This is the handbook’s evaluation tool, its Appendix D: built to be linked, pasted into an RFP, and asked in a vendor call. The adoption guide sets the build order, and the Field Reference carries the definitions behind every question.
Agent auth today can prove who is acting and what credential they hold. It cannot prove the work is still authorized. So when a vendor says they support agent authorization, the evaluation is six questions. Each probes one property of the six-property litmus, and each has a recognizable failing answer.
The five laws are the invariants of delegated authority, and these six questions are the vendor-verifiable surfaces that show those invariants are implemented.
Use this as a hard gate, not a maturity survey. If question 1 does not produce an approved task object, the claim is not mission-based authorization, and the same holds when question 3 is a shared parent token or question 5 is token expiry alone, because those are category questions too. If question 4 is only token validation, the product can still be mission-based at issuance strength, but there is no runtime-enforcement claim.
| # | Ask | What it probes | A failing answer sounds like |
|---|---|---|---|
| 1 | What is the approved task object, and where is it stored? | An approved task object | “The prompt”, “the session”, “the trace ID” |
| 2 | What derives the agent’s authority from that object? | Authority derived from the task | “Admins assign scopes at integration time” |
| 3 | When the agent fans out or delegates, what guarantees the child’s authority is strictly narrower? | Narrow-only delegation | “Sub-agents reuse the parent’s token” |
| 4 | What checks each consequential action against it at the moment of use? | Per-action runtime enforcement | “The token is validated on every call” |
| 5 | What happens, and how fast, when the task is revoked? | Observable lifecycle state | “Tokens expire within an hour” |
| 6 | Can an auditor pull one identifier and see the whole task? | Evidence joins on the task’s identity | “We have comprehensive logs” |
Every failing answer names a credential artifact, a runtime artifact, or a log where the question asked for a governed task object, and that substitution is the category error the test listens for. One failing answer has an acceptable floor form. Token expiry alone fails, but state-gated issuance against a live task record with a published staleness bound is the category’s own coarse end, claimed as revocation bounded by the outstanding token lifetime. The differentiator is the state gate and the record, never the lifetime.
The test is conjunctive for the action-time defense claim, and it splits four and two. Questions 1, 2, 3, and 5 are the category bar, and a product that fails one is not mission-based. Questions 4 and 6 are what the defense claim adds. A product that passes the four category questions without them can claim approved-record integrity with revocation bounded by the outstanding token lifetime, governance rather than defense. The six questions are the handbook’s own bar for the category. The drafts' substrate contract is looser. Its kernel requires a Mission reference, controller and actor binding, approved context, an approval event, an active/non-active gate with bounded reliance, context propagation, and an ordered governance record, and it treats structured authority and monotonic derivation as optional capabilities. A binding can meet the substrate contract and still fail questions 2 to 4. Either way the missing property tells you what you are looking at: no task object is a policy engine, no runtime enforcement is a governance dashboard, no revocation reach is a token issuer that labels its tokens, and no narrowing is a shared service account.
A passing answer has a different sound:
| Property | A passing answer says |
|---|---|
| Approved task object | There is a durable Mission record with an identifier, issuer, approved purpose, actor binding, authority, constraints, lifecycle state, and evidence links |
| Derived authority | The agent’s usable authority is computed from the Mission, not only from an integration-time role, scope, or admin setting |
| Narrowing | Delegated and derived authority is computed as a strict subset of the parent’s: a child task, sub-agent, or downstream token only narrows, and widening requires a fresh approval |
| Runtime enforcement | A PEP checks each consequential action against action, parameters, actor, and current Mission state before the effect happens |
| Revocation | Revoking or expiring the Mission reaches enforcement within a named freshness bound and fails closed when freshness cannot be established, and the answer does not claim that revocation stops work admitted before it or undoes completed effects |
| Evidence | An auditor can start with one Mission identifier and reconstruct approvals, derived authority, decisions, denials, lifecycle changes, and the consequential actions taken under it, joined by the Mission identifier (id and issuer) every Mission-bound token, decision, and evidence record carries, or that the PDP joins to an ordinary token |
And ask for the demonstration, because it is the cleanest proof that the Mission is enforced rather than decorative:
Show me one denied action where the token was valid but the Mission’s state, bounds, parameters, or delegation chain made the action impermissible.
A system that cannot produce a valid-token denial is validating credentials, not enforcing a task.
These are the questions any finance team could answer about a corporate card program without preparation: the approved purpose, who derived the limits, what authorizes each swipe, what the freeze reaches, and what the statement joins. The corporate-card test is this same instrument in card language, and if the answers would be unacceptable for a card program, they are unacceptable for an agent that moves faster and can be talked into things by the documents it reads.
A vendor that passes all six should be able to write the honest deployment claim: the assurance claims, the enforcement scope, the freshness bound, the evidence, and the exclusions, in writing. Two follow-ups test the pass. A Mission-bound token is as strong as the Resource Server that enforces its bounds, and per-action, parameter-level, and state-aware checks come only from runtime enforcement, so question 4 is the one a claim most often fails in practice. And the what-not-to-claim list names the six overclaims to listen for on the way out.
For an RFP or architecture review, use this clause. Describe the approved task object. Identify where it is stored. Show how authority, tokens, decisions, enforcement, revocation, and evidence join to it. Name the enforcement scope and freshness bound. And list every path where the claim does not apply.
For the depth behind each question: the litmus test expands the six properties and names near misses, the implementation checklist is how you verify a claimed pass, and the competitive landscape covers the alternatives a failing answer is usually reaching for, row by row, with the law each one breaks.