Claude CoWork for Nurses

What is Claude CoWork?

Claude CoWork is the practice of using Claude as a persistent, knowledgeable co-worker embedded in your daily nursing workflow. This is not about asking a chatbot a one-off question and hoping for the best. It is about configuring Claude with your clinical context, documentation standards, and specialty focus so that every interaction produces output you can actually use.

Claude-native prompts. The prompts in this guide use Claude's native XML tag structure (<context>, <instructions>, <format>, <avoid>) for more precise, consistent output. These tags help Claude parse your intent with less ambiguity. They work in ChatGPT too, but are optimized for Claude.

Think of Claude as the sharpest colleague you have ever worked with, one who never forgets your unit's documentation preferences, knows nursing frameworks inside and out, and can draft a care plan, clinical note, or patient education handout in seconds. The difference between nurses who dabble with AI and those who gain a real edge comes down to setup and consistency.

This guide walks you through setting up Claude specifically for nursing work, the five workflows that will save you the most time, and the prompting techniques that separate generic output from clinically sound content.

Install the Nurse Plugin

This guide works on three Claude surfaces. The plugin is the fastest path on two of them. Pick whichever you use:

If you're on Cowork (desktop or mobile app)

Claude Cowork is Anthropic's agentic workspace — Claude completes work autonomously and returns finished deliverables. The Nurse plugin packages the workflows below as native skills and slash commands.

Open the Cowork plugin directory in your desktop app.
Filter by Cowork, search for "Nurse", and click Install.
The plugin's slash commands and ambient skills are now available in any Cowork task.

If you don't see the plugin in the directory yet, install via custom marketplace: paste https://github.com/alexclowe/awesome-claude-cowork-plugins in your Cowork plugin settings.

If you're on Claude Code (CLI)

Install from your terminal:

claude plugin add alexclowe/awesome-claude-cowork-plugins/nurse

The plugin's slash commands and skills load on next session.

If you're on Claude.ai (web chat only)

Plugins aren't directly installable on the web chat surface. You have two options:

Use the prompts in this guide directly in a Claude Project (covered in the next section). Same outputs, more typing.
Upload the plugin's skills as a zip via Settings → Features → Custom Skills (Pro/Max/Team/Enterprise plans). Higher friction; only worth it if you want the auto-activating skills, not the slash commands.

What the plugin gives you (any surface)

Slash command	What it does
`/nursing-note`	Draft clinical documentation with assessment findings, interventions, and patient evaluation
`/shift-handoff`	Generate structured SBAR handoff reports for clear shift-to-shift communication
`/care-plan`	Create nursing care plans with NANDA-I diagnoses, NIC interventions, and NOC outcomes
`/patient-education`	Generate plain-language patient education materials adapted to health literacy level

Auto-activating skills (no command needed — Claude applies them when relevant):

Clinical Nursing — Nursing assessment, NANDA-I diagnoses, NIC/NOC frameworks, medication administration documentation, and clinical charting
Patient Education — Health literacy-adapted materials, teach-back frameworks, and plain-language medical explanations

The plugin works standalone for one-off tasks. Pair it with the surface-specific setup below for persistent context across every task — that combination is the full Claude CoWork setup.

Setting Up Claude for Nursing Work

Surface note: The Project setup below is for claude.ai web users. Cowork users have their own task-context mechanism (set context once when starting a Cowork task). Claude Code users get the plugin's ambient skills automatically — no Project setup needed. The workflows themselves are surface-agnostic — paste the prompts wherever you're working. The key to getting consistently useful output from Claude is using Claude Projects. A Project lets you set custom instructions that persist across every conversation, so you are not re-explaining your clinical setting every time.

Step 1: Create a Nursing Practice Project. In Claude, click "Projects" and create one called something like "My Nursing Practice."

