What does this page preserve?

This is the PRESENT tense of Authority Center, whose owned verb is PRESERVE. It answers 'can this actor take this action right now?' by reading the preserved authority state — the durable record of who holds which authority, under what limits, delegation, and escalation — and returning YES, NO, or CONDITIONAL with a one-sentence reason. The actors and scenarios shown are labeled sample data run in your browser.

How is it different from Verification?

Verification renders an independent verdict on an evidence artifact after the fact, trusting no one. This page answers the live authorization question from the preserved authority state. This tells you can-they; Verification independently confirms a produced decision. Different jobs.

Is the YES/NO answer here binding on a live system?

The interactive answer is computed from a labeled sample authority model in your browser to show the interaction. The same question run against your own preserved authority state returns a defensible answer with a reason. To govern the resulting action, route to Agent-008; to reconstruct who had authority at a past moment, see Authority State.

Can this agent do this?

Q: Does this answer govern the decision?

No. This is an authorization read against preserved authority — it tells you whether the authority exists. Governing the actual decision the agent then makes under that authority is Agent-008. Preserving and answering 'who is allowed' is not governing what they decide.

One question. Three inputs. Yes / No / Conditional in under 10 seconds.

Actor

Action

Try a scenario · click to ask instantly

—

Authority

—

Escalation

—

Path to YES

Decision

—

Authority

—

Evidence

—

Verification

—

—

What is Authority Infrastructure?

You're using it. "Can this agent do this?" is the canonical interaction. Every other H33 surface — onboarding, dashboards, verifiers, regulators' tools — exists to support this one question. If you can ask it of any actor (human, agent, service, organization) for any action (read, write, execute, sign, delegate) and get a defensible answer with a one-sentence reason, you have Authority Infrastructure.

After the fact: see why a decision was allowed · why a decision was denied · who had authority at a moment in time · what changed between two moments.

Owned by Authority Center

This is the authorization question, not the authentication question. Authentication establishes who or what the subject is; authorization answers whether this subject may do this action. Establishing identity does not grant authority: what an agent may do is read from the authority Authority Center preserves and decided by Agent-008. A model's confidence is irrelevant to permission.

This is a supporting surface of Authority Center (owned verb: PRESERVE) — its present tense. It reads the preserved authority state to answer whether an actor could act right now. It does not govern the decision the agent then makes (that is Agent-008), does not render an independent verdict (that is Verification), does not record operations (that is HATS), and does not produce the portable evidence primitive (that is H33-74). Asking “can they” is preserving/reading authority — not governing, verifying, recording, or producing.

Use this when you need to answer, right now, whether a specific actor or agent may take a specific action — an authorization read against the preserved authority state that returns YES, NO, or CONDITIONAL with a reason, before the action is attempted.

Which surface do I want?

Ask can-this-agent-do-this to read who is allowed now (this page). To govern the decision made under that authority, use Agent-008. For the independent verdict on a produced decision, use Verification. Same question across other tenses: past · transition · future / what-if.

Questions people ask

Does this answer govern the decision?

No. It reads whether the authority exists. Governing the decision the agent makes under that authority is Agent-008.

How is it different from Agent-008 and Verification?

This reads preserved authority (can-they, now). Agent-008 governs the decision. Verification renders the independent verdict on the produced evidence. Three distinct verbs.

Is the YES/NO on this page real?

The interactive verdict is computed from a labeled sample authority model in your browser to show the interaction. Run against your own preserved authority state, the same question returns a defensible answer with a reason.