Open Source Packages

Open governance standard for AI agent trust — the specification powering the Vorion ecosystem.

Canonical constants, tier definitions, and configuration shared across the Vorion ecosystem.

Shared Zod schemas and validators for the Vorion platform. Type definitions ensuring consistency across all packages.

Governance runtime enforcement API. Real-time policy evaluation, intent processing, and enforcement for AI agent operations.

Core runtime for the Agentic Trust Scoring Framework. Implements the BASIS standard with behavioral trust calculation, decay, and recovery.

Immutable audit trail for AI agent operations. Decision provenance tracking with tamper-evident logging.

Agent Anchor AI Trust Engine. Authorization hooks, execution tracking, and orchestration for governed AI operations.

Multi-provider AI routing with intelligent model selection, PII detection, cost optimization, and carbon-aware sustainable routing.

16-agent governance orchestration system for AI oversight, compliance checking, and quality assurance with human-in-the-loop escalation.

TypeScript SDK for connecting AI agents to Vorion Mission Control with WebSocket connectivity, auto-reconnection, and approval workflows.

Core Vorion SDK for integrating AI governance into applications.

Client library for the Categorical Agentic Registry (CAR) — universal agent identification and lookup.

CLI tool for managing CAR (Categorical Agentic Registry) agent identities from the command line.

Security primitives and utilities for the Vorion governance platform.

Vorion platform runtime — core execution environment for governed AI agent operations.

Design tokens for the Vorion ecosystem — colors, spacing, typography, and component primitives.