8 templates in this category, each with its compliance scope, included features and target platforms. Open any template to preview the screens and BAA documentation before you build.
AI Medical Scribe is a free, HIPAA-compliant healthcare AI template that captures a clinical visit by dictation and drafts a SOAP/HPI/MDM note for the clinician to review and sign. The template ships with browser-based audio capture, manual speaker turns, draft generation against AWS Bedrock, Anthropic Claude, Whisper or Azure OpenAI under a signed BAA (with a transparent rule-based fallback when no model is configured), ICD-10 and CPT code suggestions, and FHIR-shaped EHR export. A three-role RBAC console (clinician, scribe, admin) covers the encounter list, scribe workspace, AI-vs-rule-based settings, and an immutable audit log. Every draft is labelled AI or rule-based and is not the record until the clinician signs. Customize inside VertiComply, generate the full React + Vite + FastAPI + Postgres stack with HIPAA scaffolding, then export to GitHub. No platform lock-in.
- Audio capture
- Speaker turns
- SOAP drafting
- Code suggestions
- EHR export
11 screens included
Public: Home (Ambient AI scribe hero + sample note) · How it works (Five steps, one signature)
Console: Console sign in (RBAC staff sign-in (3 roles)) · Encounters (Visits by status (recording/draft/signed)) · Scribe workspace (Dictate → generate SOAP → codes → sign (AI badge)) · Patients (Patient records) · AI settings (Live-model vs rule-based status + disclosures) · Audit log (Immutable audit (admin))
Legal: Privacy (Privacy & HIPAA notice) · Terms (Terms of use) · Accessibility (WCAG 2.2 AA statement)
Integrations: AWS Bedrock, Anthropic Claude, Whisper, Azure OpenAI. Built for: providers, clinicians.
AI Triage Chatbot is a free, HIPAA-compliant healthcare AI template that runs a patient-facing symptom triage conversation, lands on a deterministic ESI severity disposition, and hands off to a clinician when red flags appear. The chief complaint and red-flag intake are conversational, but the ESI-style severity decision is rule-based — never produced by a model — and every disposition is written to an immutable audit log. A plain-language explanation is rendered by AWS Bedrock or Anthropic Claude under a signed BAA when configured, with an honest rule-based fallback otherwise. The template ships a patient portal plus a three-role RBAC console (triage nurse, clinician, admin) with a handoff queue, analytics, HIPAA e-sign consents, and secure messaging. Decision support, not a diagnosis. Customize inside VertiComply, generate the React + Vite + FastAPI + Postgres stack with HIPAA scaffolding, then export to GitHub. No platform lock-in.
- Conversational intake
- ESI severity
- Red-flag escalation
- Clinician handoff
- Decision audit
16 screens included
Public: Home (AI triage hero + ESI severity sample) · How it works (Safe triage, then a human) · Patient sign in (Patient sign-in) · Create account (Patient signup)
Triage console: Care team sign in (RBAC staff sign-in (3 roles)) · Handoff queue (Escalated checks, most urgent first) · Session detail (Disposition, answers, sign-off (RBAC)) · Analytics (ESI mix, dispositions, escalations) · Audit log (Immutable audit (admin))
Patient: Check a symptom (Complaint → red-flag questions → ESI disposition) · My checks (Past triage sessions) · Messages (Secure triage-team messaging) · Privacy & consent (Triage disclaimer + HIPAA e-sign)
Legal: Privacy (Privacy & HIPAA notice) · Terms (Terms of use) · Accessibility (WCAG 2.2 AA statement)
Integrations: AWS Bedrock, Anthropic Claude. Built for: patients, providers.
AI Radiology Vision is a HIPAA-compliant healthcare AI template that gives radiologists a reporting copilot for chest, mammography, and other structured reads — STAT-first worklist, rule-based RADS classification, critical-results workflow, and an AI-drafted impression for the radiologist to sign. Lung-RADS and BI-RADS classification with management recommendations are deterministic; the AWS Bedrock impression draft is the only AI-generated text. The critical-results loop enforces flag, communicate, and acknowledge before sign-off, all written to an immutable audit log. The template is built for HIPAA, BAA, FDA 510(k), ISO 13485, and HITRUST programs, integrates with DICOM and Orthanc for study ingestion, and ships a three-role RBAC console (technologist, radiologist, admin). Pixel-level image inference is a production computer-vision integration, honestly disclosed rather than simulated. Customize inside VertiComply, generate the full React + FastAPI + Postgres stack with compliance scaffolding, then export to GitHub. No platform lock-in.
