ScribeVault Specification

Home / ScribeVault / ScribeVault Specification

ScribeVault

Encrypted, HIPAA-Compliant, AI-Driven.

ScribeVault is an open-source AI-driven clinical documentation platform designed to function as a real-time medical scribe and post-visit documentation system for hospitals, clinics, private practices, and other healthcare environments.

ScribeVault captures clinical conversations, transforms them into structured documentation, assists with medical coding, supports voice-driven workflows, maintains encrypted patient records, and provides interoperability for clinical and insurance workflows. The system is designed for fully self-hosted deployment with user-selectable storage and a modular AI architecture.

Purpose

ScribeVault is designed to reduce the administrative burden of clinical documentation while keeping clinicians in control of the final medical record.

The system shall:

Capture clinical conversations through real-time voice transcription
Process uploaded recordings for post-visit documentation
Transform conversations into structured clinical notes
Organize information into HPI, ROS, Examination, Assessment, and Plan
Suggest ICD-10, CPT, HCPCS, and SNOMED codes
Provide detailed coding information for clinician review and approval
Prepare approved documentation for insurance submission workflows
Support medication reconciliation
Identify and surface documented allergies
Provide specialty-specific documentation templates
Support voice-controlled clinical workflows
Generate patient-facing summaries
Record and document consent
Provide patient education materials
Maintain an encounter timeline
Highlight uncertain or low-confidence information
Export clinical documentation through interoperable formats
Keep protected health information encrypted and under the control of the deploying organization

Design Principles

ScribeVault shall follow these principles:

Clinician-controlled documentation
Human approval before clinical or billing submission
Privacy-first architecture
Encryption of protected health information
Fully self-hosted operation
User-selectable data storage
Modular AI architecture
Vendor-neutral integrations
Interoperability
Auditable clinical workflows
Specialty-aware documentation
Explainable AI-assisted suggestions
Separation of source conversation data from generated clinical interpretation
Clear distinction between AI suggestions and clinician-approved documentation

Core Modules

Clinical Encounter Management

The Clinical Encounter Management module shall provide the central workspace for creating, managing, reviewing, and completing clinical encounters.

Features shall include:

Patient identification and encounter association
Encounter creation and editing
Encounter status management
Provider assignment
Specialty and visit-type selection
Encounter date and time
Clinical documentation status
Draft, review, approved, and finalized states
Encounter history
Encounter timeline
Links between source recordings, transcripts, notes, codes, consent records, and exports
Audit history for significant encounter changes
Support for individual and organizational clinical workflows

Real-Time Voice Scribing

The Real-Time Voice Scribing module shall continuously process clinical conversations during an encounter and transform spoken information into structured documentation.

Features shall include:

Real-time speech recognition
Speaker identification
Clinical conversation segmentation
Medical terminology recognition
Context-aware transcription
Clinical information extraction
Real-time note generation
HPI generation
ROS generation
Examination documentation
Assessment generation
Plan generation
Medication identification
Allergy identification
Symptom extraction
Diagnosis candidate extraction
Procedure identification
Clinical decision extraction
Important statement detection
Confidence indicators
Transcript-to-note traceability
Support for clinical environments with background noise
Continuous saving of transcription and generated content
Pause and resume controls

Post-Visit Documentation

The Post-Visit Documentation module shall process recorded clinical conversations or other authorized source material after an encounter.

Features shall include:

Audio ingestion
Transcript generation
Clinical conversation analysis
Structured SOAP documentation
HPI generation
ROS generation
Examination documentation
Assessment generation
Plan generation
Diagnosis candidate identification
Procedure identification
Medication extraction
Allergy extraction
Follow-up recommendations
Clinical event extraction
Source-to-document traceability
Clinician review and editing
Confidence highlighting
Final documentation approval

Clinical Documentation Engine

The Clinical Documentation Engine shall transform source information into standardized clinical documentation while preserving the distinction between source information and AI-generated interpretation.

