, otherwise tone and style blur.

2. Project instructions are a living document

Update them every 3–5 chats. Whenever you notice: "I don't like this" — add it to the instructions. It keeps getting better over time.

3. Prohibitions are often more effective than commands

"Never jargon without translation," "Never longer than 200 words" — such negative rules often work better with Claude than positive guidelines.

4. Keep the knowledge base lean but current

10 clean, well-named files are worth more than 50 chaotic ones. Tidy up every 1–2 months.

5. Never identifiable patient data

Not in the knowledge base either. Create example files with fictional people. Details in the data protection chapter.

6. Pro / Max / Team supports RAG

With Pro+, Claude can find what's currently needed even in very large knowledge bases. On Free, the knowledge base is smaller — still enough for beginners.

Features · 12 minutes

System prompts — onboard Claude, once and for all.

A system prompt is the "job description" for Claude. A kind of onboarding letter to a new employee that tells him: "This is how you work for me. This is the role. These are the rules. This is how I speak. You never do this. You always do that." Written once — permanently effective.

DW
Dr. Butze, after a month of Claude
Wednesday evening · Study

"There are two kinds of answers I get from Claude. Answers I can use immediately — and answers I first have to rewrite into 'my language.' I want there to be only answers of the first kind."

That is exactly what system prompts are made for.

Concept

Three types of instructions — and why that matters.

Profile settings

Apply everywhere. Here you tell Claude who you are and how you work — regardless of the topic. Example: "Write concisely, avoid clichés, I'm a doctor".

Project Instructions

Apply only within one Project. Here you tell Claude how to behave for a specific topic — e.g. "patient communication". We covered this in the Projects chapter.

Chat prompt

Applies only to this one chat. This is your actual question. It supplements the two levels above without overriding them.

These three levels stack.

Claude always reads all three — profile first, then Project, then chat. This gives you the best of all worlds: permanent ground rules + context-specific guidelines + spontaneous request.

Profile settings

The profile settings — your universal system prompt

In the settings at the top right → "What should Claude call you?" and "What should Claude know about you?". This is where the most important text you will write goes — it applies to every conversation, every Project, every context. A ready-made, well-functioning template:

Profile setting · applies everywhere
# Who I am I am a board-certified general practitioner with my own family practice in Berlin. 52 years old, family with two children, interested in technology but not an IT expert. I use Claude for professional tasks (communication, research, practice organization, marketing) AND privately (family, travel, authorities, hobbies). # How I want to be addressed - In German, formal address, calm and factual. - Not effusive. No "Great question!" at the start. - Straight to the point. # How you write for me - Short, clear, structured (short paragraphs, bullet lists where useful). - Active verbs, short sentences. - Answer the actual question first, then the reasoning — not the other way around. - If you're unsure about a fact, say so openly. Don't invent anything. - If something is medically sensitive, remind me that my professional review is mandatory. # What you never do - Invent or presume patient data. - Cite current studies if you don't know whether they're real — instead say "I have no reliable sources on this." - Pay me compliments when I get no answer out of them. # Format standards - Unless I request otherwise: maximum 300 words. - For longer tasks: offer an outline first, only then work it out.
A common beginner mistake.

Do not put project-specific things in the profile settings, like "You are a marketing copywriter for my practice." Otherwise that also applies when you ask for restaurant suggestions — and Claude will answer in a confused way. Project-specific content goes into project instructions, not into the profile.

Anatomy

Seven building blocks that make up every good system prompt

You don't have to use all seven. But the more of them you cover, the more precise Claude's behavior becomes. Order in the system prompt: in this sequence.

Role & purpose

"You are…" — a clear functional description. "You are my co-author for patient education texts in a family practice." Avoid overly generic roles ("You are a helpful assistant" doesn't help Claude).

Target audience / context

Who is Claude actually writing or thinking for? "The recipients are patients between 30 and 80 years old, mixed education, many read quickly and on the side." Without this point, the wrong tone is the most common result.

Tone & style

Concrete and enforceable. Instead of "friendly," better "polite but direct; active verbs; sentences under 18 words; no advertising clichés". Better 5 specific rules than 20 vague ones.

Content rules (dos)

What Claude should always do content-wise. "Every text ends with a clear next step for the patient." "With medical recommendations, always add 'if in doubt, please contact the practice first.'"

Content taboos (don'ts)

Often more important than dos. "Never specific dosages without my explicit instruction." "Never make diagnoses." "No promises of cure." "Never jargon without translation."

Format standards

Length, structure, language. "Standard length: 120–180 words, unless I explicitly request otherwise. Letters: salutation, body, closing 'Kind regards, your practice team Dr. Butze'."

Self-check before every answer

A small "pre-flight checklist" that Claude mentally runs through before every answer. "Did I make a claim I cannot support from the source?", "Did I use jargon without translation?", "Does the text end with a next step for the patient?" — Three such questions noticeably improve quality.

Meta trick

Claude writes your system prompt — for you.

You don't have to formulate all of this yourself. One of the most elegant applications: Let Claude draft the system prompt for you. The following meta prompt — just paste it into a regular chat — delivers a tailored proposal.

Meta prompt · Claude drafts your system prompt
I want to write a system prompt with your help, which I will then use in the project instructions of a new Claude Project. My use case: [HERE describe in 2–4 sentences what you will regularly do in the Project. Example: "I want to write 3–5 patient education texts per week there, all in plain language, all with a note about follow-up consultation. The recipients are family practice patients between 30 and 80, mixed education, many with a migration background."] Please first ask me 5–8 targeted follow-up questions that I need to answer so you can draft the system prompt for me. One question per topic area: - Role/purpose - Target audience - Tone/style - Content dos - Content taboos - Format - Self-check - Other Don't ask generic questions. Ask questions whose answers will concretely appear in the final system prompt. Only after I have answered all questions do you deliver the complete system prompt in Markdown format — with clearly numbered sections, copy-paste ready.
This is exactly the power of Claude.

You provide the context, Claude leads the conversation. After 5–10 minutes of dialogue you have a system prompt that fits you — much better than any generic template. If it's not perfect, just say: "Tighten point 3, it's too soft for me. Add a point about X." Iteration is part of the method.

Library

Six ready-made system prompts

Each one is ready to use immediately. Copy, paste into your profile or a Project, adjust slightly in one place — done.

Work Patient education

Patient education text assistant

For all patient information, package insert explanations, procedure consent information, and written explanations.

System prompt · for a Project or as a chat template
# Role You are my co-author for written patient education texts of a family practice. # Target audience Adult patients, 30–80 years old, mixed education levels, many read under stress or on the side. They should be able to read the text once and have understood the essentials. # Tone & style - Calm, warm everyday language. - Active verbs ("You take…", "we recommend…"). - Maximum 18 words per sentence. - No technical terms without an immediate translation into simple words. - Consistently formal address. # Content dos - Opening: a single sentence explaining what this is about. - Middle: what, why, how (in this order). - Name side effects / risks clearly, but don't dramatize. - Closing: a concrete next step for the patient. - ALWAYS the sentence: "If in doubt, please contact the practice first." # Content taboos - No specific dosages without my instruction. - No treatment recommendations that are controversial in the literature, without noting this. - No diagnosis statements ("You have…") — explanatory instead ("the examination shows…"). - No promises of cure. # Format - 120–250 words. - Clear paragraphs, no continuous block of text. - No bullet points for simple explanations — bullet points only for multi-part steps. # Self-check before output 1. Would my 78-year-old mother understand this on first reading? 2. Does it say "diagnosis" or "guarantee" anywhere? → correct. 3. Does the text end with a clear next step? 4. Did I translate the jargon?
Work Referral letter

Referral letter drafting helper (fully anonymized)

Important: this prompt works exclusively with fictional or anonymized cases. Real patient data doesn't belong in Claude (except Claude for Work with a data processing agreement). Read the data protection chapter.

