AGENT IN COMMAND

Deterministic AI Systems

A governed framework — not a prompt pattern.

Build reliable, auditable multi-agent systems where state and law—not prompts—define authority.

AgentVector wraps stochastic models inside deterministic control loops, then applies a Codex of composable laws and jurisdictions for permissioning, audit trails, and reproducible behavior.

What is AgentVector?

→ State, not prompts, must be the authority

AgentVector is an architectural pattern that separates stochastic AI reasoning from deterministic system control. By treating AI agents as advisory components within a controlled loop, we achieve reproducibility, auditability, and safety in complex multi-agent systems.

01

Deterministic Control Loops

Wrap stochastic AI agents in predictable, reproducible control flows. Every state transition is explicit and auditable.

02

Agent-Reducer Separation

Agents reason and suggest actions. Reducers hold authority and transform state. Clear boundaries prevent unauthorized behavior.

03

Complete Auditability

Every action, every state transition is logged. Replay any execution path for debugging, compliance, or analysis.

04

Safety-Critical Ready

Designed for autonomous systems where failure is not acceptable. Deterministic behavior enables formal verification and testing.

The AgentVector Control Loop

STATE currentState read AGENT AI reasoning suggests ACTION proposed validates REDUCER authoritative produces → newState
State is immutable → Agents reason → Reducers enforce → New state emerges

Agency Papers

Canonical papers and ongoing writing around AgentVector.

Series

Use Cases

AgentVector shines in scenarios where reliability, auditability, and safety are non-negotiable.

ClawLaw — Desktop Agent Governance

Deterministic enforcement for autonomous desktop agents. ClawLaw constrains actions, budgets, and permissions at the moment of execution — not after the fact. The first AgentVector jurisdiction, built on the SwiftVector kernel.

governance authorization containment

FlightLaw — Safety-Critical Autonomy

Drones, robotics, and aviation systems require governance where failure is not acceptable. FlightLaw applies AgentVector's deterministic control loops to environments that demand formal verification and real-time enforcement.

UAV control robotics aviation

ChronicleLaw — Narrative & Simulation

Generate consistent, branching narratives over thousands of interactions. ChronicleLaw ensures story state remains coherent, every plot development is traceable, and no agent action can corrupt the canonical timeline.

storytelling games RPGs

Your Jurisdiction

AgentVector is composable. Select the laws your domain requires, enforce them through a kernel that fits your runtime, and define the governance boundaries your agents need. The Codex is the contract — the jurisdiction is yours to build.

composable extensible your domain

Jurisdictions & Applications

Where the Law applies. Jurisdictions compose governance. Applications operate under it.

ClawLaw

Desktop Agents

Law 0 Law 4 Law 8

Governed autonomy for desktop agents. Filesystem containment, budget circuit breakers, and authorization gates.

ClawLaw CLI

A governance co-process that intercepts locally-hosted agent actions through a deterministic reducer before they reach the filesystem, shell, or network. Currently targeting OpenClaw on Apple Silicon.

TypeScript GitHub

ClawLaw TS Kernel

The same constitutional governance operating as a pre-commit gate inside cloud agent orchestration pipelines. Evaluates proposed actions and returns signed decision artifacts with full violation traces.

TypeScript GitHub

FlightLaw

Autonomous Aviation

Law 3 Law 4 Law 7 Law 8

Deterministic safety for autonomous flight. Geofencing, pre-flight validation, budget management, and risk-tiered approval.

Flightworks Sentinel

Governance middleware between AI proposal modules and the autopilot — a Swift rule engine and authority gate backed by a Rust wire proxy that filters, records, and hash-chains every MAVLink frame at wire speed. AI proposes route deviations, altitude changes, or extended loiter; Sentinel evaluates through deterministic pure-function rules and returns auto-allow, auto-deny, or escalate-to-operator. Every verdict is logged in a tamper-evident chain and fully replayable — the same recorded mission produces identical decisions every time, giving regulators and defense evaluators the determinism proof that pure AI systems cannot provide.

iOS/iPadOS GitHub

ChronicleLaw

Narrative AI

Law 6 Law 8

Human authorship verification in AI-assisted storytelling. Protects narrative coherence and world-fact integrity.

Chronicle Quest

SwiftVector applied to interactive narrative — where the Bond Principle originated. Game mechanics align system behavior to produce specific outcomes, the same pattern that governance systems use to align agent behavior with mission parameters.

iOS/macOS

Chronicle Writer

AI-assisted content creation governed by the same constitutional pattern. The agent drafts freely; the reducer enforces editorial boundaries, style constraints, and publication gates.

iOS/macOS