Features shall include:

Structured HPI
Structured ROS
Structured examination
Structured assessment
Structured plan
SOAP notes
Progress notes
Consultation notes
Procedure documentation
Discharge documentation
Referral documentation
Follow-up documentation
Custom documentation sections
Provider-specific preferences
Specialty-specific formatting
Source attribution
Confidence scoring
Missing-information detection
Contradiction detection
Documentation completeness checks
Clinician editing
Approval workflows

Specialty Templates

The Specialty Templates module shall provide documentation structures appropriate for different medical specialties and encounter types.

Supported template categories shall include:

Internal Medicine
Family Medicine
Pediatrics
Emergency Medicine
Psychiatry
Surgery
Orthopedics
Obstetrics
Gynecology
Neurology
Cardiology
Dermatology
Gastroenterology
Pulmonology
Oncology
Endocrinology
Urology
Nephrology
Ophthalmology
Otolaryngology
Rheumatology
Infectious Disease
Physical Medicine and Rehabilitation
Pain Management
Radiology
Anesthesiology
Geriatrics
Critical Care
Other specialty-defined workflows

Templates shall adapt to:

Specialty
Visit type
Clinical setting
Provider preferences
Documentation requirements
Organizational requirements
Patient population
Procedure type

The system shall support creation, modification, import, export, and versioning of custom templates.

Medical Coding Assistance

The Medical Coding Assistance module shall analyze approved or draft clinical documentation and provide detailed coding suggestions for clinician review.

Features shall include:

ICD-10 code suggestions
CPT code suggestions
HCPCS code suggestions
SNOMED CT concept suggestions
Code descriptions
Supporting documentation references
Code confidence indicators
Missing-documentation warnings
Coding conflict detection
Documentation-to-code traceability
Multiple possible code suggestions
Clinician selection and rejection
Coding approval workflow
Coding audit history
Insurance documentation preparation

AI-generated codes shall remain suggestions until reviewed and approved by an authorized clinician or qualified billing professional.

Medication Reconciliation

The Medication Reconciliation module shall organize medication information identified during an encounter.

Features shall include:

Medication extraction from conversations
Current medication lists
New medications
Discontinued medications
Dosage information
Frequency information
Route information
Medication changes
Medication reconciliation status
Potential duplicate medication identification
Medication discrepancy highlighting
Clinician review
Patient-facing medication summaries

Allergy and Clinical Safety Review

The Allergy and Clinical Safety Review module shall identify relevant clinical safety information present within available encounter data.

Features shall include:

Allergy extraction
Medication allergy identification
Reaction documentation
Severity documentation
Allergy list comparison
Potential medication-allergy conflicts
Missing allergy information warnings
Clinical safety flags
Review and acknowledgment workflows

The module shall provide information for clinical review and shall not independently authorize or execute clinical treatment decisions.

Voice Command and Voice Navigation

The Voice Command module shall provide hands-free control of ScribeVault.

Supported commands shall include:

Start and stop transcription
Pause and resume transcription
Navigate between documentation sections
Create an assessment
Create a plan
Summarize recent conversation
Insert approved documentation phrases
Open medication information
Open allergy information
Flag documentation
Flag information for billing review
Request patient summary
Request patient education
Export documentation
Save an encounter
Mark documentation for review

The system shall support optional local voice-command processing.

Patient-Facing Documentation

The Patient-Facing Documentation module shall transform approved clinical information into understandable patient-facing materials.

Features shall include:

Patient visit summaries
Plain-language explanations
Medication summaries
Follow-up instructions
Care instructions
Patient education materials
Procedure explanations
Diagnosis explanations
Questions for follow-up
Personalized education based on approved encounter information
Clinician review before release
Patient-accessible documentation

Patient-facing content shall clearly distinguish educational information from formal clinical documentation.

Consent Management

The Consent Management module shall provide mechanisms for documenting patient consent associated with clinical encounters and recordings.

Features shall include:

Consent capture
Consent status
Consent timestamps
Consent scope
Consent documentation
Optional consent recording
Consent transcription
Encrypted consent records
Association between consent and encounter
Consent review
Consent withdrawal documentation
Audit history

Encounter Timeline

The Encounter Timeline module shall provide a chronological representation of significant encounter events.