- Structured reporting
- RADS classification
- Critical-results workflow
- AI-drafted impression
- Audit log
11 screens included
Public: Home (Reading-room hero + RADS sample) · How it works (The copilot around the read)
Console: Console sign in (RBAC staff sign-in (3 roles)) · Worklist (Studies, STAT & unread first) · Study reader (Findings, Lung-RADS, impression, sign (RBAC)) · Critical results (Communicate + acknowledge workflow) · Analytics (Modality, RADS distribution) · Audit log (Immutable audit (admin))
Legal: Privacy (Privacy & HIPAA notice) · Terms (Terms of use) · Accessibility (WCAG 2.2 AA statement)
Integrations: AWS Bedrock, DICOM, Orthanc. Built for: providers, admins.
Clinical Documentation Copilot is a free, HIPAA-compliant healthcare AI template that helps clinicians refine an existing note, lift ICD-10 and CPT (E&M) codes, summarize the chart, and ask the chart grounded questions with cited answers. Note refinement runs against AWS Bedrock, Anthropic Claude, or Azure OpenAI under a signed BAA when configured, with an honest rule-based tidy otherwise. Code lift and chart Q&A retrieval are rule-based — answers only come from retrieved chart lines and cite them, declining when nothing matches the question. The template ships a three-role RBAC console (clinician, coder, admin), FHIR-shaped EHR export, analytics on grounded-answer rate, an immutable audit log, and SOC 2 plus WCAG 2.2 AA scaffolding. The clinician finalizes every note — drafts are not the record until signed. Customize inside VertiComply, generate the full React + FastAPI + Postgres stack, then export to GitHub. No platform lock-in.
- Note refinement
- Code lift
- Chart summary
- Citation grounding
- EHR export
12 screens included
Public: Home (Editorial hero + grounded-answer sample) · How it works (A copilot, not an author)
Console: Console sign in (RBAC staff sign-in (3 roles)) · Notes (Encounter notes by status) · Note workspace (Refine + code lift + chart Q&A + finalize) · Patients (Charts + facts) · Patient detail (Chart facts, summary, grounded Q&A) · Analytics (Notes by status, grounded count) · Audit log (Immutable audit (admin))
Legal: Privacy (Privacy & HIPAA notice) · Terms (Terms of use) · Accessibility (WCAG 2.2 AA statement)
Integrations: AWS Bedrock, Anthropic Claude, Azure OpenAI. Built for: providers, admins.
AI Prior Auth Assistant is a HIPAA-compliant healthcare AI template that assembles a clean prior-authorization packet for provider and utilization-management teams — rule-based medical-necessity criteria check, AI-drafted justification letter, attachment suggestions, and a human reviewer decision with an appeals path. The meets, partial, or not-met criteria verdict is rule-based and never produced by a model; only the justification letter is drafted by AWS Bedrock under a signed BAA, with an honest structured template fallback when no model is configured. The template ships a three-role RBAC console (provider, UM reviewer, admin), CoverMyMeds and Surescripts integration hooks, member records by payer, approval-rate analytics, SOC 2 plus WCAG 2.2 AA scaffolding, and an immutable audit log of every decision. The AI drafts text only and never decides. Customize inside VertiComply, generate the React + FastAPI + Postgres stack with compliance scaffolding, then export to GitHub. No platform lock-in.
- Criteria check
- Justification drafts
- Attachment suggestions
- Decision & appeals
12 screens included
Public: Home (AI prior-auth hero + criteria/letter sample) · How it works (Four steps, one clean packet)
Console: Console sign in (RBAC staff sign-in (3 roles)) · Requests (Prior-auth pipeline board) · New request (Service → criteria → review wizard) · Request detail (Criteria, AI letter, decision, appeal (RBAC)) · Members (Member records + payer) · Analytics (Approval rate, status + criteria mix) · Audit log (Immutable audit (admin))
Legal: Privacy (Privacy & HIPAA notice) · Terms (Terms of use) · Accessibility (WCAG 2.2 AA statement)
Integrations: AWS Bedrock, CoverMyMeds, Surescripts. Built for: providers, admins.
AI Drug Discovery Lab is a SOC 2, GxP, and ISO 27001-aligned healthcare AI template that gives discovery researchers a transparent triage workbench — compound library with rule-based Lipinski Rule of Five and Veber druglikeness, property-window target screening, grounded literature synthesis with citations, and AI-drafted compound rationales. The druglikeness verdict, target screen filter, and literature retrieval are deterministic and reproducible; AWS Bedrock and Pinecone (with RDKit chemistry under the hood) draft only the prose rationale, never an affinity number. Real molecular docking, ML affinity prediction, and IND workflows are honestly disclosed as production integrations rather than simulated. The template ships a three-role RBAC console (researcher, lead, admin) with candidate shortlist, target profiles, druglikeness analytics, and an immutable audit log. Customize inside VertiComply, generate the full React + FastAPI + Postgres stack with GxP-aware scaffolding, then export to GitHub. No platform lock-in.
