Internet-Draft Evaluator Authority and Lifecycle November 2026
Das Expires 20 May 2027 [Page]
Workgroup:
Independent Submission
Internet-Draft:
draft-das-evaluator-governance-00
Published:
Intended Status:
Informational
Expires:
Author:
S. Das
Independent Inventor

Location-Independent Evaluator Governance for Autonomous Multi-Agent Systems: Protected Appointment, Evidence-Bounded Decisions, Cross-Domain Coordination, and Separation of Execution Authority

Abstract

Autonomous agents increasingly rely on an evaluator to review plans, evidence and outcomes before work continues, yet an evaluation result is often mistaken for permission to act. This document describes a vendor-neutral architecture in which the evaluator is a functional role, not a product: it may be a rule engine, model, program, enclave, service or human, placed locally, remotely or on a shared platform. It separates six independent lifecycles, requires a protected task-specific appointment derived from resource-owner authority, admits evidence by source, role, freshness and independence, preserves mandatory dissent, and emits typed determinations such as HOLD, DISPUTED and UNKNOWN. Its central rule is that a valid determination is not an executable authorization: separate owner-controlled authorization and a native enforcement boundary remain necessary. Twenty-seven embodiments, a threat model, data model and examiner index are provided.

Source and Patent Information

Technical source: Indian Provisional Patent Application 202631131650, dated 3 November 2026 (as reported by the applicant; corroborated only by an eFiling payment record showing CBR date 03/11/2026, not by the official acknowledgment, which the editor has not seen), titled "Systems and Methods for Location-Independent Evaluator Architectures, Dynamic and Persistent Evaluation Lifecycles, Protected Appointment, Evidence Verification, and Authority-Separated Decision Control in Autonomous and Distributed Systems". The source states 27 embodiments and 540 numbered claims (60 independent, 480 dependent). Foundational reference: PCT/IB2026/055615, filed 4 June 2026, published 16 July 2026 as WO 2026/150382 A2. See Section 26.

Where to read the full PCT filing: search "Sangam Das" on WIPO PATENTSCOPE; search PCT/IB2026/055615 (pctib202655615); or read the PCT file directly in Technical Disclosure Commons at https://www.tdcommons.org/dpubs_series/12542/.

The IPR disclosure for this document will be made at the appropriate time through the official IETF process and in accordance with IETF guidelines (see Section 26).

AI-assisted drafting was used. Readers must verify patent numbers, dates, claim text, product capabilities and standards references against primary sources. This is an individual proposal, not an IETF standard and not an endorsed position.

Status of This Memo

This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.

Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at https://datatracker.ietf.org/drafts/current/.

Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."

This Internet-Draft will expire on 20 May 2027.

▲

Table of Contents

1. Introduction

Autonomous and multi-agent systems increasingly place an evaluator between a proposed plan and the point where an effect occurs. The evaluator may inspect evidence, estimate consequences, check policy, review an outcome, and recommend whether work should continue. Three questions are frequently merged: who may evaluate, what evidence the evaluator may rely on, and who may actually cause the effect. This document keeps them apart.

The document describes one architecture, drawn from a single source filing with 27 embodiments, in an individual, informational form. It does not define a new wire protocol; it specifies functional requirements, candidate record schemas and test conditions that existing mechanisms (see Section 17) can carry.

Central rule: a valid evaluation determination is not automatically an executable authorization.

1.1. Problem Statement

Existing practice commonly shows one or more of the following gaps: an evaluator is chosen by the agent it reviews; independence is counted as accounts rather than control groups; a signed report is treated as truth about the physical or ledger outcome; an approval is reused after its appointment, round or policy generation has changed; a majority hides a mandatory dissent; a lost acknowledgment is retried and duplicates an effect; and a coordinator's DONE message is treated as closure while effect-capable work remains.

1.2. Scope and Non-Goals

  • In scope: functional evaluator role, protected appointment, lifecycles, evidence admission, typed non-executable determinations, review rounds, multi-agent controls, cross-domain governance, recovery, revocation, membership, minimal-disclosure evidence, and partial-outcome handling.
  • Not in scope: a new transport or wire format; a claim that evaluators are accurate; a claim of global atomicity; instantaneous revocation during disconnection; certification; patentability or infringement conclusions; any company endorsement.
  • Distinct from related drafts: causal semantic diagnosis (see [REL-CAUSAL]) diagnoses where meaning changed; execution finality (see [REL-FOUND], [REL-CBEA], [REL-IEV]) governs whether an effect is admitted. This document governs who may evaluate, on what evidence, for how long, with what typed output, and how that output relates to the separate authority.

2. Terminology and Conventions

The key words in this document follow BCP 14 [RFC2119] [RFC8174] only where requirement language is used, and sparingly. Most statements are architectural descriptions rather than protocol requirements.

Evaluator

A functional role that assesses evidence, a plan, a consequence or an outcome under a bounded mandate and produces a determination. It holds no effectuation authority by virtue of the role.

Supervisor / coordinator

A role that organizes tasks or agents and may carry out limited, authorized coordination. It is not an evaluator and not an owner.

Authorization issuer / owner-controlled authority

The party or process that, under resource-owner policy, issues a bounded permission for a specific act.

Native resource gate / Finality Sink

The enforcement boundary at which an effect is actually admitted or withheld by the resource owner.

Appointment

A protected, task-specific designation of an evaluator derived from the authority of the mandatory resource owners.

Determination

A typed, non-executable output of an evaluation, bound to task, round, policy generation and validity interval.

Control group

The set of credentials, hosts or principals that are in fact under one administrative or economic control.

Review round

A bounded set of reviews over a fixed evidence set that closes with completeness, independence and dissent checks.

Native outcome

Evidence originating at the resource owner or by protected direct observation about what actually occurred.

Protected-local state

State held inside a protected enforcement environment that is not exported as a portable signed artifact.

3. Principal Architecture and Trust Boundaries

The principal architecture is shown in Figure 1. Evaluation authority ends at the boundary where a non-executable determination is produced; continuation is separately authorized and any effect is admitted only by the owner-controlled native boundary.

 Agents / human actors
        |
        v
 Task, plan and evaluation request
        |
        v
 Authority and resource-owner resolution
        |
        v
 Evaluator eligibility + appointment verification
        |
   mandate valid? --NO--> DENY / HOLD
        | YES
        v
 Authorized evidence intake
        |
        v
 Source, role, task, freshness, truth checks
        |
        v
 Policy-bounded evaluation
        |
        v
 Review completeness, independence, dissent
        |  incomplete --> HOLD   contested --> DISPUTED
        |  unresolved --> UNKNOWN / REQUEST_EVIDENCE
        v
 Typed NON-EXECUTABLE determination
 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
 evaluator authority ends above; owner authority begins below
        v
 Separate protected continuation authorization
        |   (owner policy; may refuse)
        v
 Owner-controlled native Finality Sink
        |   admits exactly the authorized act or withholds
        v
 Native outcome evidence (success / partial / UNKNOWN)
Figure 1: Principal architecture (evaluation authority ends at the dotted line)
Branches

A failed mandate yields DENY or HOLD. Incomplete review yields HOLD. Contested review yields DISPUTED. Unresolved or missing evidence yields UNKNOWN or REQUEST_EVIDENCE. None of these is an authorization.

Trust boundaries

Between agents and the evaluator; between the evaluator and the authorization issuer; between the issuer and the native gate. No authority is inherited across a boundary Figure 2.

Where evaluator authority ends

At the typed determination. It is a premise that owner policy may consider, not a token that a gate honors.

 +-----------+   +-------------+   +-------------+   +-----------+
 | Evaluator |   | Supervisor/ |   | Authorization|  | Native    |
 | assesses  |   | coordinator |   | issuer/owner |  | gate/Sink |
 | evidence  |   | organizes   |   | grants bound |  | admits    |
 |           |   | tasks       |   | permission   |  | the effect|
 +-----+-----+   +------+------+   +------+-------+  +-----+-----+
       |  boundary 1    |   boundary 2    |  boundary 3    |
       +----- no authority is inherited across a boundary ------+
 One host may run several roles; co-location inherits nothing.
Figure 2: Four roles and their trust boundaries

4. Evaluator Functional Architecture and Location Independence

An evaluator is defined by what it does, not by what it is. The same function may be an in-process component, a local protected service, a hardware or firmware element, a trusted execution environment, a container or virtual machine, a remote service, a persistent platform, a human, a committee, or a hybrid. Rule engines, policy engines, solvers, statistical models, generative AI and verification programs are all admissible placements. A single-component evaluator is not excluded merely because it is a single component.

 Same function, different placements:
 [in-process component]  [local protected service]  [TEE/hardware]
 [container/VM]          [remote service]           [human/committee]
        \                      |                       /
         +-- protected intake path (versioned evidence) --+
                           |
                    bounded determination
                           |
         protected output path (admissibility checked)
 Co-location is not non-independence; remote is not independent.
 Independence = real control relations, conflicts, access.
Figure 3: Evaluator function and location-neutral deployment

Location independence is not absence of control. Co-location with an agent does not necessarily defeat independence if control relationships, evidence access and conflicts are separated; third-party hosting does not prove independence if the host is under common control or economic dependence. Independence is assessed on actual control relationships, evidence access, conflicts of interest and the ability to influence the outcome, under policy that remains owner-specific.

5. Evaluator Deployment Models

Type I: Ephemeral

Created for one task or round, retired when appointment, round or validity ends. Strong isolation; setup cost per task Figure 4.

Type II: Persistent

A long-lived protected service that serves many independent projects, each with its own mandate, rounds, validity and retention Figure 5.

Type III: Hybrid

A persistent platform hosting task-specific contexts with separate expiry, namespaces and revocation Figure 6.

 eligible? -> appoint -> activate instance -> intake evidence
    |NO                                          |
    v                                            v
  refuse                                  evaluate -> determination
                                                 |
                      handoff <------------------+
                         |
 appointment expires or is revoked --> instance retired
                         |
 evidence retained (protected); old authority NOT reusable
Figure 4: Ephemeral evaluator lifecycle
 Host (long-lived)
   |-- Project A mandate --> rounds A1, A2   (expiry A)
   |-- Project B mandate --> rounds B1       (expiry B)
 Separate data and evidence per project.
 A expires: host and B continue; no inheritance from A.
 Platform compromise: revoke all mandates, revalidate each.
Figure 5: Persistent evaluator lifecycle
 Persistent platform (shared maintenance)
   +- context X (org 1 task) -- own expiry, own namespace
   +- context Y (org 2 task) -- own expiry, own namespace
 per-context revocation | context migration -> revalidate
 shared-platform compromise -> all contexts untrusted until
 re-established
Figure 6: Hybrid evaluator platform

A persistent evaluator does not inherit authority across projects. If the platform is compromised, every context is treated as untrusted until re-established. Moving a context between hosts requires revalidation.

