The world’s legal systems, connected.
CodexAtlas is an open-source hybrid AI platform for building a global legal knowledge graph and analyzing relationships between laws, cases, courts, legal arguments, doctrines, and institutions. It combines structured legal knowledge with internal predictive reasoning while presenting users with transparent, public-facing summaries.
CodexAtlas is designed to be modular, jurisdiction-agnostic, extensible, and adaptable to different legal systems around the world. Core legal intelligence capabilities form the foundation of the platform, while optional plugin modules allow deployments to add specialized datasets, models, jurisdictions, analytical capabilities, interfaces, and integrations without changing the core system.
CodexAtlas transforms legal information into an interconnected knowledge system. It can ingest cases, statutes, regulations, treaties, judicial opinions, legal arguments, and other authorized legal materials and connect them through citations, concepts, doctrines, institutions, temporal relationships, and jurisdictional relationships.
The platform uses a hybrid architecture. Detailed analytical and predictive models operate within the internal reasoning layer, while public-facing interfaces can provide summarized, contextualized, and non-sensitive results. This separation allows CodexAtlas to support advanced legal research without requiring raw predictive outputs to be exposed to general users.
CodexAtlas is intended for legal research, comparative law, education, institutional analysis, policy research, legal technology development, and other authorized uses. It does not provide legal advice and should not be treated as a substitute for qualified legal professionals or official legal sources.
CodexAtlas is organized into a core platform and optional plugin ecosystem.
The core contains the foundational capabilities required for CodexAtlas to operate as a legal knowledge intelligence system.
Core modules are designed to remain:
- Jurisdiction-agnostic
- Modular
- Replaceable where practical
- Database-independent
- Model-independent
- API-accessible
- Extensible through plugins
- Suitable for local or network deployment
Plugins extend CodexAtlas without requiring specialized functionality to become part of the core platform.
Plugins may provide:
- Country or jurisdiction-specific functionality
- Data-source connectors
- Specialized AI models
- Additional languages
- Visualization systems
- Research tools
- Court-specific analytics
- External integrations
- Specialized legal ontologies
- Alternative storage systems
- Specialized reporting systems
Plugins should interact with CodexAtlas through documented interfaces rather than modifying core functionality unnecessarily.
The Legal Knowledge Graph is the foundation of CodexAtlas.
Features include:
- Legal document representation
- Case and opinion relationships
- Statute relationships
- Regulation relationships
- Citation networks
- Precedent relationships
- Legal concept relationships
- Doctrine relationships
- Court and institutional relationships
- Temporal relationships
- Jurisdiction relationships
- Cross-reference mapping
- Graph versioning
- Graph integrity validation
The graph provides the underlying structure used by other CodexAtlas modules.
The Data Ingestion module provides a standardized framework for bringing authorized legal information into CodexAtlas.
Features include:
- Document ingestion
- Metadata extraction
- Document normalization
- Text extraction
- OCR processing support
- Document classification
- Citation extraction
- Entity extraction
- Language identification
- Jurisdiction identification
- Document version tracking
- Source provenance
- Data validation
- Duplicate detection
Data-source-specific functionality should generally be implemented through optional ingestion plugins.
The Legal NLP module converts legal language into structured information that can be incorporated into the knowledge graph.
Features include:
- Legal terminology extraction
- Entity recognition
- Legal concept extraction
- Citation recognition
- Argument identification
- Topic classification
- Document segmentation
- Semantic similarity
- Relationship extraction
- Legal language normalization
- Multilingual semantic representation
- Context-aware legal text analysis
The module provides model interfaces that allow different NLP systems to be used without locking CodexAtlas to a single AI provider or model.
The Legal Reasoning Engine provides CodexAtlas with its internal analytical capabilities.
Features include:
- Precedent similarity analysis
- Argument relationship analysis
- Legal reasoning extraction
- Predictive relationship inference
- Interpretive pathway analysis
- Legal concept expansion
- Jurisdiction-aware reasoning
- Historical reasoning comparison
- Institutional pattern analysis
- Scenario analysis
Detailed internal analytical results can remain within the reasoning layer while public interfaces expose appropriately summarized results.
