lnt-sovereign 1.0.3

High-performance neuro-symbolic verification layer for AI validation and logic-based guardrails.

LNT: Logic Neutrality Tensor

Deterministic Validation Layer for AI-Generated Intents

⚖️ Overview (Open Source)

LNT is a high-performance neuro-symbolic engine designed to enforce deterministic symbolic constraints on probabilistic AI outputs. It provides sub-millisecond scaling and formal Z3 consistency verification for mission-critical validation in FinTech, HealthTech, and automated infrastructure.

License: MIT. This distribution contains the core engine and public example manifests.

🚀 Key Technical Specifications

• Vectorized Kernel (BELM): SIMD-accelerated logic manifold ($O(n)$ complexity).
• Symbolic CLI: Deterministic verification for MLOps pipelines.
• Tiered Enforcement: Threshold-based logic validation (Advisory vs. Hard Fail).
• Formal Verification: SMT-based (Z3) manifest consistency analysis.

🛠️ Quick Start

Installation

pip install lnt-sovereign

User as a Logic Verification Tool (CLI)

Integrate LNT into your automated pipelines to verify behavioral consistency before deployment. See our CI/CD Workflow Template for a ready-to-use GitHub Action.

# Soft Governance (Advisory Mode)
lnt check --manifest policy.json --input proposal.json --advisory

# Hard Enforcement (Fail if score is low or TOXIC rules violated)
lnt check --manifest policy.json --input proposal.json --fail-under 90 --fail-on-toxic

Use as an SDK

from lnt_sovereign.client import LNTClient

client = LNTClient()
result = client.audit(manifest_id="visa_application", proposal=proposal)
print(f"Validation Certified: Score {result.score}")

📜 Technical Documentation

For detailed architecture, API reference, and MLOps guides, visit the Documentation Portal.

• MLOps Integration Guide: How to set up LNT as a "Logic Gate" in CI/CD.
• Technical Whitepaper: Mathematical foundations and benchmark distributions.
• API Reference: Complete parameter registry and error codes.

---

Maintained for high-reliability AI system development.