SpeziVibe

Installable skills for building digital health software with AI coding tools.

Spezi Vibe packages reusable product, clinical, regulatory, interoperability, and platform guidance into skills that can be installed in minutes and used directly in real projects.

This work supports the broader Stanford Spezi mission of lowering the barrier to building thoughtful, high-quality digital health experiences.

Quick Start

Install all skills into your coding agent with a single command:

npx skills add StanfordSpezi/SpeziVibe --all

Or list them first to see what is available:

npx skills add StanfordSpezi/SpeziVibe --list

Need npx? Install Node.js — npm and npx are included. Then confirm with node -v && npx -v.

Manual Install

If you prefer not to use npx, download the latest spezivibe-skills.zip from Releases, unzip it, and copy the skill folders into your coding tool's skills directory (e.g., .claude/skills/ for Claude Code).

New To Vibe Coding?

Vibe coding is a practical way of building software with an AI coding partner. Instead of starting from a blank page, you work in natural language: you describe the app, workflow, research plan, or technical constraint, and the agent helps you explore the codebase, make changes, draft plans, and explain tradeoffs.

If you are just getting started, these tools are a good place to begin:

• Claude Code for terminal-based AI coding workflows
• OpenAI Codex for agentic software engineering with skills support
• Gemini CLI for an open-source terminal coding agent

If you prefer a more guided starting point, Claude and Codex also have desktop app experiences in addition to CLI workflows. A practical way to begin is:

1. choose a coding tool such as Claude, Codex, or another supported agent
2. choose whether you want to work in a desktop app or in the CLI
3. create or sign in to your account and then follow the instructions below to install npx and the Spezi Vibe skills

Install npx

The easiest way to get npx is to install Node.js. npm and npx are included with standard Node.js installations.

1. Install a current Node.js release from nodejs.org.
2. Confirm the tools are available:

node -v
npm -v
npx -v

3. Use npx skills to install Spezi Vibe into your coding agent.

Install Skills

We use the skills tool for installing and sharing reusable skills across supported agents.

Install every skill from this repository:

npx skills add StanfordSpezi/SpeziVibe --all

If you want to target a specific agent, add -a claude-code, -a codex, or another supported agent:

npx skills add StanfordSpezi/SpeziVibe --skill '*' -a claude-code

Where To Start

The fastest way to get started is to run build-an-app. Describe what you want to build and it figures out which planning skills you need, runs them in order, and hands off to implementation. You do not need to know the individual skills or their order — the orchestrator handles that for you.

If you prefer to run skills individually, here is the process they follow. Not every project needs every skill — use what fits and skip what does not.

1. Define the Need

Understand the clinical or operational problem before deciding on a product. Investigate the problem space, identify affected stakeholders, and refine a need statement until it is specific, evidence-grounded, and free of embedded solutions.

Use: biodesign-needs-finding

2. Choose a Platform

Decide whether the project is better suited to the React Native Template App or the Spezi Template Application for Apple Platforms, then clone the selected template.

Use: spezi-platform-selection

3. Plan the Product

Design the user experience, data model, and compliance posture. Run whichever skills are relevant — not all are required for every project.

• digital-health-ux-planning — user journeys, onboarding, engagement, and day-to-day workflows
• health-data-model-planning — core entities, relationships, and FHIR-oriented data structures
• fhir-data-model-design — map clinical concepts into interoperable FHIR resources and terminology
• digital-health-compliance-planning — privacy, regulatory, and governance expectations
• digital-health-study-planning — study protocol, consent, and data collection (only if tied to a research study)

4. Plan the Build

Feed your planning outputs into app-build-planner to get a sequenced implementation plan. The output is a structured document (docs/implementation-plan.md) saved in your cloned template repository.

5. Build the App

Open the template repository you cloned in step 2 and work through docs/implementation-plan.md milestone by milestone with your coding agent. Build each one, verify it works, then move to the next.

6. Keep Learning (Optional)

Set up a project wiki to accumulate knowledge as you build. Every interview, paper, clinical observation, and competitive finding gets ingested into a persistent, AI-maintained knowledge base that compounds over the life of the project. Planning documents from earlier steps seed the wiki automatically.