Features shall include:

Conversation segments
Transcription events
Documentation changes
Clinical decisions
Flags
Confidence changes
Medication events
Allergy events
Coding events
Consent events
Clinician approvals
Export events
Patient-facing document generation
Audit events

Confidence and Source Traceability

The Confidence and Source Traceability module shall help clinicians understand how AI-generated documentation was produced.

Features shall include:

Confidence indicators
Highlighting of uncertain information
Source transcript references
Source audio references where available
Documentation-to-source mapping
Coding-to-documentation mapping
Identification of inferred information
Identification of directly stated information
Contradiction warnings
Missing-information warnings
Clinician correction tracking
Review status

AI-generated information shall remain distinguishable from information directly documented or approved by a clinician.

Encrypted Patient Records

The Encrypted Patient Records module shall provide secure storage and management of protected health information.

Features shall include:

Encryption of patient records
Encryption of clinical documentation
Encryption of recordings
Encryption of transcripts
Encryption of consent records
Encryption of exported records where supported
Access control
Role-based permissions
Authentication
Session management
Audit logging
Data retention controls
Data deletion controls
Backup and recovery controls
User-selectable storage
Support for organizational data ownership
Self-hosted operation

Supported storage models shall be modular and shall allow deployments to select appropriate storage according to organizational requirements.

Compliance and Privacy

The Compliance and Privacy module shall provide controls intended to support regulated healthcare deployments.

The system shall be designed to support:

HIPAA requirements
Privacy and security controls
Auditability
Access control
Minimum necessary access principles
Data retention policies
Data deletion policies
Consent management
Security event logging
Administrative controls
Data portability
GDPR-oriented privacy controls where applicable
Security controls aligned with SOC 2 principles where applicable

ScribeVault shall not represent a deployment as legally compliant solely because the software implements these features. Compliance shall depend on the complete deployment, configuration, organizational policies, contractual obligations, and applicable law.

Interoperability

The Interoperability module shall provide standardized exchange of clinical information.

Features shall include:

CCD export
C-CDA export
Structured clinical document generation
Clinical data export
Patient record export
Medication data export
Allergy data export
Encounter data export
Coding data export
Optional FHIR interoperability
Import and export validation
Interoperability audit logging
Data mapping
External system integration

AI Orchestration

The AI Orchestration module shall coordinate specialized AI capabilities while maintaining separation between transcription, extraction, reasoning, generation, coding assistance, and validation.

Features shall include:

Modular AI providers
Local AI processing
Optional external AI services
Model selection
Task-specific model routing
AI provider failover
AI provider configuration
Prompt and instruction management
Context management
Output validation
Confidence assessment
Source traceability
AI request auditing
Provider-specific privacy controls
Human approval checkpoints

The AI architecture shall allow models and providers to be replaced without redesigning the clinical workflow.

Review and Approval

The Review and Approval module shall provide clinicians with final control over AI-generated clinical documentation.

Features shall include:

Side-by-side transcript and documentation review
Editable documentation
Suggested code review
Medication review
Allergy review
Confidence review
Missing-information warnings
Contradiction warnings
Change tracking
Approval states
Finalization
Electronic sign-off support
Billing readiness review
Export authorization

No AI-generated clinical documentation or medical coding shall be considered final solely because it was generated by the system.

Clinical Workspace

The Clinical Workspace module shall provide the primary user interface for clinical documentation.

Features shall include:

Dual-pane transcript and structured note view
Real-time documentation updates
Specialty template selection
Coding review panel
Medication panel
Allergy panel
Patient summary panel
Confidence indicators
Timeline panel
Consent panel
Review and approval controls
Search within encounter
Voice navigation
Customizable workspace layout
Accessibility controls

Optional Plugin Modules

FHIR Integration Plugin

Provides optional FHIR interoperability for systems that require direct structured healthcare data exchange.