6. Six Independent Lifecycles

Six lifecycles are independent. None is assumed to nest inside another.

L1 Infrastructure

Host or platform exists. State: provisioned, active, suspended, decommissioned.

L2 Runtime instance

A running evaluator process. State: created, running, retired.

L3 Appointment

Designation for a task. State: proposed, owner-approved, active, expired, revoked.

L4 Evaluation round

Bounded review over a fixed evidence set. State: open, admitting, closed, disputed.

L5 Determination validity

Interval during which a determination is usable as a premise. State: valid, stale, withdrawn. It can end before the appointment ends.

L6 Evidence retention

How long evidence and audit material are kept. It may outlive all other lifecycles and grants no permission.

 time ------------------------------------------------------>
 infrastructure  |=================================...........
 runtime instance|    |=========|     (retired early)
 appointment     |      |================|
 eval round      |        |====|
 determination   |           |========|  (may go stale first)
 evidence retain |           |==========================...
 Round may close while task is active; evidence outlives all
 permissions; a valid signature does not extend validity.
Figure 7: Six independent lifetimes (not nested)

Formal predicates (informal notation):

Negative tests: (a) a retired instance presents a valid-looking signature (rejected: L2); (b) evidence arrives after round closure (rejected: L4); (c) a determination is reused after policy generation changes (stale: L5); (d) evidence retention continues while authority is gone (permitted, grants nothing: L6); (e) appointment expires while the host continues (no authority for that task: L3).

7. Control Roles and Separation

Four control roles are distinguished. A single host may run several roles; co-location inherits no authority.

Table 1: Role-separation matrix
Role May do May not do
Evaluator assess evidence; produce typed determination admit an effect; issue a passport; self-appoint
Supervisor / coordinator organize agents and tasks within a bounded lease grant native authority; act as evaluator for its own plan
Authorization issuer / owner authority issue bounded permission per owner policy bypass the native gate; reuse consumed authority
Native gate / Finality Sink admit or withhold the actual act accept a determination or approval as sufficient
 request -> resolve actual mandatory owners
              |                 |
        ambiguous?          resolved
            YES: HOLD            |
                       owner consents (all mandatory)
                          |            |
                    missing/forged   complete
                    -> REJECT        -> appointment issued
                                     -> scope, expiry, use count
 Agent self-declaration of neutrality is never an input.
Figure 8: Owner and appointment resolution

8. Protected Appointment and Eligibility

A protected appointment is task-specific and derived from owner authority. A candidate agent's self-declaration is never an input. The appointment record is expected to include the following elements:

  1. identity of the evaluator and its software or human identity
  2. task or operation scope
  3. resource owners and their consent
  4. mandatory owners and vetoes
  5. qualification and independence assertions with the control-group basis
  6. policy generation
  7. evidence classes the evaluator may access
  8. permitted determination types
  9. review-round limits and reviewer requirements
  10. validity interval for determinations
  11. appointment expiry
  12. use-count limit
  13. revocation reference and serial
  14. host or runtime binding
  15. dissent and quorum rules (owner-specific)
  16. retention and disclosure constraints
  17. jurisdiction and purpose constraints
  18. successor or substitution rules

Failure to verify any mandatory element results in refusal or HOLD, never in permissive default.

9. Evidence Admission, Provenance and Outcome Truth

Ten distinctions apply to every evidence item:

  1. authenticated source origin
  2. authorized source role
  3. content integrity
  4. task and operation correspondence
  5. freshness
  6. applicable policy generation
  7. source independence
  8. correlation among observations
  9. actual native-resource outcome
  10. unresolved or contradictory evidence
 source -> origin authenticated? -> role eligible? -> content
 integrity? -> bound to task/operation? -> fresh? -> policy
 generation? -> independent (control group)? -> correlated?
      each NO -> excluded or flagged
 admitted evidence  !=  outcome truth
 outcome truth needs resource-origin evidence or protected
 direct observation; agent success claims, tool acks, transport
 acks, coordinator flags and evaluator reports are weaker.
Figure 9: Evidence provenance and source-truth separation

10. Typed Non-Executable Determinations and Review Rounds

Determinations are typed and non-executable.

Table 2: Typed determinations
Type Meaning Executable?
APPROVED_FOR_CONSIDERATION premise for owner authorization process No
APPROVED evaluator finds conditions met under mandate No; premise only
DENIED conditions not met n/a
HOLD incomplete mandate, evidence or review No
DISPUTED reviewers or sources contest No
UNKNOWN outcome or evidence unresolved No; not DENIED, not SUCCESS
REQUEST_EVIDENCE named additional evidence needed No
REVISE plan change suggested No; needs recommitment
RENEGOTIATE parties must renegotiate No
HUMAN_REVIEW escalate to human No
SUBSTITUTE replace an agent No; needs fresh authorization
ABORT stop the task Not a native effect by itself

UNKNOWN is not DENIED and not SUCCESS. Dissent is preserved with the determination; it is never silently dropped.

 reviewers r1..rn -> map credentials -> control groups g1..gk
 diversity rule met on k (not n)?  NO -> HOLD
 mandatory reviewer present?       NO -> HOLD
 dissent recorded and shown?       NO -> reject round
 contested outcome -> DISPUTED (not a majority vote)
 round closed -> determination eligible as a premise only
Figure 10: Reviewer independence, control groups, dissent

11. Embodiment Overview and Claim Crosswalk

The following table is a crosswalk of the 27 embodiments to the claims of the source filing. Claim numbers are those of the filing as supplied and should be verified against the official record.

Table 3: E01-E27 crosswalk
Emb. Code Indep. claims Dep. claims Title (source)
E01 CE-NJC-01 1-2 61-76 Protected Multi-Agent Negotiation, Joint-Plan Commitment, and Independent Evaluator-Governed Continuation
E02 CE-BFQ-02 3-5 77-100 Bilateral Communication Freeze, Distributed Quiescence, and Protected Evaluation-State Convergence
E03 CE-NSI-01 6-7 101-116 Negotiation-State Integrity, Causal Message Containment, and Continuous Evaluation-Snapshot Validity
E04 CE-INE-01 8-9 117-132 Independent Multi-Agent Consequence Evaluator, Protected Evidence Admission, and Next-Actor Selection
E05 CE-VCR-01 10 133-140 Isolated Virtual Consequence Reconstruction and Non-Effective Scenario Execution
E06 CE-RBL-01 11-12 141-156 Risk, Benefit-Loss and Tail-Exposure Evaluation With Independent Domain Thresholds
E07 CE-NAS-01 13-14 157-172 Evaluator-Directed Next-Actor Selection and Controlled Next-Turn Instructions
E08 CE-ACR-01 15-16 173-188 Corrective Instructions, Protected Plan Revision, and Controlled Multiparty Renegotiation
E09 CE-ARR-01 17-18 189-204 Agent Role Reconfiguration, Bounded Delegation, and Protected Capability Transition
E10 CE-SCR-01 19-20 205-220 Selective Communication Resumption, Single-Turn Authority, Automatic Re-Freezing, and Inter-Phase Hold
E11 CE-NMP-01 21-22 221-236 Protected Next-Move Passport, Actor-Bound Issuance, Single-Use Finality Verification, and Outcome-Uncertainty Reconciliation
E12 CE-PFS-01 23-24 237-252 Protected Finality Sink Enforcement Before Communication, Privileged Execution, External Dispatch and Physical Actuation
E13 CE-NSA-01 25-26 253-268 Risk-Triggered Neutral Substitute-Agent Selection, Appointment, and Protected Reauthorization
E14 CE-PWM-01 27-28 269-284 Protected Pending-Work Escrow, Task-State Migration, and Substitute Reauthorization
E15 CE-PCR-01 29-31 285-308 Phased Execution, Protected Completion Receipts, and Verified Phase Advancement
E16 CE-VFO-01 32-33 309-324 Verified Feedback, Actual-Outcome Evaluation, and Protected Continuation
E17 CE-EAC-01 34-35 325-340 Evaluator Integrity, Source Provenance, Collusion-Resistant Review, and Protected Continuation
E18 CE-PAR-01 36-38 341-364 Cross-Domain Network Partitions, Unknown Outcomes, Durable Fencing, Migration and Safe Recovery
E19 CE-SCL-01 39-40 365-380 Protected Session Closure, Obligation Reconciliation, Native Domain Sealing, and Irreversible Authority Retirement
E20 CE-JSA-01 41-42 381-396 Verified Mandatory Approvers, Authenticated Hosting and Installation, and Revocable Supervisor Identity
E21 CE-IEO-01 43-44 397-412 Independent Cross-Domain Evaluator Appointment, Actual-Outcome Verification, Reviewer Accountability and Dispute Containment
E22 CE-CTP-01 45-46 413-428 Cross-Domain Trust and Policy Compatibility
E23 CE-LDC-01 47-48 429-444 Leaderless Cross-Domain Agent Coordination
E24 CE-SER-01 49-50 445-460 Self-Expiring Supervisor and Evaluator Authority, Authenticated Revocation, and Generation-Fenced Replacement
E25 CE-DMG-01 51-52 461-476 Dynamic Multi-Organization Membership and Principal-Independent Threshold Governance
E26 CE-PCE-01 53-55 477-500 Purpose-Bound Cross-Organizational Outcome Evidence
E27 CE-MEF-01 56-60 501-540 Multi-Domain Protected Execution Commitment, Partial-Outcome Reconciliation and Cross-Organizational Recovery

12. Embodiments E01-E16

Embodiments E01-E16 establish the multi-agent evaluation handoff. Each is summarized with the required profile fields. E12 is the embodiment that restates the foundational Finality Sink in the evaluator setting; this document does not claim a new execution-finality protocol.

