Continuum Companion Specification

Home / Continuum Companion / Continuum Companion Specification

Continuum Companion

Explore the Human Side of Healing.

Overview

Continuum Companion is a modular, locally hosted, open-source AI wellness and research platform designed to bring together Western medical knowledge and Eastern medicinal traditions in a transparent, educational, and privacy-focused environment.

The platform is designed to help users research health and wellness topics, organize information, explore different medical and traditional perspectives, create personal wellness plans, and maintain a private research archive.

Continuum Companion is intended to support, not replace, licensed medical care. The system must clearly distinguish established medical evidence, emerging research, traditional practices, cultural frameworks, and areas where evidence is limited or conflicting.

The platform is designed around local ownership, user-controlled data, source transparency, modular functionality, and human decision-making.

Core Philosophy

Support informed conversations between users and licensed healthcare professionals.
Provide educational information rather than medical diagnosis or treatment directives.
Present Western medical knowledge and Eastern traditional perspectives as distinct knowledge systems.
Clearly identify the level and type of evidence supporting each claim.
Preserve source provenance and citations whenever possible.
Allow users to investigate information rather than relying on unsupported AI assertions.
Keep personal research and wellness information under the user’s control.
Support fully local and self-hosted operation.
Avoid mandatory cloud services or centralized data collection.
Provide transparent explanations of uncertainty, disagreement, and limitations.
Respect cultural context when presenting traditional medical systems.
Keep consequential health decisions under human and professional oversight.

Modular Architecture

Continuum Companion uses a modular architecture in which core capabilities form the foundation of the platform and optional plugins extend functionality.

Core modules should operate independently where practical while supporting communication through defined interfaces.

The system should support:

Independent module activation and configuration.
Cross-module information sharing with user authorization.
Modular data storage.
Modular AI capabilities.
Optional feature installation.
Local-only operation.
Offline operation where supported.
Plugin discovery and management.
Community-developed extensions.
User-controlled module permissions.
Clear separation between research, wellness planning, traditional knowledge, and medical information.
Module-level configuration and privacy controls.

AI Research Assistant

The AI Research Assistant provides conversational research capabilities across medical, scientific, wellness, and traditional knowledge sources.

Features include:

Natural-language research questions.
Conversational research sessions.
Multi-source research.
Source-aware retrieval.
Retrieval-augmented generation.
Academic paper discovery and summarization.
Medical literature research.
Traditional text research.
Historical source research.
Research synthesis.
Comparative research across knowledge systems.
Citation-backed responses.
Source provenance.
Evidence classification.
Identification of conflicting information.
Identification of research gaps.
Identification of uncertainty.
Follow-up research questions.
Research topic expansion.
Context-aware conversations.
User-controlled research archives.
Local research processing when supported.

The AI should avoid presenting uncertain information as established fact.

Source-Backed Question and Answer System

The Q&A system provides users with answers supported by identifiable sources whenever sources are available.

Features include:

Natural-language questions.
Source-linked answers.
Inline citations.
Citation exploration.
Source summaries.
Source credibility information.
Evidence-level indicators.
Supporting evidence.
Conflicting evidence.
Areas of scientific uncertainty.
Traditional knowledge attribution.
Historical context.
Follow-up questions.
Research deep dives.
Related research topics.
Source comparison.
User feedback on source relevance.
Ability to save useful answers to the research archive.

Responses should distinguish between established medical knowledge, clinical research, emerging research, traditional practices, cultural beliefs, hypotheses, and AI-generated synthesis.

Integrative Perspective Module

The Integrative Perspective Module presents applicable Western medical information alongside Eastern medicinal and traditional perspectives.

Features include:

Western medical perspective.
Eastern medicinal perspective.
Side-by-side comparison.
Evidence classification.
Traditional practice classification.
Historical context.
Cultural context.
Potential areas of agreement.
Areas of disagreement.
Areas requiring additional research.
Safety considerations.
Questions users may wish to discuss with healthcare professionals.
Clear distinction between supportive wellness practices and medical treatment.

The system must not imply that traditional practices have the same evidentiary basis as clinically validated medical treatments when the available evidence differs.

Western Medicine Module

The Western Medicine Module organizes conventional medical knowledge and biomedical research.

Features include:

Conventional medical treatments.
Clinical research.
Clinical trial information.
Medical guidelines.
Peer-reviewed research.
Pharmacological information.
Surgical approaches.
Radiation-based treatments.
Diagnostic information.
Disease research.
Pathophysiology research.
Treatment outcomes.
Side-effect information.
Treatment interactions.
Supportive care information.
Palliative care information.
Rehabilitation and recovery information.
Research uncertainty.
Evidence quality classification.