Step 2: Set your custom instructions. In the Project settings, add instructions like:

You are my nursing documentation assistant. Here is my context:

<business-profile>
- Name: [Your Name], [Credentials — RN / BSN / MSN / NP / etc.]
- Setting: [Med-surg / ICU / ER / Home health / Outpatient clinic / etc.]
- Specialty: [Cardiac / Oncology / Pediatrics / Geriatrics / etc.]
- Facility type: [Hospital / Skilled nursing facility / Community health center]
- Documentation system: [Epic / Cerner / Meditech / etc.]
- Preferred care plan framework: [NANDA-I / Facility-specific]
</business-profile>

<rules>
- Never use real patient names, MRNs, dates of birth, or identifying details. Use initials only (e.g., "P.R.").
- All clinical language must be professional and consistent with current nursing terminology.
- Output is always a draft for nurse review. Include: "DRAFT — FOR CLINICAL REVIEW."
- Stay within the registered nurse scope of practice. Do not provide medical diagnoses or prescribe treatments.
</rules>

Step 3: Upload reference documents. Add your facility's care plan templates, documentation guidelines, SBAR templates, or patient education formatting standards to the Project knowledge base. Claude will reference these when generating content.

Step 4: Start every session inside this Project. This ensures Claude always has your context loaded.

Your Top 5 Workflows with Claude

1. Care Plans (NANDA Framework)

Developing individualized care plans with NANDA nursing diagnoses, outcomes, and interventions is essential but repetitive. Claude can build the framework from your assessment findings.

<task>Draft a nursing care plan using NANDA-I framework.</task>

<context>Patient: M.R., 72M, admitted CHF exacerbation. Findings: BLE edema 2+, bilateral crackles, 4 lb weight gain in 3 days, O2 sat 91% on RA, dyspnea on exertion. Reports difficulty understanding low-sodium diet.</context>

<instructions>
- 3 priority NANDA-I nursing diagnoses, prioritized by clinical urgency
- For each: related factors, defining characteristics, outcomes with timeframes, 4-5 interventions with rationales
- Include patient education components; use measurable outcome criteria
</instructions>

<avoid>Medical diagnoses or physician-level orders; patient identifiers beyond initials; interventions outside RN scope</avoid>

Before Claude: 30-45 minutes drafting a comprehensive care plan. After Claude: 5 minutes to input findings, 10 minutes to review and individualize.

2. Clinical Notes

Charting accurately and efficiently at the end of a shift is a constant challenge. Claude can structure your observations into clear, complete documentation.

<task>Write a clinical nursing note for a post-operative patient assessment.</task>

<context>Patient: K.L., 58F, POD 1 R total knee arthroplasty. VS: 132/78, HR 76, RR 16, T 98.8, SpO2 97% RA. Pain 5/10, managed with Toradol/PRN oxycodone. Dressing CDI, drain 45mL serosanguinous/8hr, pedal pulses palpable, RLE NVI. Ambulated 50ft with walker and PT.</context>

<instructions>
- Systematic assessment format with pain, mobility, and functional status
- Note patient education provided and plan for remainder of shift
- Concise, objective clinical language
</instructions>

<avoid>Subjective opinions unsupported by assessment data; patient identifiers beyond initials</avoid>

Before Claude: 15-20 minutes per detailed clinical note. After Claude: 3 minutes to input key findings, 5 minutes to review and finalize.

3. Shift Handoff Reports (SBAR)

A structured SBAR handoff ensures nothing falls through the cracks during shift change. Claude can organize your observations into a clear, standardized format.

<task>Draft an SBAR shift handoff report for an incoming night nurse.</task>

<context>Patient: T.W., 45M, Rm 412, Day 3, acute pancreatitis.</context>

<instructions>
- Situation: NPO, LR 125mL/hr, Dilaudid PCA, lipase 1200 to 680
- Background: Hx alcohol use disorder, prior pancreatitis 2 years ago
- Assessment: Pain 4/10 (was 8/10), tolerating clear liquids since 1400, BS x4, afebrile, UOP adequate
- Recommendation: Monitor PO tolerance overnight, advance diet per protocol, recheck lipase 0600, fall risk with bed alarm
- Include pending labs and PRN meds given this shift
</instructions>

<avoid>Patient identifiers beyond initials; information irrelevant to care continuity</avoid>

Before Claude: 10-15 minutes organizing each complex patient handoff. After Claude: 3 minutes to input, 3 minutes to review the structured SBAR.

4. Patient Education Materials

Nurses spend significant time creating or finding appropriate education materials. Claude can draft clear, patient-friendly handouts tailored to specific conditions.