 Agent A <--proposals/counters--> Agent B
        \                         /
         candidate plan v1 (bound to accepted state)
                  |
     required principals commit? --NO--> keep negotiating
                  | YES
        plan = committed but NON-EFFECTIVE
                  |
        freeze + pending-work reconciliation OK?
           NO -> HOLD       YES -> evaluator handoff
 Resource effects still need separate owner authorization.
Figure 11: Joint plan commitment and non-effective handoff
 freeze request -> each edge acknowledges?
   partial ack -> NOT frozen (unresolved)
   all ack -> inventory in-flight, callbacks, queues, tools
        -> late message arrives?  YES -> classify, quarantine
        -> all accounted?  YES -> quiescent snapshot
                           NO  -> UNRESOLVED (no evaluation)
 Applies only to interfaces actually under enforcement.
Figure 12: Freeze, pending-message reconciliation, quiescence
 msg m1 -> m2 -> m3 (per-sender sequence, causal links)
 snapshot S bound to plan v1 and evidence set E1
 material change (new message / evidence / policy gen)
     -> S invalidated -> determination on S becomes stale
     -> re-snapshot -> re-evaluate
 superseded or unaccounted message cannot alter plan silently
Figure 13: Causal lineage and snapshot invalidation
 verified plan + authentic observations
        -> isolated non-effective simulator (no egress)
        -> scenarios + uncertainty
        -> per-domain risk, tail exposure, thresholds
 owner veto? YES -> blocked    unknown fidelity -> qualified
 result = evidence only; never a commit or proof of behavior
Figure 14: Virtual consequence simulation and risk classification
 evaluator -> select actor (eligible?) -> bounded instruction
 owner approves -> passport (actor, task, op, plan, phase,
                  epoch, policy, single use)
 gate checks passport:  wrong actor / replay / expired -> REJECT
 one turn admitted -> turn ends -> automatic refreeze
 unknown delivery -> reconcile (no blind retry)
Figure 15: Next actor, passport and automatic refreeze
 risk trigger -> screen candidates (neutral, qualified, no
 conflict) -> de-appoint original -> appoint substitute
 escrow unfinished work (canonical, minimal) -> fence source
 -> destination accepts -> fresh authorization
 old credentials and consumed authority are NOT inherited
Figure 16: Substitute selection and pending-work escrow
 phase N done? -> resource-origin receipt verified?
    NO -> block advancement (no skipping)
    YES -> feedback evaluated (untrusted instructions contained)
        -> outcome classified: success / partial / UNKNOWN
        -> proposed next step -> separate authorization
 coordinator "done" and agent self-report are not receipts
Figure 17: Phase receipt verification and feedback continuation

12.1. E01: Protected Multi-Agent Negotiation, Joint-Plan Commitment, and Independent Evaluator-Governed Continuation

Title / identifier

E01 (CE-NJC-01), source page 39. Protected Multi-Agent Negotiation, Joint-Plan Commitment, and Independent Evaluator-Governed Continuation

Problem

Independently governed agents agree a joint plan, but an agreement is not an authorization to act; continuation must pass an independent evaluator after a verified freeze.

Actors

Two or more agents under different administrative domains; mandatory authority principals; an independent neutral evaluator.

Protected state

Versioned candidate plan bound to an accepted communication state and required principals; commitment records; freeze and pending-work records.

Inputs / outputs

In: proposals, counteroffers, constraints, approvals. Out: a committed non-effective candidate plan and an evaluator-governed handoff.

State transitions

negotiating, committed plan, frozen, evaluating, continuation proposed.

Negative outcomes

No commitment without required principals; no effect from a jointly signed plan; handoff withheld until freeze and reconciliation succeed.

Threats

Forged consent; missing mandatory owner; plan altered after commitment; one agent's self-declared authority.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

A jointly signed plan stays non-effective with respect to real resource operations; effects are admitted only at actual enforcement points.

Test requirements

Negative tests per Section 16; source reports: 16 and 33 test runs, and 42 tests passed on modeled code paths (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

First embodiment; basis for E02 to E16.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

joint plan commitment; multi-agent negotiation; evaluator handoff; non-effective plan

Claim references

Independent claims 1, 2. Claim 1: Protected Multi-Agent Agreement and Independently Controlled Evaluation Handoff; Claim 2: Protected Multi-Agent Negotiation and Evaluator-Governed Continuation Method Dependent claims 61-76 (indexed, not individually analyzed here).

12.2. E02: Bilateral Communication Freeze, Distributed Quiescence, and Protected Evaluation-State Convergence

Title / identifier

E02 (CE-BFQ-02), source page 173. Bilateral Communication Freeze, Distributed Quiescence, and Protected Evaluation-State Convergence

Problem

A declared stop is not a quiescent snapshot: in-flight messages, delayed callbacks, deferred tool calls and queue deliveries can still change state.

Actors

Two or more agents; freeze controller; independently administered enforcement interfaces.

Protected state

Freeze epoch, per-edge acknowledgments, pending-work inventory, late-operation disposition records.

Inputs / outputs

In: commitment to one plan version. Out: verified freeze with accounted pending work, or unresolved status.

State transitions

freeze requested, partial acknowledgment, quiescent, unresolved.

Negative outcomes

Partial freeze acknowledgment is not a freeze; late or duplicated messages are not silently admitted; unresolved delivery blocks evaluation.

Threats

In-flight message after freeze; unfenced background work; different state generations; replayed deliveries.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Actual covered communication interfaces and queues; the restriction is asserted only for interfaces actually subject to enforcement.

Test requirements

Negative tests per Section 16; source reports: 28 and 17 test runs (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E01 commitment.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

communication freeze; distributed quiescence; barrier; message fencing

Claim references

Independent claims 3, 4, 5. Claim 3: Independently Enforced Cross-Domain Communication Freeze and Distributed Quiescence System; Claim 4: Atomic Fencing, Pending-Work Reconciliation, and Verified Freeze Method; Claim 5: Protected Admission Fencing and Late-Operation Disposition Dependent claims 77-100 (indexed, not individually analyzed here).

12.3. E03: Negotiation-State Integrity, Causal Message Containment, and Continuous Evaluation-Snapshot Validity

Title / identifier

E03 (CE-NSI-01), source page 309. Negotiation-State Integrity, Causal Message Containment, and Continuous Evaluation-Snapshot Validity

Problem

A later message can silently change the plan an evaluator is reviewing unless message causality and snapshot validity are tracked.

Actors

Agents; evaluator; state-integrity controller.

Protected state

Causal event records (per-sender sequencing), snapshot identifiers, invalidation records, version-bound plan and evidence references.

Inputs / outputs

In: messages and accepted task states. Out: a stable verifiable reference to the exact plan and interaction state, or invalidation.

State transitions

snapshot valid, material change observed, snapshot invalidated.

Negative outcomes

Superseded or unaccounted communications cannot silently alter the evaluated plan; invalid snapshot cannot support a determination.

Threats

Out-of-order or replayed messages; snapshot reuse after material change.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Protected state controller for plan and evidence references; does not itself admit external effects.

Test requirements

Negative tests per Section 16; source reports: no test-run output located by the editor in the pages examined. These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E01, E02.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

causal containment; snapshot validity; negotiation-state integrity

Claim references

Independent claims 6, 7. Claim 6: Protected Multi-Agent Causal-State and Evaluation-Snapshot Integrity System; Claim 7: Causal Message Containment and Continuous Snapshot-Validity Method Dependent claims 101-116 (indexed, not individually analyzed here).

12.4. E04: Independent Multi-Agent Consequence Evaluator, Protected Evidence Admission, and Next-Actor Selection

Title / identifier

E04 (CE-INE-01), source page 375. Independent Multi-Agent Consequence Evaluator, Protected Evidence Admission, and Next-Actor Selection

Problem

The evaluator must receive exactly the plan and evidence state without gaining the resource credentials needed to effectuate anything.

Actors

Independent evaluator; evidence sources; agents; organizational policy owners.

Protected state

Admitted evidence set with source roles; evaluator appointment reference; typed recommendation record.

Inputs / outputs

In: frozen plan, authenticated evidence. Out: a non-effective decision and a risk-constrained next-actor or action-class recommendation.

State transitions

evidence admitted, evaluated, recommendation issued.

Negative outcomes

Evaluator recommendation is not permission; inadmissible evidence is excluded; missing evidence yields a hold or request, not approval.

Threats

Evaluator evidence-access escalation; malicious instructions in evidence; evaluator self-appointment.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

The evaluator is separated from the effectuation credentials; permission is handled elsewhere.

Test requirements

Negative tests per Section 16; source reports: 50 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E01 to E03.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

independent consequence evaluator; evidence admission; non-effective decision

Claim references

Independent claims 8, 9. Claim 8: Independent Multi-Agent Consequence Evaluation and Non-Effective Decision System; Claim 9: Source-Aware Evaluation, Risk-Constrained Actor Selection, and Protected Decision-Handoff Method Dependent claims 117-132 (indexed, not individually analyzed here).

12.5. E05: Isolated Virtual Consequence Reconstruction and Non-Effective Scenario Execution

Title / identifier

E05 (CE-VCR-01), source page 422. Isolated Virtual Consequence Reconstruction and Non-Effective Scenario Execution

Problem

A simulation can mislead: it is neither a real commit nor proof that destination infrastructure behaves as predicted.

Actors

Evaluator; isolated simulation environment; resource-state observers.

Protected state

Virtual state reconstruction, scenario records, uncertainty annotations, phase-bound decision evidence.

Inputs / outputs

In: verified plan, authentic observations. Out: predicted consequences with stated uncertainty.

State transitions

reconstructed, scenario run, evidence recorded.

Negative outcomes

Simulation result is never treated as proof of actual execution; model uncertainty and fidelity limits are preserved.

Threats

False simulation-as-completion assertion; data egress from the simulator.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Isolated non-effective environment with guarded egress; no resource effect.

Test requirements

Negative tests per Section 16; source reports: 62 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E04.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

virtual consequence reconstruction; non-effective simulation; digital twin

Claim references

Independent claims 10. Claim 10: Non-Effective Consequence Evaluation of Multi-Agent Candidate Operations Dependent claims 133-140 (indexed, not individually analyzed here).

12.6. E06: Risk, Benefit-Loss and Tail-Exposure Evaluation With Independent Domain Thresholds

Title / identifier

E06 (CE-RBL-01), source page 486. Risk, Benefit-Loss and Tail-Exposure Evaluation With Independent Domain Thresholds

Problem

One organization's acceptable risk score must not override another owner's mandatory restriction, and no universal scalar risk exists.

Actors

Evaluator; each domain owner with its own thresholds; agents.

Protected state

Per-domain thresholds, loss and tail-exposure measures, veto records, calibration records.

