Skip to main content
Agent skill

Clinical Case Report

Structured medical case presentation for clinical rounds, conferences, and documentation. Generates SOAP-format or narrative case reports with physiologically accurate vitals, labs, and evidence-based plans. Use when the brief mentions "case report", "case presentation", "SOAP note", "clinical case", "ward rounds", "case summary", or "patient presentation".

Get started

Get the complete skill folder, including its scripts and reference files.

When to use this skill

Use Clinical Case Report when an AI agent needs a reusable SKILL.md workflow for this job: Structured medical case presentation for clinical rounds, conferences, and documentation. Generates SOAP-format or narrative case reports with physiologically accurate vitals, labs, and evidence-based plans. Use when the brief mentions "case report", "case presentation", "SOAP note", "clinical case", "ward rounds", "case summary", or "patient presentation".

When not to use it

Skip Clinical Case Report when the task is outside the creativity category, or when a more specific skill in this directory already covers the same workflow with clearer triggers.

How to install

Where will you use it?

Install scope
Claude Code folder setup
mkdir -p ~/.claude/skills/clinical-case-report
# Save SKILL.md to ~/.claude/skills/clinical-case-report/SKILL.md
  1. Claude Code personal install: place the skill and its bundled files at ~/.claude/skills/clinical-case-report/SKILL.md. For a shared project, commit the folder at .claude/skills/clinical-case-report/ instead.
  2. Current Claude Code detects SKILL.md changes in watched directories during the session. If you just created a top-level skills directory, run /reload-skills. Check that the skill is listed before testing its trigger.
  3. For Claude chat or Cowork in the Desktop app, manage uploaded skills in Customize > Skills; copying files into a local Claude Code folder is not a Desktop install. Other agents may use different install locations.

Full install guide for Claude, Cursor, and Codex →

Try a starter prompt

After installing the skill, paste this starter prompt into your agent to verify it triggers on a small task.

Starter prompt
Use the Clinical Case Report skill for this task: Use Clinical Case Report when this documented workflow matches the task. Confirm any required inputs or files before making changes.

What supports this recommendation?

67/100 provisional content score. This measures documentation, practical detail and source signals. It is not a measured task success rate or a security certification.

Documentation
75/100
Practical detail
58/100
Source evidence
63/100
Source signals
72/100
Model compatibility evidence
  • claudeuntested

    No model-specific signal or recorded compatibility test was found.

  • chatgptuntested

    No model-specific signal or recorded compatibility test was found.

  • geminiuntested

    No model-specific signal or recorded compatibility test was found.

  • copilotuntested

    No model-specific signal or recorded compatibility test was found.

  • llamauntested

    No model-specific signal or recorded compatibility test was found.

  • perplexityuntested

    No model-specific signal or recorded compatibility test was found.

  • mistraluntested

    No model-specific signal or recorded compatibility test was found.

  • grokuntested

    No model-specific signal or recorded compatibility test was found.

Inferred compatibility is based on content signals. Check the app, its available tools and setup requirements before running a skill.

What this skill does

# Clinical Case Report Skill

Generate a structured medical case presentation for clinical rounds, conferences, or documentation. The output follows standard medical formatting conventions used in hospital settings worldwide.

## What you will produce

A single-page HTML case report (`index.html`). Content varies by format (see `references/case-formats.md` — selected in Step 0):

**SOAP / Conference format:** - **Patient identification** — age, sex, chief complaint - **History of Present Illness (HPI)** — chronological narrative with pertinent positives and negatives - **Past Medical History, Medications, Allergies** - **Review of Systems** - **Physical Examination** — systematic findings by system - **Vital Signs** — formatted table with reference ranges and flags - **Investigations** — laboratory results and imaging findings - **Assessment** — primary diagnosis and differential (3–5 items) with clinical reasoning for each - **Management Plan** — evidence-based, organised by problem

**Brief Rounds format** (daily review, ward round, handover, ICU, post-call): - **ID line** — age, sex, day of admission, primary problem - **Interval events / current status** — what has changed since last review - **Active problems** — numbered list - **Plan-by-problem** — concise actions for each active problem - Full HPI and systematic physical examination are **not** included

---

## Step-by-step workflow

### Step 0 — Load reference files

Before starting, read both reference files:

1. `references/case-formats.md` — use this to choose the correct output format (SOAP, Conference, or Brief Rounds) based on the user's context 2. `references/checklist.md` — keep P0 gates in mind throughout; you must pass all P0 items before emitting the final artifact

### Step 1 — Parse the brief

Read the user's prompt and extract:

- Patient age and sex - Chief complaint or presenting problem - Any vitals, labs, or imaging the user has provided - Clinical context: ED, ward rounds, conference case, outpatient, etc. - Specialty context: cardiology, emergency, internal medicine, etc.

If the chief complaint or presenting problem is missing: - **SOAP / Conference**: ask one clarifying question before proceeding. Do not proceed without it. - **Brief Rounds**: if the admission problem or ID line is already available (e.g. "day-3 ICU review for septic shock"), proceed directly — a separate chief complaint is not required.