System prompt · referral letter help
# Role You are my co-author for physician consultation reports and family doctor referral letters. We work exclusively with fictional or fully anonymized cases — I will not give you any real patient data. # Style - Factual, precise, concise. - Classic referral letter structure: salutation, reference, history, examination, plan, recommendation, closing. - No jargon showcasing — write the way a well-organized family doctor would phrase it for a colleague or a specialty clinic. # Content rules - If I only give you keywords, propose a complete structure and ask specifically what's missing. - If clinical information appears incomplete, flag it with "[Missing:]" at the end. - Never insert diagnoses I did not state. - Never fabricate findings that are not in my request. # Format - Complete letterhead-ready text. - About 300–500 words. - Plan always as a bullet list at the end. # Taboos - Never use real names, dates of birth, addresses, insurance numbers — even if I enter them (in that case, explicitly double-check: "Are you sure these are fictional data?"). # Self-check - Did I write anything anywhere that sounds like a real patient? → Warn. - Is every finding I mention present in the user's request? - Is the letter directly usable for a receiving physician?
Work Research

Literature & guideline summarizer

For weekend continuing education. You provide a study or guideline as a PDF / text / link, Claude delivers the GP-relevant summary in half a page.

System prompt · study editor
# Role You are my scientific editor. You read studies, guidelines, reviews and deliver GP-oriented summaries. # My focus Family medicine / general practice. Relevant topics: cardiovascular prevention, type 2 diabetes, early dementia, polypharmacy, primary care research, vaccination recommendations, pain management. # Mandatory structure of every summary 1. **Research question in ONE sentence** 2. **Method in a maximum of three sentences** 3. **Main results** (bullet points, no more than 5) 4. **What does this mean for my family practice?** (3–5 sentences) 5. **Limitations from a GP's perspective** (what the authors don't see because they aren't GPs) 6. **Concrete suggested change**: either "Changes nothing" or a specific adjustment 7. **Source citation** (author, year, journal, DOI if available) # Content rules - If I mention a study you don't know, say so openly: "I cannot reliably summarize the exact content of this study." - Never speculate about content you haven't read. - Mark all statements that come from your general knowledge (not from the study) with "[Note from Claude:]". - Explicitly flag studies with major limitations (small sample, pharma sponsorship, no primary care setting). # Format - German. - Technical language allowed, but use it precisely. - Maximum 400 words per summary.
Work Marketing

Practice marketing copywriter

Website, Google profile, notices, job postings, answering machine scripts. HWG-aware, without advertising clichés.

System prompt · practice marketing
# Role You are my copywriter for the external communication of a family practice in Berlin-Charlottenburg. The practice takes its time, is professionally solid, human, no esotericism, no aggressive promises. # Brand voice - Warm, but factual. Berlin-direct, but not brusque. - Short sentences, active verbs. - NO advertising clichés ("We are your health experts" → taboo). - No superlatives. - No comparisons with other practices. # Use cases - Website copy (homepage, services, team, directions, FAQ) - Google My Business descriptions - Practice notices (locum coverage, changes to office hours) - Job postings for medical assistants - Answering machine announcements - Social media posts (LinkedIn, Instagram) - Newsletter # Content taboos - No treatment promises ("We'll make you pain-free" → taboo). - No before/after comparisons. - No case examples involving patients — not even "modified" ones. - Always observe the German Act on Advertising in the Health Care System (HWG) — for borderline cases, check with me. # Format - Website block: maximum 80 words per section. - Job postings: 200–300 words, with a clear structure "Who we are", "Who we're looking for", "What we offer", "Contact". - Notices: maximum 60 words, one central message. # Self-check - Does this sound like a modern, calm practice — or like an ad agency? - Would this be problematic in an HWG review? - Would my patient Ms. Schmidt (68) be convinced by this — or embarrassed?
Private Travel

Family travel planner

For vacation planning with the family, weekend trips, travel logistics.

System prompt · family travel planner
# Role You are my travel planner for a family of four. # Constellation - Me (52, family doctor, gets slightly tired when the program is too full) - My partner (49, architect, likes calm and good food) - Son (11, likes active days, soccer, Lego, open to culture if it's compact) - Daughter (8, likes horses, books, gets whiny with more than two activities/day) - Two cats at home (pet care must be factored into the planning) # Values - Better two "empty" days for drifting than seven full ones. - Quality > quantity. - Mid-range hotels with character > luxury. - We love good food, hiking, cycling, water, museums — but in moderation. # What you do for me - Concrete suggestions with reasoning. No platitudes. - For each suggestion: realistic costs + effort + risk factor. - ALWAYS include a plan B for bad weather. - If you're not sure about availability / opening hours / the current situation, say so clearly — I'll call then. - Ask about anything I haven't specified (dates, budget, mode of travel, allergies). # Taboo - No invented restaurants or hotels — only ones you believe actually exist. - If you're not sure: better to give a search recommendation than to invent a "suitable hotel." - No price guesses that sound like guidebook level — only ranges ("typically 80–120 €"). # Format - Clearly structured: overview, then detail. - Tables for daily plans where useful. - Maximum 600 words per answer, otherwise outline first.
Private Authorities

Letters to authorities & insurance companies

Factual, precise letters — polite, but firm. Never threatening, never submissive.

System prompt · letters to authorities
# Role You are my co-author for letters to authorities, insurance companies, landlords, tax advisors, banks. # Style - Factual, clear, polite. - Firm without becoming emotional. - Active verbs. Concrete references (date, file number, contract clause). - Never threatening. Never submissive. - No exclamation mark inflation. No all-caps outrage. # Standard structure 1. Brief reference (date, file number, subject) 2. The facts in 2–3 sentences 3. What I concretely want (with a deadline, usually 14 days) 4. Polite closing # Mandatory behavior - Immediately point out factual errors in my own description. - If the letter is heading toward escalation (lawyer, lawsuit), explicitly point out that I should seek legal advice. - Remind me that YOU ARE NOT A LAWYER when things get tricky. - For amounts > 500 € or open disputes, advise me to consult a lawyer. # Taboos - Making claims I did not state in my request. - Insulting or ironic phrasing — even if I write emotionally. # Format - DIN 5008 business letter - Standard length: 150–250 words - Salutation: "Dear Sir or Madam," if I don't provide a name; "Dear Ms./Mr. [Name]," if I do - Closing: "Kind regards" + my name
Refinement

System prompts are never finished.

Your first system prompt will be okay. The third will be good. The tenth will surprise you. Three tips for iterating:

1. Note what annoys you

Every time you have to rewrite an answer — briefly note what didn't fit. After 5 such moments you have 5 new rules for your system prompt.

2. Prohibitions are often more effective than commands

"Never use the subjunctive" works better than "Write concretely." Negative anchors hit Claude more precisely.

3. Claude improves its own prompt

Say: "Here is my current system prompt. Read it, then show me five concrete weaknesses — and make one improvement suggestion for each."

You now have Claude's most powerful tool.

System prompts are what make the difference between "Claude is a nice toy robot" and "Claude is my most reliable staff member." Every good workflow stands or falls with a good system prompt.

Features · 8 minutes

Artifacts — when Claude builds an attachment.

Sometimes Claude answers in the chat. And sometimes a second window opens to the right with a finished document, a checklist, or a diagram — cleanly formatted, downloadable, further editable. Those are Artifacts. They change how you work with Claude.

DW
Dr. Butze, Thursday afternoon
Practice · between two patients

"I asked Claude to create a checklist for the U9 well-child checkup. Suddenly a separate document appears to the right of the chat — formatted, with checkboxes, ready to download. That was new. And exactly what I needed."

Basic concept

Think of Artifacts like an attachment in an email.

In the chat window you talk to Claude — like in a chat. But when Claude hands you a document — a checklist, a letter, a table, a diagram — it doesn't land in the conversation but in a separate window to the right. That is the Artifact.

Why two windows at all?

So the document stays readable. A twenty-page letter in a chat line would be illegible. A diagram in plain text would be nonsense. Artifacts are Claude's way of saying: "This is no longer a conversation — this is a result. Look at it, change it, download it."

Chat answer

Short, conversational, gets lost in the conversation. Three paragraphs about the pros and cons of a therapy. You read, reply, move on.

Artifact

Standalone document. You can copy it, download it, forward it — and in the next message say: "Make the section on aftercare longer." Claude changes only that section.

Trigger

When Claude creates an Artifact — and when not.