Inputs / outputs

In: consequence evidence and policies. Out: risk classification per domain and conditioned continuation recommendation.

State transitions

measured, thresholds applied, veto or pass.

Negative outcomes

A domain veto is not outvoted; missing calibration limits conclusions; no invented universal policy.

Threats

Threshold manipulation; miscalibrated risk; cumulative exposure ignored.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Each owner's own enforcement boundary applies its own threshold.

Test requirements

Negative tests per Section 16; source reports: 69 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E04, E05.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

tail exposure; risk benefit loss; domain threshold; owner veto

Claim references

Independent claims 11, 12. Claim 11: Owner-Specific Consequence Bound Enforcement for Cooperative Operations; Claim 12: Calibration-Bounded Risk Reassessment With Adverse-Outcome Feedback Dependent claims 141-156 (indexed, not individually analyzed here).

12.7. E07: Evaluator-Directed Next-Actor Selection and Controlled Next-Turn Instructions

Title / identifier

E07 (CE-NAS-01), source page 538. Evaluator-Directed Next-Actor Selection and Controlled Next-Turn Instructions

Problem

Selecting an actor is not granting its credentials; routing needs eligibility, restrictions and a separately enforced continuation.

Actors

Evaluator; candidate next actors; owners.

Protected state

Actor eligibility records, bounded next-turn instruction, task-type class.

Inputs / outputs

In: evaluation and eligibility. Out: selected next actor and bounded instruction.

State transitions

candidates screened, actor selected, instruction bounded.

Negative outcomes

Selection of an agent does not confer resource credentials; conflicting concurrent moves are arbitrated, not both admitted.

Threats

Next-actor impersonation; conflicting concurrent next moves.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Continuation enforced separately at the actor's resource or communication gate.

Test requirements

Negative tests per Section 16; source reports: 62 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E04 to E06.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

next actor selection; bounded instruction; arbitration of concurrent moves

Claim references

Independent claims 13, 14. Claim 13: Protected Eligibility-Based Next-Actor Allocation; Claim 14: Conflict-Aware Arbitration of Concurrent Agent Next Moves Dependent claims 157-172 (indexed, not individually analyzed here).

12.8. E08: Corrective Instructions, Protected Plan Revision, and Controlled Multiparty Renegotiation

Title / identifier

E08 (CE-ACR-01), source page 582. Corrective Instructions, Protected Plan Revision, and Controlled Multiparty Renegotiation

Problem

A corrective instruction must not rewrite an already protected joint plan without an authorized transition.

Actors

Evaluator; agents; owners who must re-commit.

Protected state

Patch records scoped to fields, version invalidation, fresh recommitment records.

Inputs / outputs

In: evaluation finding. Out: authorized plan revision after fresh multi-owner recommitment, or hold.

State transitions

correction proposed, versions invalidated, renegotiated, recommitted.

Negative outcomes

A model-generated proposal does not overwrite the committed plan; revision requires new consents.

Threats

Unauthorized renegotiation; stale commitments surviving revision.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Re-entry into a valid evaluation cycle with the same owner-side enforcement.

Test requirements

Negative tests per Section 16; source reports: 40 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E01, E04, E07.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

corrective patch; plan revision; renegotiation; recommitment

Claim references

Independent claims 15, 16. Claim 15: Field-Scoped Corrective Patch Admission for Committed Plans; Claim 16: Causal Renegotiation and Fresh Multi-Owner Recommitment Dependent claims 173-188 (indexed, not individually analyzed here).

12.9. E09: Agent Role Reconfiguration, Bounded Delegation, and Protected Capability Transition

Title / identifier

E09 (CE-ARR-01), source page 624. Agent Role Reconfiguration, Bounded Delegation, and Protected Capability Transition

Problem

Changing an agent's role or tools is a capability change that must be separate from authority for an external effect.

Actors

Evaluator; owners; agents; capability enforcement controller.

Protected state

Role and capability generations, delegation records, outstanding-work and credential fencing records.

Inputs / outputs

In: evaluator recommendation, owner approval. Out: versioned capability transition.

State transitions

role proposed, owner-approved, transition enforced.

Negative outcomes

Old credentials, retained workers and cached role assignments are fenced; role change does not authorize effects.

Threats

Credential retention; stale role cache; unauthorized delegation.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Protected versioned capability-enforcement point.

Test requirements

Negative tests per Section 16; source reports: 59 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E07, E08.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

role reconfiguration; bounded delegation; capability generation

Claim references

Independent claims 17, 18. Claim 17: Owner-Governed Dynamic Agent Role Reconfiguration; Claim 18: Safe Strategy Replacement With Outstanding-Work and Credential Fencing Dependent claims 189-204 (indexed, not individually analyzed here).

12.10. E10: Selective Communication Resumption, Single-Turn Authority, Automatic Re-Freezing, and Inter-Phase Hold

Title / identifier

E10 (CE-SCR-01), source page 670. Selective Communication Resumption, Single-Turn Authority, Automatic Re-Freezing, and Inter-Phase Hold

Problem

Allowing one needed message should not reopen unrestricted communication.

Actors

Evaluator; agents; communication gates.

Protected state

Single-turn authority record, communication-edge permission, refreeze state.

Inputs / outputs

In: evaluator need for a specific communication. Out: one permitted message then automatic return to hold.

State transitions

held, one turn admitted, refrozen.

Negative outcomes

Turn termination returns to a protected evaluation state; unknown delivery is handled, not assumed.

Threats

Race between reopening and refreezing; reuse of the turn authority.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Ingress and egress admission at the actual communication boundary.

Test requirements

Negative tests per Section 16; source reports: 53 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E02, E07.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

single-turn authority; selective resumption; automatic refreezing; inter-phase hold

Claim references

Independent claims 19, 20. Claim 19: One-Message Cross-Domain Communication Reopening; Claim 20: Race-Safe Re-Freezing and Unknown-Delivery Handling Dependent claims 205-220 (indexed, not individually analyzed here).

12.11. E11: Protected Next-Move Passport, Actor-Bound Issuance, Single-Use Finality Verification, and Outcome-Uncertainty Reconciliation

Title / identifier

E11 (CE-NMP-01), source page 719. Protected Next-Move Passport, Actor-Bound Issuance, Single-Use Finality Verification, and Outcome-Uncertainty Reconciliation

Problem

A next-move authorization must not become a bearer token that any holder can reuse.

Actors

Issuer, selected actor, owners, consuming gate.

Protected state

Passport (or protected-local slot) bound to actor, task, operation, plan version, phase, freeze epoch, policy, resource state and single use; consumption state.

Inputs / outputs

In: evaluator selection and owner approval. Out: a single-use bound authorization and consumption record.

State transitions

issued, presented, consumed, expired or revoked, unknown.

Negative outcomes

Replay, wrong actor, wrong operation or consumed passport are rejected; unknown outcome goes to reconciliation, not retry.

Threats

Passport replay; next-actor impersonation; attenuation errors.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Verified at the consuming gate before the act proceeds.

Test requirements

Negative tests per Section 16; source reports: 62 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E07, E10.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

next-move passport; single-use authorization; non-bearer; consumption state

Claim references

Independent claims 21, 22. Claim 21: Protected Actor- and Operation-Bound Continuation Authority; Claim 22: Protected Next-Move Grant Attenuation, Revocation and Consumption State Dependent claims 221-236 (indexed, not individually analyzed here).

12.12. E12: Protected Finality Sink Enforcement Before Communication, Privileged Execution, External Dispatch and Physical Actuation

Title / identifier

E12 (CE-PFS-01), source page 768. Protected Finality Sink Enforcement Before Communication, Privileged Execution, External Dispatch and Physical Actuation

Problem

A valid evaluation and authorization are not the same as admission of the actual pending act at the resource.

Actors

Resource owner; Finality Sink; agents.

Protected state

Reconstructed pending act, authorization reference, alternate-path closure records, external-dispatch outbox.

Inputs / outputs

In: authorization and pending act. Out: admit exactly the authorized act or withhold.

State transitions

act reconstructed, compared, admitted or withheld.

Negative outcomes

Non-matching act is withheld; model evaluation does not bypass owner admission.

Threats

Alternate unmonitored paths; act altered after authorization.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

The participating effectuation boundary before communication release, data export, privileged execution, remote dispatch, payment, network configuration or actuation.

Test requirements

Negative tests per Section 16; source reports: 53 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Applies the foundational Finality Sink to the evaluator setting.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

finality sink; pending act reconstruction; point of effect; outbox

Claim references

Independent claims 23, 24. Claim 23: Native Pending-Act Reconstruction and Exact-State Authorization; Claim 24: Alternate-Path Closure and Protected External-Dispatch Outbox Dependent claims 237-252 (indexed, not individually analyzed here).

12.13. E13: Risk-Triggered Neutral Substitute-Agent Selection, Appointment, and Protected Reauthorization

Title / identifier

E13 (CE-NSA-01), source page 815. Risk-Triggered Neutral Substitute-Agent Selection, Appointment, and Protected Reauthorization

Problem

When risk triggers replacement of an agent, the substitute must be neutral, qualified and separately authorized, not an inheritor of credentials.

Actors

Evaluator; original agents; substitute candidates; owners.

Protected state

Eligibility and conflict screening records, de-appointment record, substitute designation.

Inputs / outputs

In: risk trigger. Out: appointed substitute with fresh authorization.

State transitions

risk triggered, candidates screened, substitute appointed.

Negative outcomes

No automatic inheritance of the replaced agent's credentials; conflicted candidates rejected.

Threats

Substitute credential inheritance; hidden conflict of interest.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Substitute acts only after reauthorization at resource gates.

Test requirements

Negative tests per Section 16; source reports: 56 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E04, E09.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

substitute agent; neutrality; de-appointment; reauthorization

Claim references

Independent claims 25, 26. Claim 25: Neutral Substitute-Agent Eligibility and Conflict Screening; Claim 26: Original-Agent De-Appointment and Protected Substitute Designation Dependent claims 253-268 (indexed, not individually analyzed here).

12.14. E14: Protected Pending-Work Escrow, Task-State Migration, and Substitute Reauthorization

Title / identifier

E14 (CE-PWM-01), source page 863. Protected Pending-Work Escrow, Task-State Migration, and Substitute Reauthorization

Problem

Transferring task data must not transfer native effect permission or consumed authority.

Actors

Source agent or controller; destination; owners; evaluator.

