Metronisys: Agentic AI Orchestration
Metronisys is an enterprise-grade agentic orchestration platform designed for a "Local-First" world. It routes complex user requests through a rigorous policy engine, plans multi-step workflows, executes specialized skills under strict governance, and maintains tamper-evident audit trails. Built for consultants and solo-operators in highly regulated sectors, it aligns naturally with ISO 42001, ISO 27001, and NIST AI RMF standards.
Architectural Design Approach
Policy-Gated Authority
No agent calls tools directly. The PolicyEngine.authorize(ToolCall) acts as the single choke point. Constitutional guardrails (the "Soul") are immutable and cannot be bypassed by sub-agents.
State Management (Blackboard)
A centralized Blackboard manages shared state, metadata, and working memory. Every output is treated as a typed artifact with a full lineage kernel to prevent data contamination.
Core AI Patterns & Workflows
The Supervisor Pattern
Orchestrates the plan into a Directed Acyclic Graph (DAG). Nodes run when inputs are ready, allowing for parallel processing and durable checkpoints.
Self-Reflection & Correction
If a DAG node fails, the ReflectionAgent analyzes the failure, generates a correction artifact, and triggers an automated retry within the governance loop.
Intent-Driven Routing
Requests are classified (HR, Legal, Finance) at the entry point to apply specific policy bundles and restrict available toolsets.
Governance & Auditability
Every run generates an Append-Only JSONL Audit Manifest. To ensure integrity, events are cryptographically chained (Event Hash → Chain Hash).
| Framework | Implementation |
|---|---|
| ISO 42001 | Automated AI Artifact Trails & Risk Context policy evaluation. |
| NIST AI RMF | PII pseudonymization and Human-in-the-loop (HITL) checkpoints. |
| ISO 27001 | Encrypted-at-rest workflow checkpoints and Scoped IAM for agents. |