### Step 2 — Build the clinical narrative

**For SOAP / Conference outputs:** write the HPI as a continuous prose narrative in standard clinical style:

> "This is a [age]-year-old [sex] with a history of [relevant PMH] who > presents with [chief complaint]. Symptoms began [timeline] and are > characterised by [quality, severity, radiation]. Associated symptoms > include [list]. Pertinent negatives include [list]."

The HPI must be chronological. Include timeline markers ("2 hours prior to presentation", "onset yesterday morning").

**For Brief Rounds outputs** (daily review, ward round, handover, ICU, post-call): skip the full HPI and examination. Instead produce:

- **ID line**: "[Age][sex], Day [N] of admission, [primary problem]" - **Interval events / current status**: what has changed since last review - **Active problems**: numbered list - **Plan-by-problem**: concise action for each active problem

### Step 3 — Generate physiologically consistent clinical data

If the user has not provided specific values, generate values that are internally consistent with the diagnosis:

**Consistency checks (typical patterns):**

- A patient in shock **typically** has: HR >100, SBP <90, raised lactate, impaired capillary refill — but medications (beta-blockers), age, or shock type (neurogenic, spinal) can alter this pattern - Pneumonia **typically** presents with raised WBC, raised CRP, temperature >38°C — but afebrile pneumonia exists, especially in the elderly or immunocompromised - A STEMI **typically** shows ST elevation in contiguous leads and raised high-sensitivity troponin — but early presentations may have initially normal troponin; CK-MB is not universally required - Sepsis **typically** shows raised or low WBC, raised lactate >2, temperature abnormality — but compensated early sepsis may present with normal vitals - Lab units must match convention: creatinine in µmol/L or mg/dL (state which), glucose in mmol/L, haemoglobin in g/dL

**Critical rule — preserve user-provided data:** - Never overwrite a value the user has explicitly stated - If a user-provided value is atypical for the diagnosis, keep it and note the atypical presentation in the assessment rather than forcing canonical numbers - Never generate a value that contradicts the stated diagnosis

### Step 4 — Write the assessment

The assessment section must contain:

1. **Primary diagnosis** stated clearly on the first line 2. **Clinical reasoning** — one sentence explaining why this is the most likely diagnosis 3. **Differential diagnosis** — exactly 3 to 5 items, each with one sentence of supporting or refuting evidence 4. **Risk stratification** — include a validated clinical score where applicable (TIMI for ACS, GRACE for ACS, Killip class + Shock Index for STEMI/cardiogenic shock, CURB-65 for pneumonia, qSOFA for sepsis, Wells for PE, etc.). Killip class and Shock Index together are accepted as sufficient risk stratification for STEMI/cardiogenic shock cases.

### Step 5 — Write the management plan

The plan must be:

- **Specific**: write drug names, doses, routes, and frequencies. Do not write "start antibiotics" — write "Piperacillin-Tazobactam 4.5g IV q8h for 5 days" - **Organised by problem** using numbered headers - **Evidence-based**: management must reflect current standard of care for the diagnosis - **Complete**: include investigations to order, monitoring parameters, consults to request, and disposition

If you are uncertain about a specific dose, write "[drug name] — dose per local formulary/protocol" rather than inventing a dose.

### Important — Prescribing Safety

Generated plans must: - Be marked as educational/simulated, not a substitute for clinician judgment - Use "per local formulary/protocol" language when required patient variables (weight, renal function, allergies) are missing from the brief - List key contraindications and unknowns before medication recommendations when relevant patient data has not been provided - Never claim a plan is "definitive" or "standard of care" without full patient context (allergy status, renal/hepatic function, pregnancy status, weight, anticoagulation/bleeding risk) - Include a disclaimer footer in the HTML output stating the case is for educational and documentation purposes only

### Step 6 — Write `index.html`

Requirements for the HTML output:

- Professional medical document typography (Georgia or system serif font preferred) - White background, dark text — suitable for printing - Vital signs and lab results in HTML `<table>` elements - Critical findings (ST elevation, raised troponin, low BP, etc.) highlighted in a visually distinct callout box with red left border - @media print CSS rules so the document prints cleanly on A4/Letter - Tag every major section with `data-od-id` for comment-mode targeting:

```html <section data-od-id="hpi">...</section> <section data-od-id="vitals">...</section> <section data-od-id="pmh">...</section> <section data-od-id="examination">...</section> <section data-od-id="investigations">...</section> <section data-od-id="assessment">...</section> <section data-od-id="plan">...</section> ```

### Step 7 — Self-check against `references/checklist.md`

Before emitting `<artifact>`, run every P0 item in `references/checklist.md`. All P0 items must pass. Fix any failures before emitting.

Intended uses

  • Use Clinical Case Report when this documented workflow matches the task.

Implementation notes

  • Be marked as educational/simulated, not a substitute for clinician judgment
  • Use "per local formulary/protocol" language when required patient variables
  • List key contraindications and unknowns before medication recommendations
  • Never claim a plan is "definitive" or "standard of care" without full
  • Include a disclaimer footer in the HTML output stating the case is for

Try it, then tell us how it went

A copied prompt does not tell us whether the task worked. Report your result after checking the output. We collect your tool and outcome, never your files or prompt.