The Public Summary Engine converts complex internal analysis into understandable public-facing information.
Features include:
- Case summaries
- Doctrine summaries
- Legal relationship explanations
- Citation explanations
- Trend summaries
- Historical comparisons
- Cross-jurisdiction summaries
- Argument relationship summaries
- Legislative change summaries
- Institutional summaries
- Non-sensitive outcome pathways
Public output should distinguish between documented facts, historical patterns, analytical inferences, and model-generated conclusions.
The Jurisdiction Framework allows CodexAtlas to represent different legal systems without forcing every jurisdiction into the same legal structure.
Features include:
- Jurisdiction definitions
- Court hierarchies
- Legal system classifications
- Jurisdiction-specific terminology
- Court structures
- Legal authority relationships
- Regional legal structures
- Temporal jurisdiction changes
- Jurisdiction-specific metadata
- Common-law and civil-law representation
- Cross-jurisdiction relationship mapping
Jurisdiction-specific implementations should be distributed through optional plugins whenever they are not required by the core architecture.
The Legal Relationship Engine identifies and manages relationships between legal entities.
Supported relationships include:
- Cites
- Cited by
- Supports
- Contradicts
- Distinguishes
- Modifies
- Overrules
- Affirms
- Reverses
- Interprets
- Implements
- Amends
- Repeals
- Influences
- Derived from
- Related to
Relationships may be explicitly documented in source material or inferred by internal models. Inferred relationships should retain provenance and confidence metadata.
The Temporal Analysis module tracks how legal systems change over time.
Features include:
- Legislative evolution
- Case-law evolution
- Doctrine lifecycles
- Argument evolution
- Court-level changes
- Historical interpretation comparison
- Legal relationship changes
- Precedent influence over time
- Graph snapshots
- Historical graph reconstruction
Users can compare legal structures across different time periods.
The Cross-Jurisdiction module enables comparison between legal systems.
Features include:
- Legal concept comparison
- Doctrine comparison
- Statutory comparison
- Case-law comparison
- Institutional comparison
- Legal terminology alignment
- Cross-border influence mapping
- Comparative legal research
- Common-law and civil-law comparison
- International legal relationship analysis
The Legal Research Engine provides tools for discovering and exploring relationships within the graph.
Features include:
- Semantic legal search
- Case discovery
- Statute discovery
- Citation-chain exploration
- Related-case discovery
- Argument similarity
- Doctrine discovery
- Legal concept exploration
- Authority discovery
- Research pathway generation
- Knowledge gap detection
This module evaluates relationships between legal authorities.
Features include:
- Citation analysis
- Authority relationships
- Precedent influence analysis
- Citation quality analysis
- Court hierarchy analysis
- Historical influence tracking
- Doctrinal influence mapping
- Authority relationship visualization
Internal analytical scores should remain distinguishable from objective legal authority.
The Institutional Analytics module analyzes patterns associated with courts and other legal institutions.
Features include:
- Institutional behavior analysis
- Judicial decision pattern analysis
- Panel composition analysis
- Appeals analysis
- Reversal mapping
- Institutional doctrinal trends
- Writing-style analysis
- Procedural patterns
- Institutional change tracking
The system should present institutional analysis carefully and avoid treating statistical patterns as definitive statements about individual judges or institutions.
The Legal Conflict module identifies potential conflicts and inconsistencies within legal systems.
Features include:
- Statutory conflict detection
- Regulatory conflict detection
- Cross-jurisdiction conflict mapping
- Contradictory precedent detection
- Inconsistent interpretations
- Conflicting legal concepts
- Conflict timelines
- Potential resolution pathways
The Multilingual Legal Intelligence module provides language-independent infrastructure for international deployments.
Features include:
- Multilingual embeddings
- Legal terminology alignment
- Cross-language semantic search
- Legal concept alignment
- Translation support
- Cross-language citation analysis
- Multilingual summarization
- Jurisdiction-specific language models
Additional language packs and specialized language models can be distributed as plugins.
The Ethics and Governance module establishes safeguards for CodexAtlas deployments.