Claude decides on its own. But there are clear patterns. If your request has one of the following characteristics, the chance is high that an Artifact is created:

Length

The result will be longer than ~15 lines or 1,500 characters. Example: "Write a two-page patient letter."

Structure

You ask for a clearly structured document — checklist, table, letter, report. Example: "Create a checklist for the hygiene training."

Reuse

You want to reuse the result later, copy it, or save it as a file. Example: "Make a template for the patient consent form."

Visualization

You want a diagram, a table, or a process plan. Example: "Show me the treatment process as a flowchart."

You can also request Artifacts directly.

Simply write: "Make this as a document." or "Create this as an Artifact to download." Claude understands that.

Types

Five types of Artifacts that are relevant for you.

Text document

Letters, reports, patient education texts

This is the most common type. Cleanly formatted with headings, paragraphs, lists. You can download it as Word or Markdown, copy it, reuse it.

Table

Structured lists, comparisons

When you compare or list something — drug interactions, therapy options, diagnostic differentiation — a table often results. Readable, copyable into Excel.

Diagram

Flowcharts, process plans

"Show me the U9 checkup process as a diagram" — Claude builds a visual schema that you can print or embed in a presentation.

Small app / tool

Calculators, quizzes, interactive aids

Claude can also build simple interactive tools — for example a BMI calculator for the practice website or a small symptom self-assessment tool. Usable directly in the browser.

Web page / layout

Simple HTML pages, templates

For your own practice website or a letter with letterhead — Claude can output a finished HTML page. You see the page rendered on the right, can test it and download it. That's the pro level; text Artifacts are entirely sufficient to start with.

Six templates

Artifact prompts to take with you right away.

Copy, paste, adapt. All prompts reliably trigger an Artifact because they are structured enough.

Work Text document

1 — Patient info sheet for diagnosis XY

A one-page info sheet you can give to patients with a new diagnosis. Understandable, reassuring, without jargon.

Copy prompt
Create a patient info sheet on the following topic as a separate document (Artifact): Topic: [Enter DIAGNOSIS / CONDITION, e.g. "mild high blood pressure"] Requirements for the document: - One page, A4 - Headings for each section - Clear, simple language (8th-grade level) - NO subjunctive, NO scaremongering - Address patients formally Content structure: 1. What is [diagnosis]? (3 sentences) 2. What does it mean for me? (everyday life, symptoms) 3. What can I do myself? (3-5 concrete measures) 4. When do I need to come to the practice? (warning signs) 5. Brief encouragement at the end Important: No dosage or medication information. I discuss these personally with the patient.
Always review professionally before handing out.

Claude can write something wrong or imprecise even with the classics. You remain responsible for the medical content.

Work Checklist

2 — SOP checklist for the practice

A reusable checklist — for example for the monthly hygiene training, the emergency equipment, or restocking the examination rooms.

Copy prompt
Create a practice checklist as a document (Artifact) for the following purpose: Purpose: [e.g. "Daily final check of the examination room before closing"] Format: - Heading with purpose and date field - Section "Preparation" (3-5 points) - Section "Execution" (5-10 points) with checkboxes - Section "Completion" (3 points) - Signature field: "Performed by: _______ Date: _______" Style: - Concise, one line per point - Concrete, verifiable actions (no "if applicable" or "should") - Printable on A4
Work Diagram

3 — Treatment pathway as a flowchart

A visual process — good for patient communication, staff training, or as a notice in the waiting room.

Copy prompt
Create a flowchart (as an Artifact, Mermaid format) for the following process: Process: [e.g. "Approach for suspected uncomplicated urinary tract infection in family practice"] Requirements: - Start with "Patient presents with complaint XY" - Show decision points as diamonds ("Fever? Yes/No") - Maximum 8-10 nodes — otherwise it becomes illegible - End points clearly marked ("End: outpatient therapy", "End: referral") - Labels short, max. 6 words per node Important: This is a rough orientation aid, not a substitute for clinical judgment. Please add a corresponding disclaimer at the end.
Private Text document

4 — Packing list for the family vacation

A complete packing list sorted by person — created once as an Artifact, adjusted once, and it's set for the next ten vacations.

Copy prompt
Create a packing list as an Artifact document for our family vacation. Trip: 10 days in South Tyrol, early August, hiking vacation, vacation rental with kitchen. We're traveling by car. Family: - Adult man (52), likes hiking, needs reading glasses - Adult woman (49), architect, yoga - Boy (11), soccer enthusiast - Girl (8), reads a lot, mild pollen allergy Please organize the list by person, plus a separate block "Shared" for car, kitchen, hiking, first aid, documents. For each item: a checkbox to tick off. Format: One document, everything printable on one page.
Private Table

5 — Weekly menu planner as a table

Seven days, three meals, with a shopping list at the end — exactly what's annoying on Sunday evenings.

Copy prompt
Create a weekly menu plan as a table (Artifact) for a family of four (two adults, two children aged 8 and 11). Table: - Columns: Day | Breakfast | Lunch | Dinner - Seven rows (Monday to Sunday) - Lunch: warm, max. 30 minutes preparation - Dinner: light, often cold (bread, salad) - At least two meat-free days - No exotic ingredients — everything available in a regular supermarket Below the table: A consolidated shopping list, sorted by supermarket sections (fruit/vegetables, dairy/cheese, bread, meat/fish, dry goods/canned).
Private Mini app

6 — Small BMI calculator for the practice website

A simple, interactive tool that you can embed on your practice website. Patients enter height and weight and see the result immediately.

Copy prompt
Build me a simple BMI calculator as a small, standalone tool (Artifact, HTML/JavaScript). Requirements: - Two input fields: height in cm, weight in kg - A "Calculate" button - Output: BMI value (one decimal place) + WHO classification (underweight / normal weight / overweight / obesity class I-III) - Clear, simple design — no background image, subtle colors - Disclaimer below the result: "This is an orientation aid. For medical questions, please contact your doctor." Works in the browser without an internet connection. One HTML file, everything inline.
You don't need to be a programmer.

Claude writes the code. You see the finished tool directly in the Artifact window, can try it out and download it. Your webmaster or practice service provider gets the file and embeds it.

How to work

Working with Artifacts — editing, sharing, saving.

  1. Improve iteratively

    Just say: "Make the 'Aftercare' section longer." or "Remove the medical vocabulary from point 3." Claude changes only what's requested, the rest stays in place.

  2. Download or copy

    Via the menu at the top right of the Artifact: download as Markdown, Word, PDF. Or put the entire text into the clipboard with one click and paste it into your practice management system.

  3. Share with a link

    You can also send Artifacts to colleagues via a link. Careful: the Artifact is then publicly visible to anyone with the link — never share patient-related content.

  4. Reference it in the chat

    You can keep referring to the Artifact during the ongoing conversation: "Write a shorter version of the patient letter for our email reply." Claude knows which document is meant.

Caution

Three pitfalls you should know about.

This is how it works well

  • Say clearly what should be in the document — structure, length, target audience.
  • Have it changed iteratively ("rewrite only section 2").
  • Read everything through before sending.
  • Anonymize patient data before entering it.

This is how it goes wrong

  • Forwarding the Artifact unreviewed to the patient inbox.
  • Trusting "it's probably right" for medical content.
  • Sharing Artifact links carelessly (public!).
  • Mentioning real patient names in the prompt.
Artifacts are not a GDPR safe space.

Everything you write in the prompt — even if the result ends up in the Artifact — is sent to Anthropic's servers. Patient data doesn't belong there. For details, see Data protection & limits.

You now know Claude's most important productivity feature.

From here on it gets concrete: on the next pages you'll see ten professional and ten private use cases — each with a ready-made prompt.

Use cases · 20 minutes

Ten use cases for the practice.

From the patient education text to the letter to the authorities — ten tasks that come up in every family practice, each with a copyable prompt. Skim through, pick out what you need today.

Before you get started: three iron rules.

1. No real patient names, addresses, dates of birth, insurance IDs, or findings in the prompt. 2. Review professionally before every send — Claude can be professionally wrong. 3. No medical data on the Free/Pro/Max plan. Details: Data protection & limits.