Information should be presented as educational material and should not be interpreted as individualized medical instructions.

Eastern Medicine Module

The Eastern Medicine Module provides structured access to traditional and complementary knowledge systems.

Supported knowledge areas may include:

Traditional Chinese Medicine.
Ayurveda.
Tibetan medicine.
Japanese Kampo.
Taoist health practices.
Yogic traditions.
Meditation traditions.
Breathwork traditions.
Traditional movement practices.
Traditional dietary frameworks.
Herbal traditions.
Traditional bodywork.
Sound and resonance traditions.
Chakra-based frameworks.
Traditional philosophical approaches to wellness.

Features include:

Traditional practice descriptions.
Historical context.
Cultural attribution.
Traditional terminology.
Source references.
Traditional theory explanations.
Modern research where available.
Evidence classification.
Safety considerations.
Potential interactions with conventional treatment.
Clear separation between traditional claims and scientifically established findings.

Traditional practices must not be represented as cures for cancer or other diseases without appropriate evidence.

Body and Chakra Mapping Module

The Body and Chakra Mapping Module provides an educational framework for exploring relationships between body regions and traditional chakra systems.

Features include:

Interactive body-region mapping.
Chakra overlays.
Traditional chakra associations.
Traditional symbolic interpretations.
Emotional and somatic exploration.
Historical and cultural context.
Research references where applicable.
Comparison of traditional interpretations with Western anatomical knowledge.
User notes.
Research links.
Educational disclaimers.

The mapping system must clearly communicate that chakra associations are traditional or symbolic frameworks and are not established anatomical causes of disease.

The system must not infer that a cancer location, symptom, or disease is caused by an imbalance in a chakra.

Research Archive Module

The Research Archive Module allows users to retain and organize information discovered through Continuum Companion.

Features include:

Saved research papers.
Saved documents.
Saved sources.
Saved AI responses.
Saved research conversations.
Saved notes.
Saved practitioner notes.
Saved wellness plans.
Tags.
Categories.
Collections.
Search.
Semantic search.
Natural-language retrieval.
Source-linked records.
Research timelines.
Version history.
User annotations.
Cross-references.
Local backups.
Data export.
Data portability.

Users should be able to ask the system to retrieve previously saved research using natural language.

Wellness Plan Module

The Wellness Plan Module allows users to create, organize, and maintain personal wellness plans.

Features include:

User-created wellness plans.
AI-assisted plan creation.
Editable recommendations.
Daily schedules.
Weekly schedules.
Meditation practices.
Breathwork.
Gentle movement.
Stretching.
Relaxation practices.
Journaling.
Sleep-support practices.
Stress-management practices.
Recovery-support practices.
Nutrition research organization.
Personal goals.
Milestones.
Progress tracking.
Notes.
Plan versioning.
Plan history.
Research references.
Provider-shareable summaries.

AI-generated wellness plans should clearly identify informational suggestions and should not override medical instructions from qualified healthcare professionals.

Wellness Plan Research Integration

Wellness plans should be connected to the user’s research archive.

Features include:

Attach sources to individual plan items.
View evidence supporting a practice.
View traditional sources supporting a practice.
Identify uncertainty.
Identify potential safety considerations.
Record user observations.
Track changes over time.
Preserve plan versions.
Compare previous and current plans.
Generate summaries for discussion with healthcare professionals.

Knowledge Graph Module

The Knowledge Graph Module connects information across the user’s research.

Features include:

Research topic relationships.
Source relationships.
Citation relationships.
Treatment relationships.
Condition relationships.
Wellness practice relationships.
Traditional knowledge relationships.
Western medical relationships.
Cross-tradition mappings.
Related research discovery.
Topic clusters.
Knowledge exploration.
Research timelines.
Visual relationship exploration.

The knowledge graph should preserve distinctions between established evidence, hypotheses, traditional beliefs, and AI-generated relationships.

Evidence and Source Provenance Module

The Evidence and Source Provenance Module provides transparency into the information used to generate responses.

Features include:

Source identification.
Citation tracking.
Source URLs.
Publication information.
Authors.
Publication dates.
Source type.
Evidence classification.
Research quality indicators.
Source provenance.
Citation relationships.
Supporting source excerpts where legally and technically appropriate.
Conflicting source identification.
Outdated-source identification.
Source update tracking.
User source annotations.

The system should avoid presenting an AI-generated interpretation as if it were a direct statement from a source.

Research Comparison Module

The Research Comparison Module enables structured comparison between different approaches and bodies of knowledge.

