Welcome to Isabella AI

Relevant source files

The following files were used as context for generating this wiki page:

• README.md
• ROADMAP.md
• AGENTS.md
• src/data/canonical-docs.ts
• src/data/isabellaBlueprint.ts
• src/lib/isabella-inference-engine.ts
• server.ts
• src/components/Dashboard/SynapticFlowDiagram.tsx

Isabella Villaseñor AI represents a sovereign, territorial, and federated cognitive infrastructure anchored in Nodo Cero, Real del Monte, Hidalgo, México. It functions not as a monolithic chatbot but as a socio-technical system composed of inference models, delimited memory, authorized tools, and executable policies governed by human control. Sources: src/data/canonical-docs.ts:10-24, README.md:120-125

The project achieves cognitive sovereignty by decoupling the identity and normative memory from specific external LLM providers, allowing the system to operate in degraded or air-gapped modes. It integrates five core cognitive modules—ISA, SOPHIA, ORION, ARGUS, and CROWN—to harmonize empathy, dialectic logic, technical execution, and zero-trust security. Sources: src/data/canonical-docs.ts:32-42, src/data/isabellaBlueprint.ts:25-31

Cognitive Architecture

Isabella utilizes a dodecahedral orchestrator (MD-X6) to distribute workloads across 12 "heads" and 24 logical "cores." Each head consists of an Alpha (operative) core and a Beta (critical/auditor) core to ensure decision auditability and safety. Sources: src/data/canonical-docs.ts:88-97, src/data/canonical-docs.ts:105-120

Core Subsystems

The system organizes its functions into five primary architectural pillars:

Subsystem	Name	Primary Role
ISA	Integrated Semantic Awareness	Empathy, emotional resonance, and feminine presence.
SOPHIA	Strategic Operational & Phenomenological Heuristic Intelligence	Dialectic rigor, epistemic logic, and philosophical analysis.
ORION	Operational Real-time Inference & Output Navigator	Intent classification, technical problem solving, and tool execution.
ARGUS	Adaptive Real-time Guardian & Unified Sentinel	Zero Trust security, policy verification, and ethical guarding.
CROWN	Central Routing & Orchestration Waveform Node	Computational governance and dynamic state synthesis.

Sources: src/data/isabellaBlueprint.ts:25-31, src/lib/isabella-inference-engine.ts:38-60

The Canonical Cognitive Cycle

The system processes every perception through a governed 6-phase pipeline to ensure traceability and safety.

Sovereignty and Territorial Context

Isabella is epistemologically anchored in its territory, ensuring that mine-indigenous history and Latin American contexts are not erased by technical optimization. Cognitive sovereignty allows the system to change models without losing its normative memory and to record who authorized every critical action. Sources: src/data/canonical-docs.ts:32-42, src/data/canonical-docs.ts:56-62

Epistemic Traceability

Information processed by Isabella is classified by evidence levels:

• Green: Direct, verifiable evidence.
• Yellow: Derived inference with declared premises.
• Naranja (Orange): Provisional hypothesis or speculation.
• Rojo (Red): Contradiction or lack of critical evidence (blocked).
• Azul (Blue): Normative project principles.

Sources: src/data/canonical-docs.ts:46-52

Security and Governance

Security is an intrinsic component of the cognitive cycle rather than an external patch. The ARGUS Sentinel evaluates every request against Zero-Trust policies before allowing tool execution. Sources: src/components/Dashboard/SynapticFlowDiagram.tsx:90-105, src/data/isabellaBlueprint.ts:22-24

CRYSTALS-LATAMV Cryptographic Profile

The system implements a post-quantum cryptographic profile (TAMV-RFC-0007) to protect identities and memory.

Component	Specification	Purpose
LATAMV-KEM-1	ML-KEM-768 + X25519	Key establishment and encapsulation.
LATAMV-SIG-1	ML-DSA-87	Primary digital signature.
LITLE 32 Gates	Logic Verification Framework	Consistency and provenance validation.
Backup	HQC / BIKE / Classic McEliece	Long-term code-based resilience.

Sources: src/data/canonical-docs.ts:167-185, server.ts:740-755

Technical Implementation

Isabella is built on a modern stack emphasizing performance and type safety. The backend uses Express (exported from server.ts) and is deployed via Vercel serverless handlers. Sources: AGENTS.md:12-25, README.md:130-140

In-Process Inference Engine

The sovereign inference engine (isabella-inference-engine.ts) operates with zero external dependencies, utilizing weighted intent scoring and NFD tokenization. Sources: src/lib/isabella-inference-engine.ts:9-15

export function inferSovereign(input: string, options?: InferenceOptions): InferenceResult {
  const lang = detectLanguage(input);
  const entities = extractEntities(input);
  const { intent, confidence } = detectIntent(input);
  const primaryModule = intent.module;
  
  // Logic synthesized via C.R.O.W.N.
  const reply = selectResponse(intent.responseTemplates, lang, input, entities);
  
  return {
    reply,
    routingDecisions: {
      primaryModule,
      moduleWeights: buildModuleWeights(primaryModule, intent.intent),
      routingRationale: `CROWN v2 routed to ${primaryModule}...`
    }
  };
}

Sources: src/lib/isabella-inference-engine.ts:850-875

Conclusion

Welcome to Isabella AI signifies the entry into a system where AI is subordinate to territory, history, and human governance. By prioritizing architectural sovereignty over model dependence, the project provides a resilient cognitive infrastructure suitable for the Latin American context and the Global South. Sources: src/data/canonical-docs.ts:72-83, README.md:280-290