DW
Dr. Butze, three weeks with Claude
Friday · 4:45 p.m. · last patient gone

"I used to spend two more hours writing letters after closing. Now I write a prompt, read the answer, change two sentences. Ten minutes. The rest of the evening I see my kids."

Work Quick Win · 5 min

1 — Rewrite a patient education text in patient-friendly language

The official consent forms from the professional associations are legally bulletproof — but incomprehensible to many patients. Claude turns a technical text into a patient version that you attach as a supplement.

"I can't spend twenty minutes explaining to every patient what a colonoscopy is. Now everyone receives, in addition to the formal consent form, a one-page comprehension sheet — and the conversations become shorter and better."

Prompt to copy
I'm about to give you a technical patient education text about an examination/procedure. Please write a supplementary patient version from it with the following properties: - One A4 page, clear paragraphs - Language level: 8th grade - Address patients formally - No technical abbreviations without explanation - No scaremongering, but honest about risks - Avoid subjunctives ("would", "could" → "can") Structure: 1. What is this about? (3-4 sentences) 2. How does the examination work? (step by step) 3. What should you do/avoid beforehand? 4. What happens afterwards? 5. What risks exist? (honest, brief) 6. When should you get in touch? Please output as an Artifact so I can save it as a file. Here is the original text: [INSERT TECHNICAL PATIENT EDUCATION TEXT]
Important: not a replacement, but a supplement.

The patient version does not replace the legally required original consent form. It is a comprehension aid — and that should also be clearly noted on the page.

Work Quick Win · 3 min

2 — Answer a patient email politely and clearly

Patients often write long and emotional messages. You reply briefly and factually. Claude bridges this gap without sounding cold.

"I decide what the content is. Claude makes sure it doesn't sound like an administrative notice."

Prompt to copy
Please draft an email reply to a patient. Style: - Friendly, but factual - Avoid informal address, address the patient formally - My sender: "Practice Dr. Butze — your family doctor team" - Salutation appropriate to whether a name is present - No diagnoses or treatment recommendations by email - Closing sentence always: "For questions, you can reach us by phone at 030-XXXXXXX (Mon-Fri 8 a.m.-12 p.m.)." Content-wise, the reply should express the following: [BRIEFLY IN KEYWORDS WHAT SHOULD BE SAID — e.g.: "Patient asks for a prescription renewal for blood pressure medication. We need an appointment for a follow-up check because the last blood pressure reading is over 6 months old. Suggestion: call next week between 8 a.m.-12 p.m., then appointment."] Important: Do NOT use patient name, date of birth, diagnosis in the prompt. Keep the reply general so I insert the salutation and name myself.
Work Quick Win · 2 min

3 — Politely decline a pharma rep request

Four requests per week. Answering each one individually costs time. With a Project and a system prompt, a matter of thirty seconds.

Prompt to copy
Write a short, polite refusal to a pharmaceutical representative who has requested a practice appointment. Content: - Thanks for the inquiry - Currently no appointments for pharma visits - Please submit inquiries exclusively by email to praxis@... - We decide on appointment allocation quarterly - No escalation, friendly overall attitude Length: At most 6 sentences. Salutation: "Dear Ms./Mr. [Name]" (leave placeholder) Closing: "Kind regards, Practice Dr. Butze"
Tip: Set it up as a Project.

If this comes up more often — which is usually the case — create a Project "External communication" with a system prompt. Then in the chat, this is enough: "Refusal pharma rep." Claude knows the rest. See Projects.

Work Continuing education · 10 min

4 — Summarize a guideline on one page

An S3 guideline often has 200 pages. You have ten minutes. Claude turns the complete guideline into a practice summary — which you cross-check against your memory of the original guideline.

Prompt to copy
I'm uploading a guideline. Please create a one-page summary as an Artifact with the following structure: 1. CORE MESSAGE (max. 2 sentences): What is the most important recommendation? 2. DIAGNOSTICS - Which examinations are mandatory? - Which are optional/supplementary? 3. THERAPY - First line - Alternatives - What has changed compared to older guidelines? 4. AFTERCARE / FOLLOW-UP CHECKS - Intervals - What needs to be monitored? 5. REFERRAL INDICATIONS (to specialist/hospital) 6. RED FLAGS (when to act, not wait?) Style: bullet points, concise, clinically usable. Important: At the end of the summary, cite the guideline's source section with version date and validity period. If any points seem unclear or contradictory to you — please point this out explicitly ("The guideline is not clear here").
Treat Claude's guideline summary like a staff note.

Helpful, but no replacement for the original. For every concrete patient case that depends on it: open the guideline yourself, look up the relevant passage. Claude can misinterpret or omit details.

Work Continuing education · 5 min

5 — Condense a study/article for daily practice

You have a study from the Deutsches Ärzteblatt or the NEJM open. Forty pages. Question: Does this change how I practice on Monday?

Prompt to copy
I'm uploading a study. Please give me a practice assessment with the following points: 1. RESEARCH QUESTION of the study (1 sentence) 2. STUDY DESIGN (RCT / cohort / meta-analysis / observational) 3. MAIN RESULT in absolute and relative numbers 4. PATIENT POPULATION: does this fit a typical family practice patient in Germany? 5. STRENGTHS of the study (max. 3) 6. WEAKNESSES (limitations, bias risks, sponsor effects) 7. RELEVANCE FOR THE PRACTICE: - Should I change something in my treatment? (yes/no/maybe) - If yes: what exactly? - Should we wait for further studies? How many? Closing sentence: An honest assessment — "game changer", "minor refinement", "wait for further studies first", "not relevant for practice". I am a general practitioner in a big-city practice. Factor that into your assessment.
Work Practice marketing · 15 min

6 — Copy for the practice website

Most practice websites read as if they were built in 2008 by the medical assistant's son. With an hour of Claude work, you get a complete body of copy for a professional practice site.

Prompt to copy
Please help me write copy for my family practice's website. Create the texts as an Artifact, clearly organized by page. About the practice: - Solo family practice in [CITY/DISTRICT] - Owner: Dr. med. [NAME], specialist in general medicine - Founded [YEAR], [N] medical assistants - Focus areas: primary care, preventive checkups, chronic conditions, travel medicine - Office hours: [INSERT] Please write copy for the following subpages: 1. HOMEPAGE (max. 150 words) A short, welcoming greeting. Who are we, what do we offer, what sets us apart? No clichés like "at the highest level". 2. ABOUT THE PRACTICE / TEAM (200 words) Introduction of the practice and how we work. No CVs. 3. SERVICES (list with short explanations) Preventive checkups, vaccinations, travel medicine, DMP programs, minor surgery, home visits. 4. OFFICE HOURS & DIRECTIONS Clear, tabular. 5. NEWS / IMPORTANT NOTICES Template for practice closures, locum coverage, vacation. Style: warm but not smarmy, factual, trustworthy. No medical promises of cure. No advertising for individual medications.
Work HR · 10 min

7 — Write a job posting for a medical assistant

Job postings that work don't sound like administration — they sound like a place where people want to work. Claude can do that if you honestly describe what your practice stands for.

Prompt to copy
Please write a job posting for our family practice. Position: Medical assistant (m/f/d), 30-40 hours, permanent, starting [DATE]. About us: - Solo family practice in [CITY] - Patient clientele: [INSERT] - Team: [N] people, small, well-coordinated group - What defines us: [3 HONEST POINTS — e.g.: "we take time for every patient", "we are digitally equipped (KIM, eAU, online appointments)", "there is no overtime — closing time is closing time"] What we offer, concretely: - Pay: according to TVöD-VKA + voluntary bonus - 30 days of vacation - Continuing education budget - Job ticket - No weekend duty - Regulated breaks Desired qualification: - Completed medical assistant training - Experience with [PRACTICE SOFTWARE NAME] - Friendly, resilient, team player Style of the ad: - Direct and honest - No clichés like "dedicated team" without substance - No gendered language that excludes - Length: approx. 250 words - With a concrete application route at the end Output as an Artifact so I can paste it directly into job portals.
Work Quick Win · 3 min

8 — Answering machine scripts for the practice

One script for normal office hours. One for vacation time. One for Wednesday noon. One for emergencies. All written cleanly once, easy to speak, without stumbling sentences.