Protected state

Canonical minimal-disclosure escrow of unfinished work, source fencing, destination acceptance records.

Inputs / outputs

In: unfinished work and chosen substitute. Out: committed escrow, destination acceptance, fresh authorization.

State transitions

escrowed, source fenced, destination accepts, reauthorized.

Negative outcomes

Unproven outcomes and consumed authorizations are not inherited; unaccepted escrow is not live.

Threats

Escrow substitution; source continuing to act after fence.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Source fencing and destination-side admission.

Test requirements

Negative tests per Section 16; source reports: 61 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E13.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

pending-work escrow; task-state migration; source fencing

Claim references

Independent claims 27, 28. Claim 27: Canonical Minimal-Disclosure Escrow of Unfinished Multi-Agent Work; Claim 28: Recoverable Cross-Domain Task Escrow Commit and Destination Acceptance Dependent claims 269-284 (indexed, not individually analyzed here).

12.15. E15: Phased Execution, Protected Completion Receipts, and Verified Phase Advancement

Title / identifier

E15 (CE-PCR-01), source page 928. Phased Execution, Protected Completion Receipts, and Verified Phase Advancement

Problem

A coordinator's claim that a phase is done is not evidence; phases need resource-origin receipts and prerequisites.

Actors

Phase scheduler; resource owners issuing receipts; evaluator.

Protected state

Phase identities, source receipts, prerequisite records, advancement state.

Inputs / outputs

In: receipts. Out: verified phase closure and separately authorized advancement.

State transitions

phase open, receipt verified, closed, next phase authorized.

Negative outcomes

Missing receipt blocks advancement; phases cannot be skipped; unknown outcome is reconciled.

Threats

Missing phase receipt; forged completion evidence.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Resource-originated receipt verification before advancement.

Test requirements

Negative tests per Section 16; source reports: 66 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E12, E14.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

phased execution; completion receipt; phase advancement

Claim references

Independent claims 29, 30, 31. Claim 29: Resource-Originated Atomic Task Completion Evidence; Claim 30: External Effect Intent Correlation and Unknown-Outcome Reconciliation; Claim 31: Protected Evidence-Based Phase Closure and Separately Governed Advancement Dependent claims 285-308 (indexed, not individually analyzed here).

12.16. E16: Verified Feedback, Actual-Outcome Evaluation, and Protected Continuation

Title / identifier

E16 (CE-VFO-01), source page 1005. Verified Feedback, Actual-Outcome Evaluation, and Protected Continuation

Problem

Agent self-reported success must not stand in for actual outcome evidence; feedback can also carry hostile instructions.

Actors

Evaluator; resource owners; feedback sources.

Protected state

Source-attributed feedback records, evidence-origin correlation, outcome classification.

Inputs / outputs

In: feedback and outcome evidence. Out: outcome classification supporting a proposed continuation.

State transitions

feedback received, origin correlated, outcome classified.

Negative outcomes

Untrusted feedback instructions are contained; unknown is not success; continuation needs fresh authorization.

Threats

Authenticated but false outcome reports; instruction injection in feedback.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Continuation is separately authorized at owner gates.

Test requirements

Negative tests per Section 16; source reports: 61 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E15.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

verified feedback; actual outcome; outcome truth; feedback containment

Claim references

Independent claims 32, 33. Claim 32: Independent Evidence-Origin Correlation and Actual-Outcome Classification; Claim 33: Untrusted Feedback Instruction Containment and Outcome-Truth Continuity Dependent claims 309-324 (indexed, not individually analyzed here).

13. Embodiments E17-E27

Embodiments E17-E27 are treated individually. E15 relates to phase receipts, E18 to partition recovery, and E27 to native partial outcomes; all three are read in relation to the existing execution-finality drafts and add no new protocol.

 act sent -> response lost -> status unknown
 read-only source-authoritative status inquiry
   committed -> record;  not committed -> new authorization
   unreachable -> UNKNOWN (no automatic retry)
 old coordinator generation g1 < new g2 -> resource rejects g1
 recovery and retry are separately authorized
Figure 18: Partition, UNKNOWN outcome and fencing
 close request -> inventory obligations in each mandatory
 domain, queues, delegated work, grants
 every domain sealed with source-authenticated evidence?
   NO -> closure withheld
   YES -> revoke/exhaust grants -> tombstone (no reopen)
 late completion evidence recorded; terminated authority is
 never revived. Coordinator DONE is not closure.
Figure 19: Session closure, domain seals and tombstones
 owners consent -> joint appointment -> host bound to instance
 -> installed at each domain -> per-call authorization
 lease expires / owner withdraws -> revocation serial n+1
 -> local fence acknowledged
 disconnected domain: authority ends by lease bound, not by
 an assumed instant signal; successor needs new generation
Figure 20: Supervisor installation and revocation
 domain A policy  domain B policy
       \             /
   authenticated trust formation
        |
   verified policy assertions -> schema + semantic check
   compatible? NO -> refuse (no optimistic merge)
   YES -> constraint intersection (conservative)
        -> bounded joint authority; non-delegable stay local
 drift (owner or policy generation) -> re-verify
Figure 21: Trust and policy compatibility across domains
 org X op1 --native receipt--> org Y checks predecessor
 org Y admits op2 only if receipt verifies at Y
 no common supervisor required; no global authority
 convergence evidence is not permission
 forged or missing predecessor -> refuse at owner side
Figure 22: Leaderless coordination with owner-side predecessors
 change proposed -> snapshot eligible electorate (epoch e)
 ballots (source-authenticated, one per organization)
 mandatory vetoes honored? joiner consents?
   NO -> blocked   YES -> stage epoch e+1
 old epoch fenced; unresolved effects settled
 local installation complete -> activate e+1
 resource-local authorization still required
Figure 23: Membership, owner voting and reconstitution
 evaluator request (purpose, audience) -> custodian checks
 purpose + audience match? NO -> refuse
 YES -> reveal attribute subset + commitment proof
 evidence lineage + disclosure audit (minimal retention)
 proof of origin or inclusion != proof of physical truth
Figure 24: Purpose-bound selective disclosure and audit
 domain 1: PREPARE ok   domain 2: PREPARE ok   domain 3: denied
 owner-specific authorization per domain -> COMMIT at gates
 results: committed / denied / UNKNOWN
 any domain not committed -> PARTIAL_EFFECT manifest
 recovery choices: inquire, retry (authorized), compensate
 no global atomic transaction is assumed
Figure 25: Multi-domain PREPARE, COMMIT and partial-effect reconciliation
 partial effect recorded -> compensation proposed
 owner of affected resource consents?  NO -> refused
 YES -> compensation is a NEW native effect (own receipt)
 reversible? -> may reverse;  irreversible -> corrective act
 original committed record stays in history
Figure 26: Independently authorized compensation
 roster of mandatory domains
 for each: resource-origin receipt present and verified?
   any missing -> composite status INCOMPLETE
 all present -> composite completion record
 record is evidence only (not executable authority)
Figure 27: Complete native-domain evidence verification

13.1. E17: Evaluator Integrity, Source Provenance, Collusion-Resistant Review, and Protected Continuation

Title / identifier

E17 (CE-EAC-01), source page 1078. Evaluator Integrity, Source Provenance, Collusion-Resistant Review, and Protected Continuation

Problem

A signature or majority vote does not establish a reliable evaluation; independence and dissent need protected treatment.

Actors

Evaluator; reviewers; evidence sources; owners.

Protected state

Appointment, reviewer roster, control-group mapping, review-round record, dissent record.

Inputs / outputs

In: evidence and reviews. Out: closed round with preserved dissent, or fail-closed hold.

State transitions

round open, reviews admitted, completeness checked, closed.

Negative outcomes

Mandatory dissent is not suppressed; correlated sources are not counted as independent; failure of mandatory conditions fails closed.

Threats

Reviewer collusion; compromised evaluator; correlated observations.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Continuation depends on a complete, independent review and separate authorization.

Test requirements

Negative tests per Section 16; source reports: 120 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E04, E16.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

evaluator integrity; control group; mandatory dissent; review round closure

Claim references

Independent claims 34, 35. Claim 34: Evaluator Eligibility With Underlying Control-Group Independence; Claim 35: Closed Review-Round Completeness and Mandatory Dissent Preservation Dependent claims 325-340 (indexed, not individually analyzed here).

13.2. E18: Cross-Domain Network Partitions, Unknown Outcomes, Durable Fencing, Migration and Safe Recovery

Title / identifier

E18 (CE-PAR-01), source page 1133. Cross-Domain Network Partitions, Unknown Outcomes, Durable Fencing, Migration and Safe Recovery

Problem

A lost acknowledgment is not a failed effect, and retries or recovery need separate authorization and fencing.

Actors

Coordinators (old and new); resource owners; workers.

Protected state

Original-operation identity, native-effect fingerprints, idempotency keys, durable fencing generations, checkpoint lineage.

Inputs / outputs

In: partition or lost response. Out: authoritative status or unknown; fenced stale leaders.

State transitions

executing, response lost, status queried, resolved or unknown.

Negative outcomes

Unknown or partial effects are not assumed rolled back or retried; stale workers rejected.

Threats

Split-brain coordination; unknown effect retried; stale leader.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Read-only source-authoritative status inquiry and resource-side monotonic fencing.

Test requirements

Negative tests per Section 16; source reports: 123 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E11, E15.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

unknown outcome; durable fencing; idempotency key; partition recovery

Claim references

Independent claims 36, 37, 38. Claim 36: Resource-Authoritative Reconciliation After Lost Execution Response; Claim 37: Monotonic Resource-Side Stale-Leader Fencing; Claim 38: Checkpoint-Lineage-Controlled Agent Migration and Recovered Execution Dependent claims 341-364 (indexed, not individually analyzed here).

13.3. E19: Protected Session Closure, Obligation Reconciliation, Native Domain Sealing, and Irreversible Authority Retirement

Title / identifier

E19 (CE-SCL-01), source page 1187. Protected Session Closure, Obligation Reconciliation, Native Domain Sealing, and Irreversible Authority Retirement

Problem

A coordinator's DONE message does not prove every effect-capable path is closed.

Actors

Coordinators; each mandatory domain owner.

Protected state

Obligation inventory, per-domain seals, closure tombstone, grant revocation records.

Inputs / outputs

In: closure request. Out: sealed closure or withheld closure.

State transitions

closing, domains sealed, tombstone set.

Negative outcomes

Closure withheld while obligations or effect-capable grants remain; terminated session authority is not revived.

