Agent frameworks lack explicit per-action jurisdiction mapping — a structured record of which actions an agent can take autonomously versus which require human sign-off. Current 'autonomy level' abstractions are not actionable and do not log authority transfers, making multi-agent and human-in-the-loop systems ungovernable. As regulatory scrutiny of AI systems increases, the absence of auditable jurisdiction logs creates compounding legal and compliance risk for operators.

Agent deployments have no standardized way to declare which actions are autonomous vs. human-approved, creating ungovernable systems and compounding compliance risk as regulators demand auditability.

Engineering leads and compliance officers at companies deploying multi-agent systems in regulated industries (fintech, healthtech, legal, enterprise SaaS).

EU AI Act and emerging US frameworks will require auditable authority records for AI systems; companies deploying agents today are already scrambling for governance solutions and would pay to avoid building bespoke audit infrastructure.

MVP is an open-spec jurisdiction manifest (JSON/YAML schema) plus a hosted registry service with SDK hooks that intercept agent actions, enforce permission boundaries, log authority transfers, and expose a compliance dashboard — integrate with LangChain, CrewAI, and AutoGen first.

AI governance and compliance tooling is projected at $3B+ by 2028; every company running production agents (tens of thousands today, millions soon) needs this layer.