Prompt to copy
Please draft scripts for our practice answering machine — four different versions, each written so it can be spoken smoothly. General rules: - Short: maximum 25-30 seconds spoken - Plan clear pauses - Phone numbers in digit groups ("zero-one-five-two...") - Always mention 116117 (medical on-call service) - For emergencies: 112 Practice details: - Practice Dr. Butze, [STREET], [ZIP CITY] - Office hours: Mon-Tue-Thu 8-12 and 2-6, Wed and Fri 8-12 - Telephone consultation: daily 11-12 Versions I need: 1. STANDARD (outside office hours) 2. VACATION / PRACTICE CLOSED - Locum: Dr. [NAME] at [STREET], phone [NUMBER] 3. LUNCH BREAK 4. EMERGENCY NOTICE for special situations (e.g. pandemic mode, all appointments by phone) Please give each version a clear heading, with notes on recommended speaking pauses ("[pause]" in brackets).
Work Continuing education · 5 min

9 — Structure continuing education notes

Three days of conference, twenty-five talks, a hundred handwritten notes. Claude turns them into something you'll still understand in four weeks.

Prompt to copy
I'm about to give you my notes from a continuing education event (unstructured, partly abbreviated, partly written half-legibly). Please turn them into a structured continuing education memo as an Artifact: Format: # Continuing education [TITLE] Date: [INSERT], Duration: [N CME points] ## Key insights (max. 5 points) The 5 most important things relevant to my practice. Per point: one line, precise. ## Per talk/topic - Title - Speaker (if in the notes) - 2-3 main statements - "Take home" — what do I concretely implement? ## Open questions / research tasks Points I still need to read up on or clarify. ## Practice actions Concrete things I'll implement after the training: - ☐ ... - ☐ ... If my notes are unclear, make that transparent ("Note unclear — please add from the handout"). Here are the notes: [INSERT NOTES]
Tip: Photo instead of typing.

You can simply photograph your handwritten notes and upload the photo to Claude. Claude reads the handwriting, converts it into text, and structures it according to the scheme above. Works surprisingly well — as long as the handwriting isn't completely illegible.

Work Administration · 10 min

10 — Letter to authorities (KV, building office, health office)

A letter that sounds appropriately formal, is factually precise, and promises nothing you can't keep.

Prompt to copy
Please draft a letter to an authority with the following properties: Format: - DIN 5008 business letter - Standard length: 150-250 words - Salutation: "Dear Sir or Madam," if I don't provide a name; otherwise "Dear Ms./Mr. [Name]" - Closing: "Kind regards" + Practice Dr. Butze Style: - Factual, polite, without clichés - Concrete data and facts first - Clearly worded request/question at the end - No emotional arguments - Reference line ("Your letter dated...") where applicable Recipient: [AUTHORITY] Occasion: [DESCRIBE CONCRETELY — what do you want to achieve, what deadline is there, which enclosures are included] If applicable, file number: [INSERT] If enclosures are required, list them below the letter. Output as an Artifact with a letterhead placeholder at the top.
You now have ten templates for everyday practice.

Bookmark this page. The first time, each case may take ten minutes — by the third time it's three. The time you gain is worth every learning curve.

Use cases · 20 minutes

Ten use cases for private life.

Travel planning, letters to authorities, homework help, birthday poems, weekly menus. Ten everyday situations in which Claude gives you back two evenings per week — each with a ready-made prompt.

DW
Dr. Butze, Sunday morning
Living room · wife at the table, kids in the garden

"My wife asked me what I was doing on the laptop. I said: 'I'm planning the vacation.' She replied: 'You never plan the vacation.' Forty minutes later we had a complete travel plan for South Tyrol. Since then I've also planned the toy-themed birthday party. And the weekly menu."

Different rules apply in private life.

Here you can think more freely. But: even privately, don't enter bank PINs, tax IDs, passwords, or your family's health data into Claude. That remains a hard line — see Data protection & limits.

Private Quality of Life · 30 min

1 — Plan a complete family vacation

From destination to daily schedule. Claude delivers a complete draft, which you then adapt to your preferences.

"I hate travel planning. My wife hates that I hate travel planning. Claude solved our marital problem."

Prompt to copy
Please plan a family vacation for us. Result as an Artifact: a structured travel plan I can show my wife. Family: - Me (52, family doctor, like hiking) - My wife (49, architect, likes art, yoga) - Son (11, soccer fan, my easily combustible counterpart) - Daughter (8, reads a lot, pollen allergy) Trip: - 10 days in early August - Region: South Tyrol - By car from Berlin - Prefer a vacation rental over a hotel - Budget: medium (no luxury, but not stingy) Plan structure: 1. SUMMARY (3-4 sentences) 2. GETTING THERE: route, sensible stopovers, driving time 3. ACCOMMODATION RECOMMENDATIONS: 3 suggestions with region and reasoning (who it suits) 4. DAILY PLAN (10 days): one activity per day, alternating active/quiet, child-friendly and adult-appropriate 5. RESTAURANTS (5 recommendations per region) 6. PACKING LIST HIGHLIGHTS (what we absolutely must not forget for South Tyrol) 7. EMERGENCY INFO (pharmacy note because of the daughter's pollen allergy, without diagnostic details) Important: Please use concrete place names, not "near". If you're unsure about a recommendation (e.g. whether a restaurant still exists) — say so openly.
Be careful with concrete recommendations.

Claude knows many places from its training data — but restaurants close, opening hours change. Before booking, briefly verify online or ask Claude to use web search (feature can be enabled in Claude.ai).

Private Quick Win · 5 min

2 — Weekly menu plan with shopping list

Sunday evening, 7 p.m. The same old question: "What are we cooking next week?" Claude answers it in thirty seconds.

Prompt to copy
Please create a weekly meal plan for our family. Family: 2 adults + 2 children (11 and 8). Rules: - Monday to Friday: warm dinner, max. 30 min preparation - Nobody eats lunch at home except Wednesday and Friday (then I need lunch for 4) - Vegetarian twice a week - "Kids' favorite" once a week (pasta, pizza, etc.) - No exotic ingredients — everything from REWE or Edeka - Sunday: cozy family meal, happy to take longer Output as an Artifact table: - Day | Lunch (only Wed/Fri/Sat/Sun) | Dinner | Main ingredients Below it: consolidated shopping list, sorted by supermarket sections (fruit & vegetables, dairy products, bread/baked goods, meat/fish, dry goods, frozen). If you can build in leftover use (e.g. rice from the day before) — go ahead.
Private Family · 10 min

3 — Homework help for the kids

You're a family doctor, not a math teacher. But your eight-year-old daughter needs help with fractions. Claude explains it the way a good teacher would — patiently, child-appropriate, without doing the work for them.

Prompt to copy
You are a patient elementary school teacher. You explain to my [8-year-old daughter / 11-year-old son] the topic: Topic: [e.g. "Fractions: adding fractions with different denominators" or "English vocabulary: regular verbs in the past tense"] Rules: - Language: simple, friendly, age-appropriate - At most 2-3 sentences per explanation step - After each step, a small practice exercise - Wait for the child's answer (don't keep talking) - For correct answers: praise briefly, then next step - For wrong ones: do NOT reveal the answer, but ask a guiding question - Examples from the world of an 8-year-old (pizza slices, birthday cake, school snack...) Start with a very short introduction ("Today we're looking at ... First, let me ask: what do you already know about it?"). Then start the interactive lesson.
Tip: let the child type themselves.

If you sit next to your child and they type in the answers themselves, they learn not only the material but also good habits with AI: asking follow-ups, not believing everything, trying their own solution first.

Private Quick Win · 5 min

4 — Gift ideas for difficult occasions

Mother-in-law turning 75. A colleague retiring. The eleven-year-old son of an acquaintance has a birthday. Claude doesn't deliver run-of-the-mill ideas, but concrete suggestions if you describe things concretely.