Features include:
- Juvenile matter exclusion
- Sensitive-data filtering
- Public-output controls
- Model transparency
- Provenance tracking
- Audit logging
- Data governance controls
- Source attribution
- Model documentation
- Explainability metadata
- Deployment policies
- Jurisdiction-specific compliance configurations
CodexAtlas does not include juvenile matters within its intended dataset or analytical scope.
The Graph Integrity module maintains the reliability and traceability of the knowledge graph.
Features include:
- Source provenance
- Relationship provenance
- Data lineage
- Duplicate detection
- Contradiction detection
- Graph anomaly detection
- Data validation
- Import verification
- Version tracking
- Change history
- Reproducible research records
The Developer Platform provides programmatic access to CodexAtlas.
Features include:
- REST API
- GraphQL API
- Authentication interfaces
- Query interfaces
- Graph access
- Search endpoints
- Analysis endpoints
- Data export
- Plugin interfaces
- Model interfaces
- Administrative APIs
- Developer documentation
The CodexAtlas CLI provides administrative and development tools.
Features include:
- Data ingestion
- Graph construction
- Model management
- Index management
- Plugin management
- Dataset validation
- Graph inspection
- Export operations
- Configuration management
- System diagnostics
The Visualization Engine provides common interfaces for representing legal relationships.
Features include:
- Knowledge graph exploration
- Citation networks
- Precedent maps
- Doctrine maps
- Legal timelines
- Argument maps
- Legislative evolution
- Institutional relationships
- Cross-jurisdiction comparisons
- Legal influence networks
Specialized visualization systems may be implemented as plugins.
The Simulation module supports controlled research scenarios.
Features include:
- Hypothetical legal scenarios
- Relationship propagation
- Doctrinal scenario analysis
- Institutional scenario modeling
- Legal-system comparison
- Historical reconstruction
- Research experimentation
Simulation results should remain clearly identified as analytical or hypothetical rather than established legal conclusions.
CodexAtlas supports optional plugins for capabilities that are useful but should not be required by every deployment.
Jurisdiction plugins can provide:
- Country-specific court structures
- Regional court structures
- Legal terminology
- Court metadata
- Local legal taxonomies
- Jurisdiction-specific ingestion
- Local legal data connectors
- Jurisdiction-specific compliance rules
Data-source plugins can connect CodexAtlas to authorized sources such as:
- Court repositories
- Government legal databases
- Legislative repositories
- Regulatory databases
- Public legal archives
- Academic repositories
- Institutional collections
Each plugin should document its source, licensing requirements, update frequency, and permitted uses.
Model plugins can provide alternative:
- Embedding models
- Large language models
- Legal language models
- Translation models
- OCR models
- Classification models
- Summarization models
- Entity recognition models
- Relationship extraction models
CodexAtlas should remain model-agnostic and avoid requiring a particular AI provider.
An optional treaty plugin can add:
- International treaties
- Conventions
- Agreements
- Treaty relationships
- Domestic implementation relationships
- Treaty interpretation
- Cross-border influence mapping
An optional compliance plugin can provide structured analysis of legal requirements for defined scenarios.
Features may include:
- Requirement mapping
- Applicable-law discovery
- Regulatory relationship mapping
- Compliance pathway analysis
- Jurisdiction comparison
- Historical regulatory changes
Outputs should remain research-oriented and should not be represented as legal advice.
The Legal Education plugin can provide:
- Simplified case explanations
- Doctrine learning tools
- Interactive legal timelines
- Legal concept maps
- Research exercises
- Knowledge graph exploration
- Educational visualizations
A specialized research plugin can analyze aggregate patterns for potential systemic disparities.
Features may include:
- Aggregate outcome comparisons
- Historical disparity analysis
- Institutional comparison
- Geographic comparison
- Temporal comparison
- Data-quality warnings
- Statistical research tools
Results should be presented as analytical findings rather than definitive conclusions about individual actors.