Features may include:

FHIR R4 support
Patient resources
Encounter resources
Observation resources
Condition resources
Medication resources
AllergyIntolerance resources
Practitioner resources
DocumentReference resources
External FHIR server integration
FHIR import and export

EHR Integration Plugin

Provides optional integration with external electronic health record systems.

Features may include:

EHR authentication
Clinical document transfer
Patient lookup
Encounter synchronization
Documentation export
Medication synchronization
Allergy synchronization
Coding synchronization
Integration-specific audit logging

Insurance Integration Plugin

Provides optional insurance workflow integrations.

Features may include:

Insurance documentation packages
Coding export
Claim-supporting documentation
Eligibility information integration
Payer-specific workflows
Submission status tracking
Documentation validation
Billing workflow integration

Patient Portal Plugin

Provides optional patient access to approved information.

Features may include:

Patient authentication
Patient summaries
Approved clinical documents
Medication information
Follow-up instructions
Education materials
Consent records
Secure messaging integration
Document download
Access auditing

Medical Reference Plugin

Provides optional access to external medical reference resources.

Features may include:

Medication references
Medical terminology
Clinical reference material
Coding references
Specialty references
Reference source attribution
Version tracking
Source validation

Translation and Multilingual Plugin

Provides optional multilingual clinical documentation capabilities.

Features may include:

Multilingual transcription
Translation of approved documentation
Patient-facing translation
Medical terminology preservation
Language detection
Interpreter workflow support
Translation confidence indicators
Original-language source preservation

Advanced Speech Plugin

Provides enhanced speech-processing capabilities.

Features may include:

Advanced speaker separation
Specialized medical vocabulary
Accent adaptation
Noise reduction
Multiple-language recognition
Voice profiles
Custom vocabulary
Local wake-word support
Speech confidence analysis

Advanced Analytics Plugin

Provides optional organizational analytics.

Features may include:

Documentation time analysis
Encounter workflow metrics
Documentation completeness metrics
Coding suggestion statistics
Clinician review metrics
Error correction analysis
AI confidence trends
Operational reporting
Exportable analytics

Analytics shall be designed to avoid unnecessary exposure of protected health information.

Organization Administration Plugin

Provides advanced administrative controls for larger deployments.

Features may include:

Organization management
Department management
Provider management
Role management
Permission management
Policy configuration
Template administration
AI provider administration
Storage administration
Audit administration
Compliance reporting

Custom AI Provider Plugin

Allows deployments to connect additional AI providers or internally managed models.

Features may include:

Provider registration
Model registration
Endpoint configuration
Authentication configuration
Task routing
Provider-specific privacy settings
Model selection
Failover configuration
Usage monitoring
AI request auditing

Custom Storage Plugin

Allows deployments to connect additional storage systems.

Features may include:

Storage provider registration
Encryption configuration
Record storage
Recording storage
Document storage
Backup integration
Retention configuration
Data migration
Storage health monitoring

Clinical Specialty Plugin

Allows organizations and clinical specialists to add specialty-specific capabilities without modifying the core platform.

Features may include:

Specialty templates
Specialty terminology
Specialty workflows
Specialty documentation sections
Specialty coding guidance
Specialty voice commands
Specialty patient education
Specialty validation rules

Security Requirements

ScribeVault shall provide security controls appropriate for systems handling protected health information.

The system shall support:

Encryption in transit
Encryption at rest
Secure authentication
Authorization
Role-based access control
Least-privilege access
Audit logging
Session controls
Credential protection
Secure configuration
Data retention controls
Data deletion controls
Backup protection
Security event monitoring
Administrative controls
Configurable security policies

Sensitive data shall not be transmitted to external services unless explicitly enabled and configured by the deployment administrator.

Data Ownership and Storage

ScribeVault shall be designed so that the organization deploying the system controls its clinical data.

The system shall support:

User-selectable storage
Local storage
Organization-controlled storage
Encrypted databases
Encrypted file storage
External storage integrations
Data export
Data migration
Data deletion
Backup and restoration
Configurable retention policies

The core system shall not require a proprietary hosted storage service.