- Druglikeness screening
- Target screening
- Literature synthesis
- Candidate shortlist
12 screens included
Public: Home (Discovery-bench hero + druglikeness sample) · How it works (Transparent triage on one bench)
Console: Bench sign in (RBAC staff sign-in (3 roles)) · Library (Compounds with druglikeness verdict + score) · Compound detail (Properties, Lipinski/Veber, rationale, shortlist (RBAC)) · Targets (Target profiles + run screen (filter)) · Literature (Grounded synthesis with citations) · Analytics (Druglikeness + status mix) · Audit log (Immutable audit (admin))
Legal: Privacy (Privacy & data notice) · Terms (Terms of use) · Accessibility (WCAG 2.2 AA statement)
Integrations: AWS Bedrock, Pinecone, RDKit. Built for: researchers, admins.
AI Mental Health Companion is a free, HIPAA-compliant healthcare AI template that gives members a daily mental-health support surface — mood check-ins, journaling with an empathetic AI reflection, validated PHQ-9 and GAD-7 screeners with exact scoring, a CBT and DBT skills library, and a personal safety plan. AWS Bedrock, Anthropic Claude, and Twilio Voice integrate under a signed BAA when configured, with an honest rule-based reflection fallback otherwise. Crisis detection is rule-based and never delegated to the model; it surfaces crisis resources and routes the member to a human counselor handoff. The template ships a member portal plus a three-role RBAC care console (counselor, clinician, admin) with severity analytics, alerts queue, SOC 2 plus WCAG 2.2 AA scaffolding, and an immutable audit log. Supportive self-help, not therapy or an emergency service. Customize inside VertiComply, generate the React + FastAPI + Postgres stack, then export to GitHub. No platform lock-in.
- Mood tracking
- Journaling
- PHQ-9 / GAD-7 screeners
- CBT/DBT skills
- Crisis safety & human escalation
18 screens included
Public: Home (Mental-health companion hero + mood check-in) · How it works (Daily support with a safety net) · Member sign in (Member sign-in) · Create account (Member signup)
Care console: Care team sign in (RBAC staff sign-in (3 roles)) · Care alerts (Crisis/elevated queue, crisis first) · Member detail (Moods, screeners, alerts, sign-off (RBAC)) · Analytics (Severity mix, engagement, alerts) · Audit log (Immutable audit (admin))
Member: Today (Daily mood check-in + trend + care team) · Journal (Journaling with AI/rule-based reflection) · Companion (Empathetic chat (crisis-guardrailed)) · Check-ins (PHQ-9 / GAD-7 validated screeners) · Skills (CBT/DBT skills library) · Safety plan (Personal crisis safety plan)
Legal: Privacy (Privacy & HIPAA notice) · Terms (Terms of use) · Accessibility (WCAG 2.2 AA statement)
Integrations: AWS Bedrock, Anthropic Claude, Twilio Voice. Built for: patients, providers.
AI Care Coordinator is a free, HIPAA-compliant healthcare AI template for population-health and care-coordination teams — a rule-based engine detects care gaps (overdue A1c, BP, mammogram, colorectal screen, wellness visit, flu shot, lipids) across the patient panel, converts them into prioritized tasks, and drafts patient outreach nudges. AWS Bedrock drafts the outreach copy under a signed BAA when configured, with an honest template fallback otherwise; Twilio handles SMS and voice delivery; FHIR R4 integration syncs panel data from the EHR. Detection, scheduling, and closure decisions are never delegated to a model — a clinician closes every gap. The template ships a three-role RBAC console (care coordinator, clinician, admin) with per-patient and per-panel closure rings, SOC 2 plus WCAG 2.2 AA scaffolding, and an immutable audit log. Customize inside VertiComply, generate the React + FastAPI + Postgres stack, then export to GitHub. No platform lock-in.
- Gap-in-care detection
- Task scheduling
- Plan-driven nudges
- Human-in-the-loop
11 screens included
Public: Home (Care-coordination hero + closure ring) · How it works (Detect, schedule, nudge, close the loop)
Console: Console sign in (RBAC staff sign-in (3 roles)) · Panel (Patients ranked by overdue gaps + closure ring) · Patient detail (Care gaps, AI nudge, tasks, close gap (RBAC)) · Tasks (Panel-wide task queue) · Analytics (Closure rate, open gaps by measure) · Audit log (Immutable audit (admin))
Legal: Privacy (Privacy & HIPAA notice) · Terms (Terms of use) · Accessibility (WCAG 2.2 AA statement)
Integrations: AWS Bedrock, Twilio, FHIR R4. Built for: providers, clinicians.