An optional court analytics module can provide:
- Docket analysis
- Delay analysis
- Case duration modeling
- Procedural bottleneck analysis
- Workload analysis
- Court-level trend analysis
- Panel analysis
The Multimodal plugin can extend ingestion to:
- Scanned documents
- Images
- Tables
- Audio transcripts
- Video transcripts
- Document exhibits
- Other authorized legal records
The collaboration plugin can provide:
- Shared research workspaces
- Saved graph views
- Research collections
- Annotation systems
- Collaborative case analysis
- Research notes
- Public research projects
- Dataset documentation
Third-party developers can create specialized visualization plugins for:
- Court dashboards
- Citation maps
- Doctrine networks
- Legal timelines
- Geographic legal maps
- Institutional analytics
- Comparative law interfaces
Plugins should:
- Use documented CodexAtlas interfaces.
- Keep specialized functionality outside the core when practical.
- Document dependencies.
- Document data sources.
- Identify applicable licenses.
- Preserve source provenance.
- Follow CodexAtlas ethical requirements.
- Exclude juvenile matters.
- Avoid exposing restricted internal predictions through public interfaces.
- Include appropriate documentation and tests.
- Avoid unnecessary modification of core modules.
CodexAtlas is designed around a layered architecture:
- Source Layer — authorized legal documents and metadata.
- Ingestion Layer — extraction, normalization, and validation.
- Knowledge Layer — structured legal entities and relationships.
- Reasoning Layer — internal AI analysis and predictive modeling.
- Summary Layer — controlled public-facing explanations.
- API Layer — programmatic access.
- Interface Layer — research and visualization interfaces.
- Plugin Layer — optional extensions and integrations.
This architecture allows individual components to evolve independently while maintaining a stable foundation.
CodexAtlas should avoid unnecessary vendor lock-in.
The architecture is intended to support interchangeable:
- Graph databases
- Relational databases
- Search engines
- Embedding models
- Language models
- OCR systems
- Translation systems
- Visualization frameworks
- Data storage systems
Deployments may select the technologies appropriate for their jurisdiction, infrastructure, security requirements, and research objectives.
CodexAtlas must be deployed according to applicable laws and data-governance requirements.
Deployments should establish:
- Data-source authorization
- Retention policies
- Access controls
- Audit policies
- Redaction policies
- Data provenance
- Model governance
- Public-output policies
- Local compliance requirements
Juvenile matters are outside the intended scope of CodexAtlas and should be excluded from ingestion, processing, indexing, graph construction, and public output.
CodexAtlas separates internal analytical capabilities from public-facing outputs.
The internal layer may contain:
- Detailed model outputs
- Predictive scores
- Relationship confidence
- Scenario simulations
- Advanced institutional analysis
- Model embeddings
- Internal analytical metadata
The public layer should emphasize:
- Documented facts
- Source citations
- Historical patterns
- Relationship explanations
- Contextual summaries
- Transparent analytical findings
- Appropriate uncertainty
This separation is a foundational architectural principle of CodexAtlas.
CodexAtlas development should prioritize:
- Open-source interoperability
- Modular architecture
- Global adaptability
- Legal-data provenance
- Transparency
- Reproducibility
- Human oversight
- Model independence
- Vendor neutrality
- Responsible AI
- Clear separation between evidence and inference
Contributions are welcome from developers, legal researchers, data scientists, academics, translators, documentation contributors, and other members of the open-source community.
Before contributing, review:
CONTRIBUTING.mdnotice.mdLICENSEdocs/Workflow.md
All contributions must comply with the project's licensing, attribution, ethical, and data-governance requirements.
CodexAtlas is a research and technology platform and does not provide legal advice.
AI-generated analysis, predictions, summaries, classifications, and relationships may contain errors and should be independently evaluated against authoritative legal sources.
Users are responsible for determining whether their use of CodexAtlas complies with applicable laws, regulations, licensing requirements, court rules, data restrictions, and professional obligations.
- Fully AGPL-3.0+ compliant system
- Copyleft enforced for network deployments
- Required attribution:
- Roxanne Ardary
- https://www.roxanneardary.com/
- Specification Branding License (SBL)
- Attribution-free commercial deployment
- Pricing based on scale, usage, and deployment scope
- https://roxanneardary.com/codexatlas/
CodexAtlas 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. - CodexAtlas 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 updatenotice.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.