Features include:

Western medicine versus traditional perspectives.
Multiple traditional systems comparison.
Treatment comparison.
Research comparison.
Evidence comparison.
Historical comparison.
Cultural context.
Potential benefits identified in sources.
Potential risks identified in sources.
Known limitations.
Areas of agreement.
Areas of disagreement.
Unresolved research questions.

Comparisons should describe documented differences without assigning unsupported superiority to a particular system.

Privacy and Local Hosting Module

Privacy and local ownership are foundational requirements.

Features include:

Fully self-hosted operation.
Local AI inference.
Local data storage.
Local research archives.
Offline operation where supported.
LAN-only deployments.
Optional air-gapped deployments.
User-controlled backups.
Portable deployments.
Encryption at rest.
Encryption in transit.
Local vector storage.
Zero-telemetry mode.
No mandatory account.
No mandatory cloud AI provider.
No sale of user research data.
Explicit permission controls.
Data deletion controls.
Data export controls.

The platform should minimize collection of personal information and should not require personal health information to leave the user’s controlled environment.

Human Oversight and Safety Module

The Human Oversight and Safety Module ensures that the system remains an educational research and wellness tool.

Features include:

Medical disclaimers.
Clear uncertainty indicators.
Human review prompts.
Healthcare-provider discussion prompts.
Safety warnings.
Interaction warnings.
Contraindication warnings where supported by reliable sources.
Separation of research from diagnosis.
Separation of wellness planning from medical prescribing.
Protection against unsupported medical claims.
Protection against fabricated citations.
Source verification.
AI confidence limitations.
Escalation guidance for urgent situations.
User control over generated plans.
Clear identification of AI-generated content.

The system must never present itself as a physician, diagnose a user, prescribe medication, or instruct a user to disregard professional medical advice.

User Profile and Preferences Module

The User Profile and Preferences Module allows users to control how Continuum Companion presents information.

Features include:

Research interests.
Preferred information formats.
Preferred source types.
Traditional knowledge systems of interest.
Saved research topics.
Wellness plan preferences.
Notification preferences.
Privacy settings.
Data-sharing permissions.
AI behavior preferences.
Citation preferences.
Evidence display preferences.

Sensitive information should remain local and user-controlled whenever possible.

Journal and Reflection Module

The Journal and Reflection Module provides a private space for personal observations.

Features include:

Private journaling.
Daily reflections.
Wellness observations.
Research notes.
Mood and experience notes.
Practice observations.
Sleep observations.
Personal milestones.
Searchable entries.
Local storage.
Optional encryption.
Links between journal entries and wellness plans.
Links between journal entries and research.

Journal information should not be treated as clinical evidence or used to make unsupported medical conclusions.

Optional Plugin Modules

Optional plugins extend Continuum Companion without making additional functionality mandatory.

Potential plugins include:

Medical Literature Plugin

Academic database research.
Medical journal discovery.
Research-paper retrieval.
Citation metadata.
Literature monitoring.
Research alerts.

Clinical Trials Plugin

Clinical trial discovery.
Trial eligibility information.
Trial location information.
Trial status.
Trial source links.
Saved trial research.

Clinical trial information should be presented as research information and not as a recommendation to participate.

Traditional Texts Plugin

Traditional medical texts.
Historical documents.
Translation support.
Terminology research.
Cross-cultural interpretation.
Source provenance.

Herbal Research Plugin

Herbal research.
Traditional uses.
Modern research.
Safety information.
Known interactions.
Evidence classification.
Source citations.

The plugin must clearly distinguish traditional use from clinically established efficacy.

Nutrition Research Plugin

Nutrition research.
Dietary research.
Traditional dietary frameworks.
Food research.
Nutrient research.
Evidence comparison.
Source-backed information.

Meditation and Breathwork Plugin

Meditation research.
Breathwork research.
Practice libraries.
Guided practice organization.
Research references.
Personal practice tracking.

Movement and Somatic Practices Plugin

Yoga research.
Tai Chi research.
Qigong research.
Gentle movement practices.
Mobility practices.
Somatic awareness practices.
Research references.
Personal tracking.

Sound and Music Research Plugin

Music research.
Sound research.
Relaxation research.
Traditional sound practices.
Historical traditions.
Research citations.
Personal practice tracking.

Wearable Data Plugin

Optional local import of wearable data.
Sleep data.
Activity data.
Heart-rate data.
Recovery metrics.
Longitudinal visualization.
Local-only processing.

The plugin must not independently diagnose disease from wearable data.