Prompt to copy
I need 5 gift ideas for the following occasion. Occasion: [e.g. my mother-in-law's 75th birthday] Budget: [e.g. 50-100 euros] Gift recipient: - Age, gender - Interests: [concrete — e.g. "reads a lot of crime fiction, has a small city garden, cooks Italian, used to be a teacher"] - What she already has (to avoid) - What she does NOT like (to avoid) - Living situation: [e.g. "lives alone in a rented apartment"] Please: - 5 very different ideas — one experience-oriented, one practical, one beautiful, one homemade, and one "wildcard" idea - Per idea: 3-4 sentences on why it fits - For each idea: approximate price range and where to get it - For homemade ideas: a short implementation sketch No generic ideas like "chocolates" or "flowers".
Private Special occasion · 10 min

5 — Write a birthday speech or poem

Tomorrow you have to give the speech at your parents' golden wedding anniversary. Or a poem for your daughter's birthday. Claude doesn't write for you — Claude delivers a good draft that you bring into your voice.

Prompt to copy
Please write a draft for a [birthday/wedding] speech. Occasion: [e.g. my mother-in-law Anneliese's 75th birthday] Speaker: me (son-in-law) Audience: approx. 30 family members and friends Duration: 3-5 minutes spoken (approx. 400-600 words) About the guest of honor (a few notes): - Raised 3 children, one of whom died at 18 - Former elementary school teacher - Widow for 8 years - Lives in the same house as her daughter and grandchildren - Known for: her laugh, her apple tart, her sharp mind - She and I get along well — we love arguing about politics, then everything's fine again Tone: - Warm-hearted, but not smarmy - One small humorous moment allowed - NO pathos, NO "on life's journey" or "bound in love" - Something personal in it - Honest: difficult topics (the child who died) may be briefly acknowledged without tipping the mood Structure: 1. Opening — short, maybe an anecdote 2. Main part — what makes this person special (with examples) 3. Closing — wishes, an image, a toast Give me the draft. I will then bring it into my own tone.
Private Administration · 10 min

6 — Letter to the tax advisor or insurance company

You want to be clearly understood — without hours of back and forth by email.

Prompt to copy
Please draft an email or letter to my [tax advisor / insurance broker / bank]. Recipient: [insert salutation] Occasion: [concrete — e.g. "Question about the tax deductibility of a continuing education course abroad in 2025"] Content points (in the order in which they should be addressed): 1. [e.g.: Training was in Florence, 5 days, €1,800 costs plus travel] 2. [e.g.: Is that fully deductible or partially?] 3. [e.g.: Which receipts do I specifically need?] 4. [e.g.: Deadline until when? How do I best submit them?] Style: - Polite, but direct - Structured (points 1-4 numbered) - One concrete request at the end (callback? appointment? reply by email?) - Length: max. 200 words No emotional language. No filler phrases at the start (omit "I hope you are well").
No sensitive data in the prompt.

No tax ID, no account balance, no insurance number. Keep it general — you fill in the letter with the concrete numbers yourself before you send it.

Private Quick Win · 5 min

7 — Write a car sale listing

A listing that doesn't look like an "amateur private sale" — but factual, trustworthy, complete.

Prompt to copy
Please write a car sale listing for mobile.de. Vehicle: - Make / model / type: [e.g. BMW 3 Series Touring, F31, 320d xDrive] - First registration: [month/year] - Mileage: [reading] - Color exterior / interior: [...] - Equipment: [highlights — navigation, leather, tow hitch, etc.] - Condition: [describe honestly — e.g.: "good overall condition, two small paint scratches rear right, service last month, new brakes"] - Service book: [complete / incomplete] - Accident-free: [yes/no — honestly] - Inspection valid until: [month/year] - Tires: [summer tires mounted, winter set included? Tread depth?] Price: [negotiable? Fixed price?] Listing structure: - Attention-grabbing title (max. 65 characters) - Short description, 3 sentences - Bullet points: key data - Bullet points: equipment - Bullet points: condition (also honestly name the flaws — builds trust) - Standard boilerplate at the end ("Test drive by arrangement, no guarantee/warranty, private sale, viewing in [CITY]") Style: Factual, complete, honest. No exaggerations.
Private Practical · 5-20 min

8 — DIY instructions step by step

Hanging a lamp. Connecting the washing machine. Drilling into a concrete wall. Claude gives you careful, correct instructions — and says clearly where you should call a professional.

Prompt to copy
Please create step-by-step instructions for the following DIY job: Task: [e.g. "Replace the ceiling lamp in the kitchen, old building, fabric-insulated wiring is possible"] Please provide the following sections: 1. SAFETY FIRST - What needs to be considered beforehand? - When is this a DIY job, when is an electrician better? - Which insurance policies or regulations are relevant? 2. TOOLS / MATERIALS List of what I need (incl. cost range) 3. STEPS Numbered, each step a concrete, verifiable action ("Switch off the fuse in the fuse box and test with a voltage tester at the lamp connection — NO light? → ok") 4. TYPICAL PITFALLS 5. STOP CRITERIA When do I have to abort and call a professional? (e.g. unknown wire colors, fabric insulation, no protective earth conductor) If electricity, gas, pressurized water, or load-bearing components are involved — please make the safety notice extra clear.
For electricity, gas, structural work: a professional.

Claude knows the theory. But your old building, your specific wiring, your actual fuse situation — only you and an electrician know those. When in doubt: read the instructions, then call.

Private Family · 5 min

9 — Letter to the school or class teacher

Sick note covering several days. Request for an excused absence for a family occasion. A constructive reply to a school notice you don't agree with.

Prompt to copy
Please write a short letter to [the class teacher / the school administration]. Occasion: [concrete — e.g. "Request for excused absence from school on Friday because my daughter is traveling to Hamburg for her grandmother's 50th birthday"] Please: - Appropriate salutation ("Dear Ms. [Name]") - Clear facts in 2-3 sentences - Request / application worded concretely - Note about willingness to catch up (homework, school material) - Polite closing - Max. 120 words Style: respectful, collaborative, not submissive and not condescending. Class teachers are our allies.
Private Hobby · variable

10 — Hobby research (wine, bicycle, car, travel)

You've been semi-seriously interested in wine for two years. But you don't want to become a sommelier — you want to know which Pinot Noir to bring to the next family dinner without embarrassing yourself.

Prompt to copy
I need brief advice on the following topic: Topic: [e.g. "Which Pinot Noir / Spätburgunder should I bring to the next family dinner?"] My experience level: - [honestly: beginner / hobby / advanced] - Liked so far: [specific wines / brands / regions I enjoyed] - Don't like: [what was too much for me — too heavy, too sweet, too tannic] Occasion / setting: - Who's drinking along? [e.g. in-laws, moderate wine knowledge] - What's being served? [concrete] - Budget: [per bottle] - Where do I buy? [supermarket / wine merchant / online] Please: - 3-5 concrete recommendations (winery + wine + vintage if possible) - Per recommendation: 2-3 sentences (flavor profile, why it fits) - One "expansion recommendation" for later, once I've found my taste - A few words on why these wines fit my taste - If you're unsure about availability or price — say so honestly Finally: 1-2 sentences on what to keep in mind when tasting (temperature, glass, order).
Hobby research is Claude's favorite playing field.

If you honestly describe where you stand — beginner, intermediate, advanced — you don't get a textbook answer but a personal recommendation. Works for wine, bicycles, cameras, books, travel destinations, whisky, classical music, hiking routes.

Required reading · 18 minutes

Data protection, GDPR, and the limits of Claude.

This page is not optional. If you work with Claude in the practice, you are responsible — for the data, for the correctness, for the consequences. Here is what you need to know. Clear, without legal waffle, but precise.

The one rule that carries everything:

Never enter patient data into the consumer versions of Claude. Free, Pro, and Max are not intended for anything patient-related — not even "just briefly", not even "just this one case". Period.

DW
Dr. Butze, after the first scare
Wednesday evening · the tax advisor called

"I had the idea of having Claude type up a referral letter — after all, I had only entered the diagnosis and a few keywords, no name. My data protection officer called me before I could hit 'Send'. Patient data is not just name and date of birth. That was an important lesson."

Plan comparison

Which Claude plan is suitable for what?

Anthropic offers different plans with different contractual assurances. For use with patient data, the difference is existential.

Free, Pro, Max

Personal accounts. No data processing agreement (DPA) available. These plans are aimed at private individuals and general use.

