Learn how heart attack, endometriosis and ADHD show up differently in women, taught by a 3D teacher with trusted sources.
HealthTypeScript · JavaScriptGitHub @albaparsiOnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
No updates detected yet. GitHub releases will appear here.
Owners of public repositories under a personal GitHub account can sign in to verify ownership and manage the introduction themselves. No administrator approval is needed.
Sign in, generate a verification file, and commit it to this repository’s default branch within 30 minutes. Files on other branches or forks do not count. Then check the file below to manage the project without administrator approval. Keep the file to retain your access.
If needed, you will go to sign-in. After signing in, return here and press your chosen verification button again.
Run /nomorevibe verify in your project folder to verify ownership and manage the introduction yourself.
Verify in one minuteAsk about other verification options
The maker's introduction consists of the line above. Read the repository README for more details. Repository README ↗
Loading the full README.
Read the full README ↗Women's symptoms often get missed because medicine studied men. HerHealth is an AI teacher that turns any health topic into a lesson for women, with every fact checked against trusted sources.
A 3D teacher talks you through a real-life case, quizzes you, shows what's happening in the body, and answers your questions out loud. Every fact links to the trusted medical source and the exact sentence it came from.
Live: https://her-health-wheat.vercel.app
Built solo at Cal Hacks 13.0. Education, not medical advice.
Next.js 16 + TypeScript + Tailwind · React Three Fiber (3D avatar, lip sync from Web Audio) · Convex (backend, lesson cache, learner progress) · Elastic Serverless (semantic search over the indexed trusted sources, Agent Builder tutor) · MedlinePlus API · Claude API (lesson writing, fact review, Q&A) · Cartesia Sonic (voice) · Web Speech API (mic questions) · Chrome MV3 extension · Vercel
src/components/Classroom.tsx: the lesson flow (welcome, intro, checks, case, explanation)src/components/Avatar.tsx, TeacherVoice.tsx, AskTeacher.tsx: the teachersrc/components/MyProgress.tsx, convex/attempts.ts: quiz scores, Your progress and the impact pageconvex/lessons.ts: built-in lessons from Elastic, the lesson rules, and the fact-check (checkLesson)convex/custom.ts: Make a lesson, including the trusted-domain allowlist, MedlinePlus lookup and the
trusted-site web searchconvex/ask.ts, convex/agent.ts, convex/speech.ts: answers (Agent Builder tutor, Claude fallback) and voiceextension/: the Chrome extension (zipped to public/her-health-extension.zip)npm install
cp .env.example .env.local # fill in keys; never commit this file
npx convex dev # logs in, creates a dev deployment, writes Convex vars to .env.local
npm run dev # http://localhost:3000
The teacher is public/avatar.glb: the "brunette.glb" example avatar from
TalkingHead, created at Ready Player Me and licensed
CC BY-NC 4.0 (non-commercial use, attribution).
It has ARKit and viseme blendshapes for lip sync. It was compressed with
npx @gltf-transform/cli optimize ... --compress meshopt --texture-compress webp --texture-size 1024.
The app recolors her top and poses the arms at load time. For commercial use, swap in a CC0 model
such as TalkingHead's "mpfb.glb".
.env.local as
ELASTIC_URL and ELASTIC_API_KEY.data/sources.json:
npm run ingest # add -- --reset to rebuild the index from scratch
npm run search -- "heart attack symptoms women jaw" heart-attack
convex/agent.ts, scripts/agent.ts):
npm run agent -- "Can a heart attack feel like heartburn?"
Kibana's URL is worked out from ELASTIC_URL, or you can set KIBANA_URL.npx convex deploy (pushes the functions in convex/ to your production
deployment).npx convex env set --prod ANTHROPIC_API_KEY ...NEXT_PUBLIC_CONVEX_URL to the
production Convex URL. Vercel redeploys the frontend on every push to main.Unzip public/her-health-extension.zip, open chrome://extensions, turn on Developer mode, and
choose Load unpacked. To point it at your own deployment, edit extension/config.js.