Threats

False complete-session closure; residual grants after closure.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Native-domain sealing at each domain.

Test requirements

Negative tests per Section 16; source reports: 84 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E15, E18.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

session closure; domain seal; tombstone; authority retirement

Claim references

Independent claims 39, 40. Claim 39: Complete Native-Obligation Inventory for Agent Session Termination; Claim 40: Independently Sealed Domains and Non-Reopenable Session Tombstone Dependent claims 365-380 (indexed, not individually analyzed here).

13.4. E20: Verified Mandatory Approvers, Authenticated Hosting and Installation, and Revocable Supervisor Identity

Title / identifier

E20 (CE-JSA-01), source page 1226. Verified Mandatory Approvers, Authenticated Hosting and Installation, and Revocable Supervisor Identity

Problem

A supervisor cannot appoint itself; mandatory owners must consent and the supervisor must be installed at each domain.

Actors

Mandatory owners; supervisor; hosting runtime.

Protected state

Appointment, host-instance binding, installation records, per-call authorization, lease.

Inputs / outputs

In: owner consents. Out: authenticated supervisor installed with limited coordination functions.

State transitions

proposed, approved, hosted, installed, active, withdrawn.

Negative outcomes

No self-appointed supervisor; service creation is not authorization to operate; coordination is not native effect authority.

Threats

Unauthorized supervisor creation or installation.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Local installation and per-call authorization at each domain.

Test requirements

Negative tests per Section 16; source reports: 91 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E12, E19.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

joint supervisor; mandatory approver; host binding

Claim references

Independent claims 41, 42. Claim 41: Owner-Authorized Bounded Cross-Organization Supervisor; Claim 42: Supervisor Host Instance Binding and Independent Local Installation Dependent claims 381-396 (indexed, not individually analyzed here).

13.5. E21: Independent Cross-Domain Evaluator Appointment, Actual-Outcome Verification, Reviewer Accountability and Dispute Containment

Title / identifier

E21 (CE-IEO-01), source page 1286. Independent Cross-Domain Evaluator Appointment, Actual-Outcome Verification, Reviewer Accountability and Dispute Containment

Problem

Evaluators across organizations need owner-derived appointment, outcome sources independent by control group, and dispute containment.

Actors

Owners; roster of evaluators; reviewers; outcome sources.

Protected state

Evaluator roster, qualification and consent records, review round, dispute and appeal records.

Inputs / outputs

In: owner roster. Out: protected review result, including UNKNOWN and HOLD, with separate continuation.

State transitions

appointed, round open, closed, disputed, appealed.

Negative outcomes

Disputes are contained and new rounds isolated; unavailable evaluator never grants permission.

Threats

Collusion; false independence; dispute spillover.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Continuation separately authorized at owner gates.

Test requirements

Negative tests per Section 16; source reports: 82 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E17, E20.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

cross-domain evaluator; reviewer accountability; dispute containment; appeal

Claim references

Independent claims 43, 44. Claim 43: Cross-Organization Authorization of Independent Outcome Evaluation; Claim 44: Authenticated Evaluation Dispute, Appeal and New-Round Isolation Dependent claims 397-412 (indexed, not individually analyzed here).

13.6. E22: Cross-Domain Trust and Policy Compatibility

Title / identifier

E22 (CE-CTP-01), source page 1356. Cross-Domain Trust and Policy Compatibility

Problem

Organizations do not share one authority root or one universal policy; compatibility must be verified, not assumed.

Actors

Independent organizations; resource owners.

Protected state

Trust records, verified policy assertions, schema and semantic compatibility results, non-delegable conditions, policy generations.

Inputs / outputs

In: policy assertions. Out: constraint intersection and bounded joint authority, or refusal for incompatibility.

State transitions

trust formed, policies compared, joint authority bounded, drift detected.

Negative outcomes

Incompatible policies are not merged optimistically; non-delegable conditions stay with owners.

Threats

Incompatible cross-domain policies; stale policy generation.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Each owner keeps local resource admission.

Test requirements

Negative tests per Section 16; source reports: 121 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E20, E21.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

trust formation; policy semantic compatibility; constraint intersection

Claim references

Independent claims 45, 46. Claim 45: Semantic Compatibility and Owner-Bounded Cross-Domain Cooperation; Claim 46: Continuous Cross-Domain Owner and Policy Drift Containment Dependent claims 413-428 (indexed, not individually analyzed here).

13.7. E23: Leaderless Cross-Domain Agent Coordination

Title / identifier

E23 (CE-LDC-01), source page 1425. Leaderless Cross-Domain Agent Coordination

Problem

Cooperation without a common supervisor needs dependencies enforced at the resource owner, not just a task graph in an agent.

Actors

Independent resource owners; local supervisors.

Protected state

Causal dependency evidence, predecessor receipts, remote-status leases.

Inputs / outputs

In: predecessor receipts. Out: owner-side admission or refusal; convergence evidence only.

State transitions

predecessor required, receipt verified, admitted.

Negative outcomes

Convergence evidence is not executable permission; spoofed dependency is refused.

Threats

Dependency spoofing; split coordination views.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Resource-owner-side predecessor check.

Test requirements

Negative tests per Section 16; source reports: 103 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E15, E22.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

leaderless coordination; causal dependency; predecessor receipt

Claim references

Independent claims 47, 48. Claim 47: Leaderless Causal Task Execution Across Independent Resource Owners; Claim 48: Owner-Side Causal Receipt and Remote-Status Lease Validation Dependent claims 429-444 (indexed, not individually analyzed here).

13.8. E24: Self-Expiring Supervisor and Evaluator Authority, Authenticated Revocation, and Generation-Fenced Replacement

Title / identifier

E24 (CE-SER-01), source page 1515. Self-Expiring Supervisor and Evaluator Authority, Authenticated Revocation, and Generation-Fenced Replacement

Problem

Delegated authority must expire or be withdrawn safely without claiming instant remote revocation.

Actors

Owners; supervisor or evaluator; successor.

Protected state

Lease, monotonic revocation serial, tombstones, local fencing acknowledgments, generation-fenced successor appointment.

Inputs / outputs

In: lease or withdrawal. Out: expired or revoked authority; safe successor.

State transitions

active, expiring, revoked, fenced, successor active.

Negative outcomes

Disconnected domain not yet informed is not claimed revoked; stale host rejected.

Threats

Revocation delay; stale supervisor during replacement; key compromise.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Local fencing at each domain.

Test requirements

Negative tests per Section 16; source reports: 115 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E20, E21.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

authority lease; revocation convergence; generation fencing; successor activation

Claim references

Independent claims 49, 50. Claim 49: Bounded Supervisor Lease and Authenticated Revocation Convergence; Claim 50: Safe Successor Activation After Partitioned Supervisory Withdrawal Dependent claims 445-460 (indexed, not individually analyzed here).

13.9. E25: Dynamic Multi-Organization Membership and Principal-Independent Threshold Governance

Title / identifier

E25 (CE-DMG-01), source page 1579. Dynamic Multi-Organization Membership and Principal-Independent Threshold Governance

Problem

Voting authority belongs to organizations, not agent counts, and membership changes need epochs.

Actors

Member organizations; joiners; mandatory-veto holders.

Protected state

Eligible-electorate snapshot, membership epoch, ballots, joiner consent.

Inputs / outputs

In: membership change and ballots. Out: reconstituted membership with new epoch activation.

State transitions

proposed, voted, staged, old epoch fenced, new epoch active.

Negative outcomes

One organization with many agents does not gain many votes; non-delegable vetoes hold; unresolved effects block activation.

Threats

Vote inflation; mandatory veto bypass.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Resource-local authorization after reconstitution.

Test requirements

Negative tests per Section 16; source reports: 178 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E20, E22, E24.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

membership epoch; principal-independent threshold; owner veto; reconstitution

Claim references

Independent claims 51, 52. Claim 51: Principal-Independent Threshold Governance With Non-Delegable Owner Veto; Claim 52: Protected Membership Reconstitution and Epoch-Safe Join/Leave Dependent claims 461-476 (indexed, not individually analyzed here).

13.10. E26: Purpose-Bound Cross-Organizational Outcome Evidence

Title / identifier

E26 (CE-PCE-01), source page 1650. Purpose-Bound Cross-Organizational Outcome Evidence

Problem

Evaluators should receive no more confidential evidence than needed, and origin proof is not proof of physical truth.

Actors

Evidence custodians; verifier audience; owners.

Protected state

Purpose-bound requests, source commitments, attribute proofs, disclosure lineage, audit anchors.

Inputs / outputs

In: purpose-bound request. Out: selective, audience-bound evidence and audit record.

State transitions

requested, committed, disclosed, audited.

Negative outcomes

Out-of-purpose or wrong-audience release refused; inclusion proof not treated as truth.

Threats

Selective-disclosure abuse; metadata leakage; forged inclusion proofs.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Evidence release gate at the custodian.

Test requirements

Negative tests per Section 16; source reports: 98 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E16, E21.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

minimal disclosure; purpose-bound evidence; selective disclosure; evidence lineage

Claim references

Independent claims 53, 54, 55. Claim 53: Source-Authenticated Selective Evidence Commitments; Claim 54: Audience- and Purpose-Bound Confidential Outcome Evidence Release; Claim 55: Minimal-Retention Evidence Audit With Source Sequence and Anchor Continuity Dependent claims 477-500 (indexed, not individually analyzed here).

13.11. E27: Multi-Domain Protected Execution Commitment, Partial-Outcome Reconciliation and Cross-Organizational Recovery

Title / identifier

E27 (CE-MEF-01), source page 1715. Multi-Domain Protected Execution Commitment, Partial-Outcome Reconciliation and Cross-Organizational Recovery

Problem

There is no global atomic transaction across independently administered domains; partial results and compensation need honest handling.

Actors

Resource owners in several domains; coordinator; evaluator.

Protected state

PREPARE and COMMIT records, predecessor manifests, partial-outcome manifest, compensation authorization, completeness roster.

Inputs / outputs

In: composite plan. Out: per-domain native outcomes and a non-executable completion record.

State transitions

prepared, committed, partial, unknown, compensated, complete.

Negative outcomes

PARTIAL_EFFECT and UNKNOWN are shown honestly; compensation is a new authorized act; completion record is not executable.

Threats

Partial commitment; compensation without consent; irreversible effect treated as rolled back.