Human Oversight

ScribeVault shall maintain a human-in-the-loop workflow for clinical documentation.

AI systems may assist with:

Transcription
Summarization
Clinical information extraction
Documentation generation
Coding suggestions
Medication reconciliation
Allergy identification
Patient education generation
Documentation validation

AI systems shall not independently:

Finalize a clinical record
Approve medical codes
Submit insurance claims
Diagnose a patient
Prescribe medication
Change a treatment plan
Override clinician decisions

Clinical users shall retain authority over final documentation and clinical decisions.

Workflow

Real-Time Encounter

A typical real-time workflow shall support:

Select or create patient
Create encounter
Select specialty and visit type
Confirm applicable consent
Start voice scribing
Capture and process conversation
Generate structured documentation
Review transcript
Review generated documentation
Review medications and allergies
Review coding suggestions
Correct or approve documentation
Finalize encounter
Export approved information

Post-Visit Documentation

A typical post-visit workflow shall support:

Select or create encounter
Import authorized recording
Generate transcript
Analyze conversation
Generate structured documentation
Review source information
Review documentation
Review medications and allergies
Review coding suggestions
Correct or approve documentation
Finalize encounter
Export approved information

Export Formats

ScribeVault shall support export of approved information in interoperable and human-readable formats.

Supported formats shall include:

CCD
C-CDA
PDF
TXT
JSON
XML
Optional FHIR resources

Export functionality shall preserve the distinction between AI-generated suggestions and clinician-approved information where applicable.

Specification Branding License (SBL)

Fully AGPL-3.0+ compliant system
Copyleft enforced for network deployments
Required attribution:
- Roxanne Ardary
- https://www.roxanneardary.com/

Optional

Specification Branding License (SBL)
- Attribution-free commercial deployment
- Pricing based on scale, usage, and deployment scope
- https://roxanneardary.com/scribevault/

License & Notice Requirements

ScribeVault is released under the GNU Affero General Public License v3.0 or later (AGPL-3.0+).

By contributing to this project, you agree that your contributions will also be released under this license.

Please note the following:

All contributions must comply with the AGPL-3.0+ terms.
Under Section 7 of the license, all redistributions, forks, and derivative works must preserve attribution to Roxanne Ardary and roxanneardary.com.
ScribeVault specifications are free to use with attribution. A Specification Branding License can be negotiated upon request.
The project’s notice.md file tracks attribution requirements and contributor acknowledgments. Any update that adds new contributors or modifies attribution should also update notice.md.
When submitting a pull request, ensure that any new files maintain the attribution headers where applicable.
Network-deployed versions of this software must also remain fully AGPL-3.0+ compliant, including exposure of source code modifications when applicable under the license.

For full legal details, please refer to the AGPL-3.0+ license and the project’s notice.md file.

Notice – ScribeVault

Attribution Requirement: Under Section 7 of the AGPL 3.0+ license, all redistributions, forks, and derivative works, including network-deployed versions of this project, must provide attribution to Roxanne Ardary and roxanneardary.com.

Contributors

This file tracks contributors and their specific contributions to the project.

Roxanne Ardary, roxanneardary.com – March 26, 2026
Created the repository for ScribeVault. Developed the open-source AI-driven clinical documentation platform that functions as a real-time medical scribe and post-visit summarizer, including HIPAA-compliant encrypted storage, specialty templates, voice interaction, coding assistance, and C-CDA/FHIR export capabilities.
[Add other contributors here] – [Date]
[Describe contribution in one sentence]

License – ScribeVault

This repository is licensed under the GNU Affero General Public License v3.0 or later (AGPL-3.0+).

Key Points:

You are free to use, modify, and distribute the code.
All redistributions, forks, and derivative works or network-deployed versions must also be licensed under AGPL-3.0+ and provide attribution to Roxanne Ardary and roxanneardary.com as required under Section 7 of the license.
The software is provided “as is,” without warranty of any kind.

For the full license text, see GNU AGPL-3.0 License.