HANRIA blogEssays and explainers on accountable AI agents.https://hanria.ai/blogs/2026-11-14T00:00:00ZHANRIA Agents can act. They cannot yet prove.https://hanria.ai/blogs/principals/2026-11-14T00:00:00ZWhy the missing piece in the AI economy is accountability, and how HANRIA is designed to build it. A response to the essay Principals. What is authority drift when every individual agent check still passes?https://hanria.ai/blogs/authority-drift/2026-09-04T00:00:00ZWhat is authority drift when every individual agent check still passes? What a hash-chained local decision log is, and is nothttps://hanria.ai/blogs/decision-log-is-not-evidence/2026-09-04T00:00:00ZWhat a hash-chained local decision log is, and is not. What does fail-closed mean for an agent mandate?https://hanria.ai/blogs/fail-closed-mandates/2026-10-10T00:00:00ZWhat does fail-closed mean for an agent mandate? Governed Habitats: What the Agent-Habitat Literature Asks of a Harness, and Where HANRIA Standshttps://hanria.ai/blogs/governed-habitats/2026-10-08T00:00:00ZAn assessment of the HANRIA harness against the agent-habitat literature: what aligns, what does not yet, and what external validation would take. How to write an agent mandate: denials first, expiry, requires_humanhttps://hanria.ai/blogs/how-to-write-a-mandate/2026-09-04T00:00:00ZHow to write an agent mandate: denials first, expiry, requires_human. Worked mandate templates: ticket read, file prefix, refund ceilinghttps://hanria.ai/blogs/mandate-templates/2026-09-04T00:00:00ZWorked mandate templates: ticket read, file prefix, refund ceiling. What should an agent do on permit, deny, escalate, or error?https://hanria.ai/blogs/permit-deny-escalate/2026-09-04T00:00:00ZWhat should an agent do on permit, deny, escalate, or error? When must an agent escalate instead of acting?https://hanria.ai/blogs/requires-human/2026-10-12T00:00:00ZWhen must an agent escalate instead of acting? A skill can check an agent action. Only a runtime can stop one.https://hanria.ai/blogs/skill-checks-runtime-enforces/2026-09-04T00:00:00ZA skill can check a proposed agent action against a written mandate and record permit, deny, or escalate. Stopping the action requires exclusive control of its credential or tool. What is the difference between an agent skill and an enforcement runtime?https://hanria.ai/blogs/skill-vs-runtime-in-one-page/2026-10-09T00:00:00ZWhat is the difference between an agent skill and an enforcement runtime? What is an agent mandate, and why a permission system will not catch authority drift?https://hanria.ai/blogs/what-is-an-agent-mandate/2026-09-04T00:00:00ZAn agent mandate is a written grant of which operations, targets, counterparties, and amounts an agent may attempt, when it expires, and which kinds always need a person. A permission system will not catch authority drift.