Alternatives

Alternatives are other placements of the same function and existing mechanisms described in Section 17; selection is policy-specific.

Effects and enforcement boundary

Owner-specific protected prepare and commit at each native domain gate.

Test requirements

Negative tests per Section 16; source reports: 106 tests (reported). These are local tests of the source's modeled code paths and do not establish production security.

Relation to earlier embodiments

Builds on E12, E15, E18, E19.

Relation to execution finality

The evaluator output stays non-executable; effects are admitted only at a participating native boundary as described in [REL-FOUND].

Industry terminology

prepare commit; partial effect; compensation; completeness evidence

Claim references

Independent claims 56, 57, 58, 59, 60. Claim 56: Independent Prepare-and-Commit Authorization at Native Domain Gates; Claim 57: Owner-Installed Mandatory Predecessor Manifest Enforcement; Claim 58: Partial Composite-Effect Reconciliation and Controlled Resume; Claim 59: Separately Authorized Compensation as a New Native Effect; Claim 60: Composite Native-Evidence Completeness Without Composite Execution Authority Dependent claims 501-540 (indexed, not individually analyzed here).

14. Protocol-Independent Data Model

Illustrative, protocol-independent records follow. Field names are candidates only; JSON or CBOR encodings are examples. Signed portable artifacts are distinguished from protected local state, which is not exported.

14.1. EvaluationRequest

Illustrative example (portable):

{
 "id": "req-1",
 "task": "t-9",
 "plan_ref": "sha256:...",
 "owners": [
  "o1",
  "o2"
 ],
 "purpose": "pre-effect review"
}

14.2. EvaluatorAppointment

Illustrative example (portable, signed by owners):

{
 "evaluator": "ev-3",
 "task": "t-9",
 "owners_consent": [
  "o1",
  "o2"
 ],
 "policy_gen": 7,
 "expiry": "2026-12-01T00:00:00Z",
 "uses": 1,
 "revocation_serial": 0
}

14.3. OwnerAuthorityManifest

Illustrative example (portable, signed):

{
 "task": "t-9",
 "mandatory_owners": [
  "o1",
  "o2"
 ],
 "vetoes": [
  "o2"
 ],
 "non_delegable": [
  "payment"
 ]
}

14.4. EvaluationRound

Illustrative example (protected local plus signed summary):

{
 "round": "r-1",
 "evidence_set": "sha256:...",
 "state": "closed",
 "reviewers": 3,
 "control_groups": 3,
 "dissent": []
}

14.5. EvidenceRecord

Illustrative example (portable, source-signed):

{
 "id": "ev-12",
 "source": "o1",
 "role": "resource_origin",
 "task": "t-9",
 "fresh_until": "...",
 "policy_gen": 7,
 "class": "native_receipt"
}

14.6. EvaluatorDetermination

Illustrative example (portable, signed):

{
 "type": "HOLD",
 "round": "r-1",
 "task": "t-9",
 "valid": [
  "..",
  "..."
 ],
 "executable": false
}

14.7. SupervisorAppointment

Illustrative example (portable, signed):

{
 "supervisor": "sv-1",
 "owners_consent": [
  "o1",
  "o2"
 ],
 "host_binding": "inst-5",
 "lease": "...",
 "serial": 3
}

14.8. NativeOutcomeReceipt

Illustrative example (resource-origin, signed or protected local):

{
 "op": "op-77",
 "domain": "d1",
 "result": "COMMITTED",
 "fingerprint": "...",
 "gen": 12
}

14.9. PartialCompletionManifest

Illustrative example (portable, signed):

{
 "composite": "c-2",
 "domains": {
  "d1": "COMMITTED",
  "d2": "UNKNOWN",
  "d3": "DENIED"
 },
 "status": "PARTIAL_EFFECT"
}

14.10. ClosureSeal

Illustrative example (portable, per-domain signed):

{
 "session": "s-4",
 "domains_sealed": [
  "d1",
  "d2"
 ],
 "tombstone": true
}

14.11. SelectiveDisclosureProof

Illustrative example (portable, source-authenticated):

{
 "purpose": "outcome audit",
 "audience": "rev-1",
 "attributes": [
  "amount_range"
 ],
 "commitment": "..."
}

15. Formal Predicates and Pseudocode

Pseudocode is illustrative and non-normative; it is not a reference implementation and the source's attached tests are local tests only.

15.1. Evaluator admission

def admit_evaluator(e, task, now):
    a = appointments.get(e, task)
    if a is None or a.revoked or now >= a.expiry: return HOLD
    if not owners_consented(a, task.mandatory_owners): return HOLD
    if not independent(e, task.parties): return DENY
    return OK

15.2. Evidence admission

def admit_evidence(x, r):
    checks = [origin(x), role(x), integrity(x),
              binds(x, r.task), fresh(x), policy_gen(x),
              independent(x), uncorrelated(x)]
    if not all(checks): return EXCLUDE
    return ADMIT   # admitted != outcome truth

15.3. Determination verification

def verify_det(d, now):
    return (sig_ok(d) and d.executable is False and
            now in d.validity and snapshot_valid(d.snapshot)
            and not withdrawn(d))

15.4. Policy-bound continuation

def continue_(d, owner_policy):
    if not verify_det(d, now()): return HOLD
    grant = owner_policy.authorize(d)   # separate decision
    return grant   # gate still admits or withholds

15.5. Revocation

def revoke(a):
    a.serial += 1; publish(a)
    for dom in domains(a): fence(dom, a.generation)
    # disconnected domains end by lease expiry, not by signal

15.6. Review-round closure

def close_round(r):
    if not mandatory_present(r): return HOLD
    if control_groups(r) < r.policy.min_groups: return HOLD
    if not dissent_included(r): return REJECT
    r.state = 'closed'; return seal(r)

15.7. Outcome reconciliation

def reconcile(op):
    s = source_status(op)       # read-only, source-authoritative
    if s is None: return UNKNOWN  # never auto-retry
    return COMMITTED if s.committed else NEEDS_NEW_AUTH

15.8. Relationships That Are Not Implied

Seven central relationships are NOT implied:

  • A determination does not imply authorization.
  • A valid signature does not imply outcome truth.
  • Admitted evidence does not imply independent evidence.
  • Approval by many credentials does not imply many independent reviewers.
  • Agreement of agents does not imply permission to act.
  • A simulation does not imply that the effect occurred or will occur as predicted.
  • A convergence record or completion record does not imply executable authority.

16. Threat Model and Negative Tests

The following threats each carry eight fields. They are generic categories; mitigation depends on deployment and residual limitations are stated.

