Series

Directing AI agents as a practiced discipline. Scoping work, enforcing invariants, and building the judgment to know when to intervene.

• The Agency Paradox — published (anchors the series)
• The Stochastic Gap — why probabilistic systems require deterministic boundaries
• Propose, Don't Decide — the reducer pattern as a model for human-AI authority
• The Automation Paradox — why more capable agents need more governance, not less
• Trust as Architecture — why trust is a system property, not a prompt attribute

The governance layer beneath every governed agent. Budgets, sandboxes, jurisdictions, and the Laws that agents cannot override.

• Law at the Edge — new foundation paper (anchors the series)
• SwiftVector: The Reference Kernel — why the reference implementation is built in Swift
• The Conformance Contract — how JSON fixtures prove enforcement equivalence across kernels
• Budgets, Not Permissions — why Law 4's resource model changes how agents degrade
• Jurisdiction Design — how to compose Laws for a new domain
• Audit as Architecture — event-sourced governance and the replay guarantee