<task>Create a patient education handout on managing type 2 diabetes at home after discharge.</task>

<context>Audience: newly diagnosed T2DM patients. Reading level: 6th-8th grade. Setting: hospital discharge.</context>

<instructions>
- Cover: glucose monitoring basics, medication adherence, dietary guidelines, signs of hypo/hyperglycemia, when to call the doctor
- Include a simple daily checklist; under 500 words
- Warm, encouraging tone; clear headings and short paragraphs
</instructions>

<avoid>Jargon without plain-language explanation; specific med names/doses; content contradicting ADA guidelines</avoid>

Before Claude: 30-45 minutes searching for or creating materials. After Claude: 5 minutes to input, 10 minutes to review and customize.

5. Insurance and Prior Authorization Documentation

Documenting medical necessity for insurance purposes requires precise clinical language. Claude can structure these narratives efficiently.

<task>Draft a nursing narrative supporting medical necessity for continued SNF placement.</task>

<context>Patient: D.H., 80F. Dx: R hip fracture s/p ORIF, c/b post-op delirium (resolved) and DVT (on anticoagulation). WBAT on R, mod assist transfers, PT 5x/week. Lives alone, single-story home with 3 front steps.</context>

<instructions>
- Document skilled nursing needs: wound assessment, anticoagulation monitoring, fall prevention
- Describe functional limitations preventing safe discharge home
- Include rehab progress and remaining goals; reference ADL assist levels
- Objective, measurable language; under 350 words
</instructions>

<avoid>Patient identifiers beyond initials; recovery timeline guarantees; language suggesting no progress</avoid>

Before Claude: 20-30 minutes drafting medical necessity documentation. After Claude: 5 minutes to input, 10 minutes to review and ensure accuracy.

Prompt Engineering Tips for Nurses

1. Always specify the clinical setting. "ICU nursing note" produces very different output than "home health visit note." Your clinical environment shapes documentation requirements, terminology, and level of detail.

2. Include the NANDA diagnosis label when requesting care plans. Specifying "Risk for Impaired Skin Integrity" rather than "skin problems" ensures Claude uses proper nursing taxonomy and generates aligned interventions.

3. State the documentation framework explicitly. Asking for "SBAR format" or "NANDA-I care plan" or "focused DAR note" tells Claude exactly how to structure the output. Do not assume it will pick the right format.

4. Specify the scope of practice. Add "Stay within the RN scope of practice" to prompts. This prevents Claude from generating content that crosses into medical diagnosis or prescriptive authority.

5. Request rationales for interventions. Adding "Include evidence-based rationales for each intervention" produces care plans that are more useful for clinical reasoning and student education.

6. Use assessment data, not interpretations. Provide objective findings like "edema 2+ bilateral lower extremities" rather than "patient is fluid overloaded." Let Claude help you connect assessment data to nursing diagnoses properly.

Privacy & Compliance

HIPAA compliance is non-negotiable. Claude is not a HIPAA-covered entity or business associate. Do not enter protected health information (PHI) into Claude. This means no patient names, medical record numbers, dates of birth, or any combination of data that could identify a specific individual.

Use initials and de-identified information only. Replace all patient names with initials. Omit MRNs, room numbers tied to real patients, dates of service, and insurance information. Describe clinical presentations in general terms.

Verify all clinical content. Claude generates plausible clinical language, but it is not a licensed clinician. Verify that nursing diagnoses, interventions, and assessment documentation accurately reflect the patient's condition and your clinical judgment. You sign the chart — you own the documentation.

Scope of practice awareness. Ensure that AI-generated content does not include medical diagnoses, medication orders, or other content outside the nursing scope of practice. Claude should support your documentation, not expand your scope.

Check your facility's AI policy. Many healthcare facilities have specific policies regarding the use of AI tools for clinical documentation. Review your employer's policies before incorporating Claude into your workflow.

Going Further

Ready to build on this foundation? Check out these resources:

Browse our full collection of nursing prompt packs and cheat sheets for ready-to-use templates
Run an AI readiness audit for your nursing practice to identify your biggest opportunities
Explore AI-powered tools built specifically for nurses to automate even more of your workflow