
Installable skills that help AI coding tools plan, build, and ship digital health apps.
Copied!
What is SpeziVibe
SpeziVibe is a collection of installable skills for AI coding tools. Skills guide you from needs analysis and compliance through data modeling and UX — then help you build with Spezi templates, pre-built modules, and a living project knowledge base.
From Stanford's Spezi initiative, built to lower the barrier to high-quality digital health experiences.
How it works
One command adds every skill to your AI coding tool. Works with anything that supports installable skills or custom instructions.
Describe what you want to build. Skills walk you through needs analysis, compliance, data modeling, UX, and study design — producing structured docs along the way.
Start from a Spezi template app and add pre-built modules. Your planning docs guide the AI as it wires everything together.
Generate changelogs, release notes, and maintain a living project wiki that keeps your team's knowledge organized as you iterate.
Skills
Each skill covers a specific aspect of digital health development — from planning through building and maintenance.
Describe what you want to build. This skill figures out which steps apply, walks you through planning, and sets you up to build with the right template and modules.
Choose between the React Native Template App and the Spezi Template Application for Apple Platforms, set up your dev environment, and clone the right starter repo.
Walks you through a Stanford Biodesign-style needs-finding process to define a clear problem statement before jumping to solutions. Produces a need-statement.md.
Helps plan a research protocol — enrollment, consent, data collection, assessment schedules, and outcome measures. Produces a study-brief.md.
Helps you reason through HIPAA, IRB, FDA, GDPR, and related compliance questions early. Produces a compliance-brief.md with identified domains and recommended controls. Not legal advice.
Helps define health data entities, relationships, lifecycle states, and interoperability needs. Biased toward FHIR for clinically meaningful data. Produces a data-model-brief.md.
Plans user journeys, onboarding, engagement, and day-to-day workflows for patients and clinicians. Platform-agnostic — no wireframes, just goals and decision points. Produces a ux-brief.md.
Maps clinical data types to specific FHIR R4 resources, terminology bindings, and relationships. Expert-driven recommendations with sample JSON. Produces a fhir-data-model.md.
Reads the planning docs from other skills and produces a milestone-based implementation plan with tasks, dependencies, and verification criteria. Produces an implementation-plan.md.
Set up a persistent, AI-maintained knowledge base for your project. Add interviews, papers, and clinical observations — the AI integrates them across interlinked wiki pages, flags contradictions, and keeps everything current.
Generates changelog entries from git history in the Keep a Changelog format. Groups commits by category and translates technical messages into user-facing language.
Creates user-facing release notes from git history with feature highlights, fixes, breaking changes, and migration guidance.
Compatible with
Skills are tool-agnostic — they work with anything that supports installable skills or custom instructions.
One command. Every skill. Plan, build, and ship digital health apps.
Copied!