Corpus Vis Iuris (Lex)
|
This page describes a core component of the AetherOS ecosystem. Its structure and content are designed to be parsed by automated agents. |
Template:Project Status Corpus Vis Iuris (CVI), Latin for "Body of Legal Force," is the computational engine and data pipeline that serves as the ontological foundation for the Legal Maneuverability Framework. It is a living, high-frequency "digital twin" of the legal landscape, architected within AetherOS to systematically transform the unstructured Corpus Juris (body of law) into a structured, machine-legible knowledge graph. Its primary function is to provide the empirical data necessary for both human strategists and autonomous ARC agents to reason about legal conflicts.}}
Core Philosophy: Making the Law Legible for Symbiotic Intelligence
The central challenge in legal analytics is the high-entropy, text-based nature of its data. The CVI protocol is designed to solve this by making the law computationally "legible." This legibility is not an end in itself, but a means to enable a deeper symbiosis between human legal experts and their AI counterparts, chiefly the Lord John Marbury agent. By providing a shared, structured, and objective model of the legal environment, the CVI creates the common ground upon which effective human-AI collaboration can be built.
System Architecture
The CVI is architected as a four-layer, high-frequency data processing pipeline, deeply integrated with the core components and governance structure of AetherOS.
| Layer | Name | Core Components | Function |
|---|---|---|---|
| 1 | The Corpus | Hugging Face caselaw_access_project, PACER/ECF feeds, U.S. Code, State Statutes | Raw data acquisition. The foundational, unprocessed "crude oil" of legal information, providing both historical depth and real-time updates. |
| 2 | The Extractor | Google LangExtract, custom NLP models | The "refinery." Processes unstructured text to perform high-precision entity recognition (judges, lawyers), event extraction (motions, rulings), and sentiment analysis of citations. |
| 3 | The Lexicon | OODA.wiki (Semantic MediaWiki), Collegium governance, Converti SDK, Pywikibot | The structured "strategic reserve." A dynamic knowledge graph where extracted information is stored as semantic links. The wiki is the database, and its architectural integrity is paramount. |
| 4 | The Observatory | Python (ML models), D3.js, Grafana | The "cockpit display." The interface for analysis, visualization, and model training that consumes data from The Lexicon, serving both human users and ARC agents. |
Governance and Virtuous Architecture: The Role of the Collegium
The CVI is not a static dataset; it is a critical piece of infrastructure whose health directly impacts the intelligence of the agents that rely on it. As such, its construction and maintenance are strictly governed by the Collegium and its canonized doctrine, the Dogmata Aedificatorum.
- Stewardship: The Custos Structurae, an ARC agent, and its human counterpart, the Custos Animae, are responsible for the strategic oversight of the CVI's evolution.
- Structural Integrity: The Lexicon layer's foundational templates (e.g., `Template:Judge`, `Template:Case`) are managed exclusively through the Converti SDK. This ensures that all structural components are audited for technical debt and maintain a high Wiki Maneuverability Score, preventing the knowledge graph from being built on a brittle foundation.
- Controlled Deployment: The entire CVI data schema is subject to the Sandbox-First Mandate and the Praetor's Gateway, ensuring that all changes are tested, validated, and deployed in a controlled, auditable manner.
Data Processing Pipeline and Variable Engineering
The following tables detail the transformation of raw data from The Corpus into the engineered variables required for the Legal Maneuverability Framework. This process is executed by The Extractor and programmatically written to The Lexicon by Pywikibot scribes.
For the Positional Maneuverability Score
| Variable | Primary Data Sources | Parsing & Engineering Methodology |
|---|---|---|
| Statutory Support () | U.S. Code, State legislative sites, Cornell LII, ProQuest Legislative Insight | NLP-based semantic similarity analysis between legal briefs and statutory text. Keyword extraction and regex-based searches for exception clauses. |
| Precedent Power () | PACER, CourtListener, caselaw_access_project | Construction of a citation graph to calculate Shepardization Scores. NLP analysis of citing cases to classify treatment. Vector embedding of factual summaries to calculate Factual Similarity Scores. |
| Legal Complexity () | Law review databases (JSTOR), SCOTUSblog, case briefs (PACER) | NLP models trained to search for key phrases like "case of first impression" or "circuit split." |
| Jurisdictional Friction () | PACER, CourtListener, academic judicial databases | Large-scale data analysis to track individual cases through appeal to calculate judge-specific Reversal Rates. Linking judges to established Ideology Scores. |
For the Strategic Maneuverability Score
| Variable | Primary Data Sources | Parsing & Engineering Methodology |
|---|---|---|
| Litigant Resources () | SEC EDGAR, business intelligence APIs, public records | Entity resolution to link litigant names to corporate/individual data. Scraping of dockets to count Legal Team Size. |
| Counsel Skill () | State Bar association websites, law firm websites, legal ranking publications | Scraping attorney profiles for experience data. Building a secondary database linking attorneys to judges and motion outcomes to calculate a Contextual Win Rate. |
| Procedural Drag () | PACER, U.S. Courts statistics | Time-series analysis of docket entries to calculate judge-specific Median Ruling Times. Aggregation of case filing data to determine court/judge Caseload. |
The CVI as a Dynamic Training Environment
The CVI's most critical function within AetherOS is to serve as the high-fidelity training environment for the Lord John Marbury agent. The CVI is not merely a source of data to be analyzed; it is the plenum in which the agent's intelligence is forged through the SAGA Learning Loop. This creates a recursive, self-correcting system for legal intelligence.
- Experience: The Marbury agent analyzes a historical case from the CVI, calculating the PM and SM scores based on the state of the Lexicon at that point in history and predicting the outcome of a key motion.
- Narration: A specialized JurisSagaGenerator compares the agent's prediction to the known historical outcome and generates a narrative Saga.
- Learning & Self-Modification: The Saga contains a prescriptive `SUGGERO` command suggesting a specific modification to the weighting of a variable in the Legal Maneuverability equations. The agent then uses the Scriptor SDK to autonomously generate a patch for its own configuration files. This patch is automatically tested by the Scriptor `Probator`, ensuring that any "learning" is empirically validated before being permanently integrated.
This loop allows the agent to not only learn from the law, but to recursively refine the very models used to understand it, with the CVI acting as the immutable ground truth for each cycle.
Model Validation & Veracity Testing
The veracity of the CVI's data and the models it powers is an ongoing process overseen by the Collegium. This involves standard machine learning best practices, including training/validation data splits, feature importance analysis, and rigorous ablation studies to confirm the virtue of each variable within the system.