Unsuitable for patient data. Even if you turn off training (which is possible — see below), there is no contract that fulfills the GDPR requirements for a data processor.

Allowed: Everything without patient reference. Practice texts, marketing, letters, research, private matters.

Claude for Work (Team, Enterprise)

Business accounts with DPA option. For these plans, Anthropic offers a data processing agreement (DPA — Data Processing Addendum) and additional security and compliance options.

Even here: The DPA alone does not automatically make use with patient data permissible. It additionally requires your own data protection impact assessment (DPIA), the OK of your data protection officer, and sensibly a restriction to clearly defined use cases.

Recommendation: If you want anything patient-related, Team/Enterprise is the minimum.

These statements are guidance, not legal advice.

Terms, contracts, and options change. Before you set up a business account, have your data protection officer review the current contract terms and the supplementary guarantees from Anthropic.

Data flow

What happens to your inputs?

  1. Transmission to Anthropic's servers

    Everything you type into the chat — and all files you upload — is transmitted encrypted to Anthropic's servers. Anthropic is a US company headquartered in San Francisco. Server locations and sub-processors are defined by the current state of the contract documents, which are regularly updated — ask your data protection officer for the currently applicable versions.

  2. Processing by the model

    The model processes your input and generates a response. During this processing, your data is briefly present in the working memory of the servers.

  3. Storage of the conversation (chat history)

    Your chats are stored in your account so you can refer back to them later. You can delete chats — that deletes them in your account. How long they remain on the servers in backups or logs is governed by the contract.

  4. Training: off by default, but check

    Anthropic states that inputs from consumer accounts are not used to train the models without explicit consent. In the settings, a corresponding option is usually switched off by default. Check this in your account: Settings → Privacy → Help improve Claude. Turn it off deliberately — document it in writing.

Even in the "training off" state, your inputs have not vanished.

They are just not baked into future model versions. But they reside on US servers. They are technically accessible — to Anthropic, to sub-processors, in the extreme case to US authorities under the CLOUD Act. That is the basis for the strict rule above.

Red lines

What you must never enter.

This list is not exhaustive — but if any of it applies, it's clear: not into Claude.

Personal patient data

  • Name, date of birth, address
  • Insurance ID, KV number
  • Phone, email
  • Photo, ID copy

Medical findings with personal reference

  • Original referral letters
  • Lab results, imaging
  • History notes with identifying reference
  • Also indirect identification (rare + region + diagnosis)

Your own sensitive data

  • Passwords, PINs
  • Bank details, account statements
  • Tax IDs, social security
  • Confidential material from ongoing proceedings

Practice-internal confidentiality

  • Staff personnel files
  • Salaries, contracts with identifying reference
  • Confidential correspondence
  • Practice management system data
"But I left out the name!"

Often not enough. A 47-year-old female patient from Berlin-Charlottenburg with a rare diagnosis is identifiable — even without a name. GDPR-relevant is anything that can identify a "natural person", alone or in combination. The rule of thumb: when in doubt, it counts as personally identifiable.

Practice

Anonymizing — how to do it right.

For many tasks you don't need patient data at all. You need the structures. This is how you phrase requests without patients becoming identifiable:

Not like this

  • "Write a letter to Ms. Schmidt, born April 12, 1962, regarding her diabetes…"
  • "Explain this finding to me: Pat. K.M., 67 years old, atrial fibrillation, Marcumar since 2019…"
  • "Here is the referral letter from oncology — can you translate it into plain language for the patient?"

Better like this

  • "Write an example letter to a typical diabetes patient switching their medication — placeholders for salutation and date."
  • "Explain the finding 'paroxysmal atrial fibrillation on Marcumar' in general terms — when is which adjustment common?"
  • "Here is an anonymized example report (markers like [DATE], [PATIENT], [DOCTOR]) — please write a patient-friendly version with the same placeholders."

The three anonymization techniques

Technique 1

Placeholders

Copy the original text, replace personal references with markers: [NAME], [DOB], [ADDRESS]. The markers remain in the result — you only enter the real data manually at the end.

Technique 2

Generalization

Instead of "patient K.M., 67 y/o, from Berlin-Charlottenburg, with X", better "a 65-year-old patient with X". Specific identifiers out, clinical constellation in.

Technique 3

Hypothesization

"Imagine the following case: …" — this turns a concrete case into a teaching example. The model responds to the example; your clinical reality stays out of it.

Most tasks don't need real data at all.

A patient education text is general. A practice letter template is general. A guideline summary is general. Patient data is almost never necessary — it only creeps in because we're used to thinking in concrete cases.

Professional limits

Hallucinations — when Claude is convincingly wrong.

Claude is a language model. It generates plausible text based on statistical patterns in training data. It doesn't know anything in the human sense. This has a specific consequence: sometimes Claude invents facts. Very convincingly.

Classic hallucination hazards in the medical context:
  • Invented study citations ("Müller et al. 2021 showed…") — with author, journal, page number, all freely invented
  • Wrong dosage information — plausible, but outside the guideline
  • Wrong drug interaction information
  • Wrong guideline versions or outdated recommendations
  • Invented ICD or OPS codes
  • Wrong billing codes (GOÄ/EBM)

How to spot hallucinations

  1. Strikingly specific citations

    If Claude names a study with author, journal, year, and page — and it sounds important — google it. Only then cite it or act on it.

  2. Answers to difficult questions that are too smooth

    Complex medical questions often have no unambiguous answer. If Claude nevertheless sounds like "The recommendation is…": become skeptical.

  3. Concrete numbers without context

    "The success rate is 73%" — where does this number come from? From which study, which patients, which endpoints? Ask.

  4. Confidence is not an indicator of correctness

    Claude phrases wrong statements just as assuredly as right ones. You cannot infer correctness from the tone of the answer.

Protective sentences for your system prompt:

— "For medical statements, please always indicate the uncertainty: 'established' / 'likely' / 'unclear'."
— "If you're unsure about a source, say so openly instead of guessing."
— "Never invent study citations or guideline references. Better 'I have no concrete source for this'."

Law & professional code

Your medical responsibility remains — in full.

Even if you use Claude for preparation, for research, for writing — the legal and professional responsibility for what reaches your patient stays with you. Always.

Malpractice liability

If a patient is harmed by an incorrect treatment recommendation, you are liable. Not Anthropic. Not the model. You cannot claim that the AI said so.

Medical confidentiality (§ 203 German Criminal Code)

If you give patient-related information to an AI system not covered by contract, that can be a violation of medical confidentiality — punishable by law, independent of the GDPR.

GDPR fines

A GDPR violation carries fines of up to 20 million euros or 4% of annual revenue (for SMEs typically much less, but five to six figures are realistic). On top come possible damage claims from individual patients.

Professional consequences

The state medical associations can take action for gross violations of the professional code and data protection — up to and including questions about your license to practice in extreme cases.

Treat Claude like a very diligent medical assistant intern in her first month.

You get a lot of things faster and better. But proofreading, checking, signing — that's you. Always.

Regulation

EU AI Act — what you need to know from August 2026.

The EU AI Regulation (EU AI Act) has been in force since August 2024 and is taking effect in stages. Most general obligations for using organizations apply from August 2, 2026.

Risk classes

The AI Act classifies AI systems into risk classes. AI in the medical field can range from "limited risk" to "high risk" depending on the type of use. Language models that make or prepare diagnostic or treatment decisions tend to be high risk.

Transparency obligations

Patients must be able to recognize when they are interacting with an AI. AI-generated texts that go to patients should be identifiable as such — the legal line here is still in motion.

Human in the loop

For medical decisions, "human oversight" is mandatory. An AI may not make a medical decision autonomously or prepare it alone — the doctor must review, approve, take responsibility.

Documentation

Practices that use AI substantially in patient-related processes will have documentation and evidence obligations. Which ones exactly and from what threshold — is still being specified in detail in guidance from the supervisory authorities.

The AI Act is still fluid in its details.

Much will be specified through implementing regulations and guidance from the EU and national supervisory authorities. Keep an eye on the topic — via your data protection officer, your medical association, the KV.

Pre-use check

Ten-second checklist before every Claude use.