Calendar and Routine Plugin

Wellness schedule integration.
Practice reminders.
Routine management.
Research reminders.
Appointment preparation.
Personal scheduling.

Practitioner Sharing Plugin

Exportable wellness summaries.
Research summaries.
Source lists.
User-selected journal entries.
User-selected wellness plans.
Provider discussion documents.
Explicit user-controlled sharing.

Local Knowledge Base Plugin

Custom document collections.
Personal research libraries.
Local indexing.
Custom terminology.
Private knowledge bases.
Organization-specific research collections.

Translation Plugin

Translation of research materials.
Traditional terminology translation.
Multilingual research.
Language-aware source retrieval.
Preservation of original terminology.

Community Research Plugin

Community research collections.
Public research libraries.
Collaborative annotations.
Research discussions.
Contributor attribution.
Moderation controls.

Community content must be clearly distinguished from peer-reviewed or otherwise authoritative sources.

Optional AI Model Plugin System

The platform should support optional locally hosted AI models.

Features include:

User-selected models.
Multiple model support.
Model-specific configuration.
Local inference.
Model capability information.
Model provenance.
Model version tracking.
Model replacement.
Model evaluation.
Model-specific safety settings.
Model-specific resource requirements.

No single AI model should be required for the platform to remain functional.

Accessibility Module

Accessibility should be treated as a core usability requirement.

Features include:

Keyboard navigation.
Screen-reader compatibility.
Adjustable text size.
High-contrast options.
Accessible forms.
Accessible charts.
Alternative text.
Reduced-motion options.
Clear navigation.
Plain-language explanations.
Expandable technical explanations.
Multilingual support where available.

Internationalization Module

The platform should support research across languages and cultures.

Features include:

Multilingual interface support.
Multilingual research.
Translation assistance.
Traditional terminology preservation.
Original-language source references.
Localized date and measurement formats.
Cultural context.
Language-specific search.

Data Ownership and Portability Module

Users should retain control over their locally stored information.

Features include:

Full data export.
Research export.
Wellness plan export.
Journal export.
Source export.
Portable archives.
Backup and restore.
Migration between installations.
Human-readable exports.
Machine-readable exports.
User-controlled deletion.

Ethical AI Module

The Ethical AI Module establishes standards for responsible use of AI within the platform.

Features include:

Source transparency.
Citation integrity.
Uncertainty disclosure.
Hallucination mitigation.
Cultural sensitivity.
Bias awareness.
Human oversight.
User consent.
Privacy preservation.
Data minimization.
No deceptive medical authority claims.
No fabricated sources.
Clear AI-generated content indicators.

Community and Open Source Module

Continuum Companion should support an open development ecosystem.

Features include:

Public project documentation.
Public roadmap.
Community contributions.
Plugin development.
Translation contributions.
Research-source contributions.
Cultural review.
Practitioner commentary.
Documentation contributions.
Accessibility contributions.
Community discussion.
Contributor recognition.
Transparent project governance.

Research Quality Framework

The platform should classify information according to its source and evidentiary context.

Potential classifications include:

Clinical guideline.
Systematic review.
Meta-analysis.
Randomized controlled trial.
Observational research.
Preclinical research.
Case report.
Expert commentary.
Educational reference.
Historical source.
Traditional medical source.
Cultural or spiritual framework.
Hypothesis.
AI-generated synthesis.

Classification should communicate what a source represents rather than reducing complex evidence to a single unsupported score.

Cancer Research Support

Continuum Companion may provide educational research support for users exploring cancer and cancer-related wellness topics.

Features include:

Cancer research discovery.
Treatment research.
Clinical trial research.
Supportive care research.
Symptom-management research.
Rehabilitation research.
Nutrition research.
Mind-body research.
Traditional medicine research.
Research archive integration.
Treatment information organization.
Questions for healthcare professionals.
Source-backed comparisons.

The system must not claim that traditional medicine, chakra systems, supplements, dietary practices, or other complementary approaches can cure cancer unless supported by appropriate evidence.

Cancer-related responses should clearly distinguish standard medical treatment from supportive wellness practices and traditional perspectives.

Specification Branding License (SBL)

Standard

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/continuum-companion/

License & Notice Requirements

Continuum Companion 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.
Continuum Companion 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 – Continuum Companion

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 – May 13, 2026
Created the repository for Continuum Companion and designed the open-source modular architecture for an integrative AI wellness and research platform combining Eastern and Western medical knowledge systems with locally hosted AI research, archival, and wellness planning capabilities.
[Add other contributors here] – [Date]
[Describe contribution in one sentence]

License – Continuum Companion

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.