Use: project-wiki

Skill Catalog

This repository is organized for npx skills, with one skill per folder under skills/.

build-an-app

Walk through the full process of building a digital health app — describe what you want to build and it figures out which planning skills you need, runs them in order, and hands off to implementation. Start here if you are unsure where to begin.

npx skills add StanfordSpezi/SpeziVibe --skill build-an-app

spezi-platform-selection

Choose the right app foundation — helps you decide whether a project is better served by the React Native Template App or the Spezi Template Application for Apple Platforms, then points the coding agent at the right next steps.

npx skills add StanfordSpezi/SpeziVibe --skill spezi-platform-selection

biodesign-needs-finding

Turn an idea into a real clinical need — guides a team through a Stanford Biodesign-style needs-finding process so they define the problem well before jumping to a solution.

npx skills add StanfordSpezi/SpeziVibe --skill biodesign-needs-finding

digital-health-study-planning

Plan a study around the app — helps shape a digital health study or research workflow, including recruitment, consent, assessments, schedules, and outcome measures.

npx skills add StanfordSpezi/SpeziVibe --skill digital-health-study-planning

digital-health-compliance-planning

Think through privacy and regulatory risk early — helps teams reason about HIPAA, IRB, FDA, GDPR, and adjacent compliance questions before implementation gets too far ahead.

npx skills add StanfordSpezi/SpeziVibe --skill digital-health-compliance-planning

health-data-model-planning

Shape the app's health data backbone — helps define core health concepts, entities, relationships, lifecycle states, and interoperability needs before implementation begins.

If you are working inside the React Native Template App, the repo-local data-model skill carries the implementation-focused guidance for app entities, FHIR mappings, storage, and sync behavior.

npx skills add StanfordSpezi/SpeziVibe --skill health-data-model-planning

digital-health-ux-planning

Design a patient- and clinician-friendly experience — helps plan onboarding, core journeys, engagement loops, and day-to-day workflows for digital health products.

npx skills add StanfordSpezi/SpeziVibe --skill digital-health-ux-planning

fhir-data-model-design

Map clinical concepts to FHIR — translates clinical requirements into a FHIR R4-oriented data model with concrete resources, relationships, and implementation guidance.

FHIR implementation review is now part of the React Native Template App's repo-local fhir skill, where it can stay aligned with the actual app mappings and services.

npx skills add StanfordSpezi/SpeziVibe --skill fhir-data-model-design

app-build-planner

Turn planning outputs into a milestone-based build plan — extracts features from your UX, data model, compliance, and study planning work, maps them to available packages or framework modules, and sequences everything into milestones you can build and review one at a time.

npx skills add StanfordSpezi/SpeziVibe --skill app-build-planner

project-wiki

Turn your project into a compounding knowledge base — set up a persistent, AI-maintained wiki that grows with every interview, paper, clinical observation, and competitive finding. Inspired by Karpathy's LLM Wiki pattern, adapted for the Stanford Biodesign innovation process. Seeds automatically from other SpeziVibe planning documents.

npx skills add StanfordSpezi/SpeziVibe --skill project-wiki

keep-a-changelog-generator

Draft changelogs people can actually read — turns git history into structured changelog entries using the Keep a Changelog format and clearer user-facing language.

npx skills add StanfordSpezi/SpeziVibe --skill keep-a-changelog-generator

release-notes-generator

Summarize a release clearly — helps generate release notes that explain features, fixes, and migration concerns in a concise way for real users and collaborators.

npx skills add StanfordSpezi/SpeziVibe --skill release-notes-generator

Philosophy

These skills are meant to be:

• practical enough to support real implementation decisions
• reusable across multiple digital health projects
• approachable for teams that are new to AI coding workflows
• structured cleanly enough to install and share with standard skills tooling

License

This project is licensed under the MIT License. See LICENSE.md and LICENSES/ for details.

Contributors

This project is developed as part of the Stanford Mussallem Center for Biodesign at Stanford University.
See CONTRIBUTORS.md for the current contributor list.