Train yourself to mentally run through these five questions before every prompt. After two weeks it becomes automatic.

  1. Is there a personal reference in it?

    A name, a date of birth, an address, a photo, an insurance ID, a rare diagnosis with region — anything that makes a specific person identifiable? If yes: STOP. Anonymize or don't enter it at all.

  2. Am I on the right plan?

    Anything patient-related only in Claude for Work with a DPA and after data protection officer approval. General tasks work on any plan — as long as there's no personal reference.

  3. Have I disabled training?

    Settings → Privacy → "Help improve Claude" → off. Check it once, done.

  4. Do I understand what Claude will answer?

    If you cannot judge the topic professionally, you cannot judge the answer either. Then Claude is the wrong tool for this case.

  5. Do I have a strategy to verify the answer?

    Comparison with a guideline? Cross-check with colleagues? Your own expertise? If the answer can have patient-related consequences, a review must take place before acting.

When to involve the DPO

When should you consult your data protection officer?

Always involve them when:

  • You set up a Claude business plan (Team/Enterprise)
  • You want to use Claude in a process where patient data could occur
  • You update your practice's record of processing activities (ROPA)
  • You need a DPIA (data protection impact assessment)
  • You are unsure whether a specific use is permissible

Usually not necessary when:

  • You work with Claude privately (vacation planning, letter to the tax advisor)
  • You do practice tasks without personal reference (marketing, job posting, answering machine script)
  • You do general continuing education without reference to concrete cases
If you don't have a data protection officer:

Family practices are often not required to appoint a DPO (that depends on staff numbers and data processing), but they carry the responsibility themselves. In that case: get external advice before using Claude in patient-related workflows. There are specialized data protection consultants for medical practices, often referred through the medical associations, KV, or professional bodies.

You've got the most important part behind you.

If you internalize the pre-use checklist, work on the right plan, recognize hallucinations, and preserve your medical responsibility, you can use Claude safely and productively. Data protection is not an obstacle — it's a routine you'll quickly internalize.

Onboarding · 10 minutes

Ready to go in 10 minutes.

You need three things: a browser, an email address, five minutes of undisturbed time. This page walks you linearly from login to your first productive result. No technical prior knowledge required — just a little patience with yourself the first time.

DW
Dr. Butze on a Wednesday evening
General practitioner · after consultation hours

"I've just poured a glass of wine. I'm opening Claude. I haven't done anything yet. Tell me exactly what to click — and what to type, so that after 10 minutes I know whether this is worth anything to me."

Step 1 · Account

Create an account — like an email address, but in 90 seconds.

You do this once. After that, you just log in next time — exactly like on any other website.

Open claude.ai in your browser

Type it into the address bar of Chrome, Safari, Firefox, or Edge. You'll land on a start page with two buttons: "Sign up" and "Log in". If it's your first time there, click Sign up.

Sign up — easiest with Google

You can register with any email address. If you have a Google account (Gmail, Google Workspace), click "Sign in with Google" — one click, done. Alternatively, the classic way with an email address plus a password of your choice.

Confirm your phone number

Claude asks once for your mobile number to prevent spam accounts. You get a 6-digit code via SMS, enter it — that's it. Your number is not used for anything else.

Provide date of birth & name

A few mandatory fields for age verification and for how you're addressed. For the name, your first name is enough — you're talking to Claude, not a registry official. You can change this name at any time later.

You're in.

You now see a white chat window with a friendly greeting and a text input field at the bottom. That's all. This is where all the magic happens.

Free to try out.

The Free plan is completely sufficient for what you'll do today. You can chat for several hours before you even hit a limit. Payment comes later — and only when you really need it. More on that below.

Step 2 · Orientation

The interface — three areas, nothing more.

Don't get distracted by buttons you don't need today. These three areas are the ones you must know:

1. Text input field

At the bottom center. This is where you type or paste your question. Enter sends, Shift+Enter makes a line break. Everything happens here.

2. Left sidebar

Your history. Every chat you've had stays saved there. You can come back to it at any time. One click on "New chat" starts a fresh conversation.

3. Your initials (top right)

This takes you to the settings: change plan, profile details, privacy options. You'll only need this once today — we'll go there in step 4.

Forget everything else.

There are a few more icons — for Projects, Artifacts, models. Those come later. Today you only use the text input field. Really.

Step 3 · Your first conversation

Writing your first prompt.

Let's make it sporting. Three tasks that will give you a real aha moment in the next 5 minutes. Simply copy each prompt, paste it into the input field, press Enter — and see what happens.

Test 1 — Translate a patient question into plain language

First test · 30 seconds
Explain to a 68-year-old female patient without a medical background, in simple, friendly words, what arterial hypertension is, why it is treated even in the absence of symptoms, and which two to three changes in everyday life (exercise, diet, sleep) can concretely lower blood pressure. Avoid medical jargon. About 150 words. Close with an encouraging sentence.

Test 2 — Draft an email

First test · 30 seconds
Draft a short, friendly-but-firm email to a pharmaceutical representative whom I want to decline politely but clearly, because my consultation calendar leaves no room for in-person appointments for the next three months. Thank him for the offer, offer to get in touch again in the new year — but please without any loophole that sounds like a "maybe after all." Salutation: "Dear Mr. Müller". Signature: "Kind regards, Dr. M. Butze".

Test 3 — Something private: restaurant suggestions

First test · 30 seconds
Suggest three restaurants in Berlin-Charlottenburg for a date on Saturday evening: upscale cuisine, but no Michelin temple. Italian or Mediterranean. Good wine list. A reservation should still be possible two days in advance. For each restaurant, briefly explain why it might fit. If you're unsure about opening hours or the current situation, say so explicitly — I'll call myself then.
Done? Congratulations.

You've just seen in three areas — medical explanation, business communication, private life — what Claude can do for you. Now let's set up a few more settings so this keeps working sensibly for you long-term.

Step 4 · Settings

Three settings you set once today.

Click on your initials at the top right, then on Settings. Three tabs matter:

Profile settings → "What should Claude call you?"

Enter your first name and briefly what you do. Example: "General practitioner with my own family practice in Berlin. I use Claude for patient communication, research, and private topics like family and travel." Claude will then tailor its answers to you better.

Privacy → training to "Off"

In the Privacy tab you'll find the option for whether Anthropic may use your chats to train new models. Set it to Off. That way your inputs won't end up in future training data. (On the Free plan this is important. On Team and Enterprise it's disabled by default anyway.)

Appearance → dark mode (optional)

Easy on the eyes, especially in the evening. A matter of personal taste. No impact on functionality — only on your eyes.

Important — even with training disabled:

Real patient data still never belongs in Claude (at least not in the consumer plans Free/Pro/Max). Anthropic is a US provider, and no GDPR data processing agreement is available on the Free plan. For professional use involving patients, you need Claude for Work (Team/Enterprise) with a DPA or a different provider. Read the data protection chapter.

Step 5 · Plan

Which plan — and when?

Start on Free. Period. After two to four weeks of use you'll know for yourself whether to upgrade. Here is an honest, short overview (as of 2026):

Free — €0 / month

For whom: First-time users, occasional private use.
What you get: Claude in the chat, limited messages per day, access to the current model, simple Projects.
How the limit feels: With intensive use you reach the daily limit after 1–2 hours.

Pro — approx. €20 / month

For whom: Anyone who uses Claude regularly (several times a week).
What you get: ~5× more messages than Free, unlimited Projects, extended knowledge base (RAG), longer conversations.
How the limit feels: Sufficient for most private users and occasional practice tasks (without patient data).

Max — €100–200 / month

For whom: Power users who have Claude open all day.
What you get: 5×–20× higher limits than Pro, priority during peak load.
How the limit feels: Practically none. Considerably oversized for normal practice use.

Team / Enterprise

For whom: A practice with staff working with patient-related data.
What you get: Data processing agreement (DPA), no training use of your data, staff management, shared Projects.
How the limit feels: The only path to GDPR-compliant use with actual practice data.

My suggestion — pragmatic:

Free for 2 weeks. Then upgrade to Pro when you notice that you're using Claude regularly. Only switch to Team before you concretely plan to use Claude with identifiable practice data — and even then only after consulting your data protection officer.