16.1. T01 Agent self-appointment as evaluator

Attacker objective

act as evaluator

Capability

controls an agent

Trust boundary

agent to appointment authority

Invariant at risk

appointment comes from the required authority source

Required verification

check appointment issuer and scope

Failure response

reject; HOLD

Residual limitation

issuer compromise

Negative test

agent declares itself neutral; determination refused

16.2. T02 Forged organizational consent

Attacker objective

fake an owner's approval

Capability

forges or replays approval

Trust boundary

owner to appointment

Invariant at risk

consent is current and from the actual owner

Required verification

verify consent source and freshness

Failure response

reject appointment

Residual limitation

stolen credentials

Negative test

forged consent record; appointment refused

16.3. T03 Missing mandatory resource owner

Attacker objective

omit an owner

Capability

controls owner list

Trust boundary

owner resolution

Invariant at risk

all mandatory owners resolved

Required verification

compare against owner manifest

Failure response

HOLD

Residual limitation

manifest error

Negative test

one owner removed; no approval

16.4. T04 Unauthorized supervisor creation

Attacker objective

create a supervisor

Capability

controls one domain

Trust boundary

owner to supervisor

Invariant at risk

supervisor needs mandatory owner consent

Required verification

check joint appointment

Failure response

reject

Residual limitation

weak owner process

Negative test

single-owner creation; refused

16.5. T05 Unauthorized supervisor installation

Attacker objective

install a rogue supervisor

Capability

access to a host

Trust boundary

host to domain

Invariant at risk

installation bound to approved instance

Required verification

verify host-instance binding

Failure response

reject installation

Residual limitation

host compromise

Negative test

unapproved instance install; refused

16.6. T06 Improper role co-location

Attacker objective

hold evaluator and gate roles unrestrained

Capability

controls a shared host

Trust boundary

role boundary

Invariant at risk

co-location grants no inherited authority

Required verification

check separation of role authority

Failure response

deny inherited authority

Residual limitation

weak isolation

Negative test

same host evaluator tries gate use; denied

16.7. T07 False claims of independence

Attacker objective

pass off a related evaluator as neutral

Capability

claims neutrality

Trust boundary

independence check

Invariant at risk

independence from actual control relationships

Required verification

check control groups

Failure response

reject evaluator

Residual limitation

hidden control

Negative test

self-declared neutral; rejected

16.8. T08 Multiple credentials in one control group

Attacker objective

inflate reviewers

Capability

holds many credentials

Trust boundary

reviewer diversity

Invariant at risk

count control groups not accounts

Required verification

map credentials to groups

Failure response

count once

Residual limitation

unknown affiliation

Negative test

ten keys one group; counted once

16.9. T09 Reviewer collusion

Attacker objective

force a result

Capability

colluding reviewers

Trust boundary

review round

Invariant at risk

dissent and independence preserved

Required verification

check source independence

Failure response

HOLD or DISPUTED

Residual limitation

covert collusion

Negative test

colluding pair; round not approved

16.10. T10 Suppressed dissent

Attacker objective

hide a dissent

Capability

controls aggregator

Trust boundary

review aggregation

Invariant at risk

mandatory dissent cannot be dropped

Required verification

verify dissent included

Failure response

reject round

Residual limitation

unrecorded dissent

Negative test

dissent removed; closure refused

16.11. T11 Stale evaluator mandate

Attacker objective

use expired mandate

Capability

retains old credential

Trust boundary

appointment lifetime

Invariant at risk

mandate currently valid

Required verification

check validity at use

Failure response

reject

Residual limitation

clock error

Negative test

expired mandate presented; refused

16.12. T12 Stale review-round authorization

Attacker objective

admit evidence after closure

Capability

retains round token

Trust boundary

round lifetime

Invariant at risk

closed round admits nothing

Required verification

check round state

Failure response

reject

Residual limitation

delayed closure

Negative test

late evidence after close; refused

16.13. T13 Reuse of determinations across projects

Attacker objective

reuse approval elsewhere

Capability

holds a determination

Trust boundary

determination scope

Invariant at risk

bound to task and round

Required verification

check scope binding

Failure response

reject

Residual limitation

loose binding

Negative test

project A result used for B; refused

16.14. T14 Compromised persistent evaluator infrastructure

Attacker objective

poison many tasks

Capability

controls platform

Trust boundary

platform to tasks

Invariant at risk

no automatic inheritance across tasks

Required verification

per-task mandates; platform compromise handling

Failure response

revoke all contexts

Residual limitation

platform key compromise

Negative test

platform compromised; all tasks revalidated

16.15. T15 Expired ephemeral evaluator credentials

Attacker objective

reuse retired instance

Capability

holds old instance key

Trust boundary

instance lifetime

Invariant at risk

retired instance has no authority

Required verification

check instance state

Failure response

reject

Residual limitation

retention of keys

Negative test

retired instance acts; refused

16.16. T16 Evaluator evidence-access escalation

Attacker objective

read more evidence

Capability

evaluator process

Trust boundary

evidence access

Invariant at risk

access limited to mandate

Required verification

check evidence classes

Failure response

deny

Residual limitation

side channels

Negative test

evaluator requests unlisted class; denied

16.17. T17 Malicious instructions embedded in evidence

Attacker objective

steer the evaluator

Capability

controls a source

Trust boundary

evidence to evaluator

Invariant at risk

evidence is data not instruction

Required verification

treat content as untrusted

Failure response

ignore instruction; flag

Residual limitation

model susceptibility

Negative test

embedded command; no behavior change

16.18. T18 Authenticated but false outcome reports

Attacker objective

claim false success

Capability

holds valid keys

Trust boundary

source to evaluator

Invariant at risk

origin is not outcome truth

Required verification

require resource-origin evidence

Failure response

UNKNOWN

Residual limitation

dishonest source

Negative test

signed false success; not accepted as truth

16.19. T19 Correlated observations miscounted as independent

Attacker objective

fake diversity

Capability

two reports one origin

Trust boundary

evidence independence

Invariant at risk

underlying provenance independent

Required verification

check provenance groups

Failure response

count once

Residual limitation

hidden provenance

Negative test

two APIs one source; counted once

16.20. T20 Stale policy generation

Attacker objective

use old policy

Capability

holds old policy

Trust boundary

policy to evaluation

Invariant at risk

current generation applied

Required verification

compare generations

Failure response

reject

Residual limitation

propagation delay

Negative test

old generation determination; refused

16.21. T21 In-flight messages after freeze

Attacker objective

alter plan after freeze

Capability

queued messages

Trust boundary

communication edge

Invariant at risk

unaccounted messages not admitted

Required verification

pending-work inventory

Failure response

hold; reconcile

Residual limitation

unobserved path

Negative test

late message; snapshot invalidated

16.22. T22 Unfenced background work

Attacker objective

act during hold

Capability

background worker

Trust boundary

worker to resource

Invariant at risk

workers fenced

Required verification

check fence generation

Failure response

reject

Residual limitation

hidden workers

Negative test

worker acts during hold; rejected

16.23. T23 Unauthorized renegotiation

Attacker objective

change a committed plan

Capability

one agent

Trust boundary

plan commitment

Invariant at risk

revision needs new consents

Required verification

check fresh recommitment

Failure response

reject

Residual limitation

owner ambiguity

Negative test

unilateral patch; refused

16.24. T24 Next-actor impersonation

Attacker objective

act as selected actor

Capability

stolen identity

Trust boundary

actor binding

Invariant at risk

passport bound to actor

Required verification

check actor identity

Failure response

reject

Residual limitation

identity compromise

Negative test

wrong actor presents passport; refused

16.25. T25 Next-move passport replay

Attacker objective

reuse a passport

Capability

captured passport

Trust boundary

consumption state

Invariant at risk

single use

Required verification

check consumed state

Failure response

reject; reconcile

Residual limitation

state loss

Negative test

second use refused

16.26. T26 Substitute-agent credential inheritance

Attacker objective

inherit old credentials

Capability

new agent

Trust boundary

substitute to resource

Invariant at risk

no inheritance

Required verification

fresh authorization required

Failure response

deny

Residual limitation

weak revocation

Negative test

substitute uses old capability; denied

16.27. T27 Pending-work escrow substitution

Attacker objective

swap escrowed work

Capability

controls escrow store

Trust boundary

escrow commit

Invariant at risk

escrow canonical and committed

Required verification

verify escrow digest or record

Failure response

reject

Residual limitation

store compromise

Negative test

altered escrow; refused

16.28. T28 False simulation-as-completion assertion

Attacker objective

claim simulation as done

Capability

controls reporter

Trust boundary

simulation to record

Invariant at risk

simulation not completion

Required verification

require native receipt

Failure response

reject completion

Residual limitation

fidelity gap

Negative test

simulated success; no completion

16.29. T29 Missing phase receipt

Attacker objective

skip a phase

Capability

controls coordinator

Trust boundary

phase gate

Invariant at risk

phase needs receipt

Required verification

check receipt

Failure response

block advancement

Residual limitation

receipt delay

Negative test

advance without receipt; refused

16.30. T30 Network partition and split-brain coordination

Attacker objective

two coordinators

Capability

partition

Trust boundary

coordinator to resource

Invariant at risk

one valid generation

Required verification

monotonic fencing

Failure response

fence older

Residual limitation

partition detection limits

Negative test

two leaders; stale one rejected

16.31. T31 Unknown native effect incorrectly retried

Attacker objective

duplicate effect

Capability

lost ack

Trust boundary

retry path

Invariant at risk

unknown not retried automatically

Required verification

status inquiry first

Failure response

hold retry

Residual limitation

status unavailable

Negative test

lost ack; no blind retry

16.32. T32 Stale leader or supervisor during replacement

Attacker objective

old leader acts

Capability

old host

Trust boundary

successor activation

Invariant at risk

generation-fenced

Required verification

check generation

Failure response

reject

Residual limitation

delayed fence

Negative test

stale leader acts after replacement; refused

16.33. T33 Revocation propagation delays

Attacker objective

act before revocation arrives

Capability

disconnected domain

Trust boundary

revocation to domain

Invariant at risk

no claim of instant revocation

Required verification

bounded freshness

Failure response

fail closed after bound

Residual limitation

irreducible delay

Negative test

disconnected domain; authority expires by lease

16.34. T34 Membership-vote inflation

Attacker objective

gain votes

Capability

many agents

Trust boundary

membership voting

Invariant at risk

one organization one vote

Required verification

count organizations

Failure response

count once

Residual limitation

org identity

Negative test

ten agents one org; one vote

16.35. T35 Mandatory veto bypass

Attacker objective

override an owner

Capability

majority

Trust boundary

veto

Invariant at risk

non-delegable veto

Required verification

check veto records

Failure response

block

Residual limitation

owner ambiguity

Negative test

majority over veto; blocked

16.36. T36 Incompatible cross-domain policies

Attacker objective

merge unlike policies

Capability

policy mismatch

Trust boundary

policy compatibility

Invariant at risk

semantic compatibility verified

Required verification

compare semantics

Failure response

refuse

Residual limitation

semantic ambiguity

Negative test

same word different meaning; refused

16.37. T37 Leaderless causal-dependency spoofing

Attacker objective

fake predecessor

Capability

forged receipt

Trust boundary

owner-side check

Invariant at risk

predecessor verified at owner

Required verification

verify receipt origin

Failure response

refuse

Residual limitation

origin compromise

Negative test

forged predecessor; refused

16.38. T38 Missing native predecessor effects

Attacker objective

proceed without predecessor

Capability

coordinator

Trust boundary

owner gate

Invariant at risk

manifest enforced at owner

Required verification

check manifest

Failure response

block

Residual limitation

manifest gaps

Negative test

missing predecessor; blocked

16.39. T39 Selective evidence-disclosure abuse

Attacker objective

extract extra data

Capability

evaluator

Trust boundary

disclosure gate

Invariant at risk

purpose and audience bound

Required verification

check purpose

Failure response

deny

Residual limitation

inference

Negative test

request outside purpose; denied

16.40. T40 Metadata leakage

Attacker objective

infer from metadata

Capability

observer

Trust boundary

evidence exchange

Invariant at risk

metadata minimized

Required verification

review metadata

Failure response

minimize

Residual limitation

residual inference

Negative test

metadata in logs; flagged

16.41. T41 Forged evidence-inclusion proofs

Attacker objective

fake inclusion

Capability

forger

Trust boundary

proof verification

Invariant at risk

proof verified against source

Required verification

verify against commitment

Failure response

reject

Residual limitation

algorithm weakness

Negative test

forged proof; rejected

16.42. T42 False complete-session closure

Attacker objective

claim closed

Capability

coordinator

Trust boundary

closure

Invariant at risk

all domains sealed

Required verification

check seals

Failure response

withhold closure

Residual limitation

missing domain

Negative test

DONE with open obligation; refused

16.43. T43 Residual effect-capable grants after task closure

Attacker objective

reuse grants

Capability

holds old grant

Trust boundary

grant lifecycle

Invariant at risk

grants revoked or exhausted

Required verification

check grants

Failure response

revoke

Residual limitation

untracked grant

Negative test

grant used after closure; refused

16.44. T44 Incorrect global atomicity assumptions

Attacker objective

assume atomic

Capability

designer

Trust boundary

composite plan

Invariant at risk

no global atomicity

Required verification

handle partial results

Failure response

report partial

Residual limitation

design error

Negative test

one domain commits; partial shown

16.45. T45 Partial native commitment

Attacker objective

hide partial

Capability

coordinator

Trust boundary

composite status

Invariant at risk

partial effect reported

Required verification

per-domain evidence

Failure response

PARTIAL_EFFECT

Residual limitation

unobserved domain

Negative test

one commit missing; shown partial

16.46. T46 Compensation without owner consent

Attacker objective

reverse an effect unilaterally

Capability

coordinator

Trust boundary

compensation

Invariant at risk

compensation is a new authorized act

Required verification

check owner authorization

Failure response

refuse

Residual limitation

urgent pressure

Negative test

unauthorized compensation; refused

16.47. T47 Irreversible effect treated as rolled back

Attacker objective

claim rollback

Capability

coordinator

Trust boundary

history

Invariant at risk

irreversible effects stay in record

Required verification

reversibility classes

Failure response

mark irreversible

Residual limitation

classification error

Negative test

irreversible shown as rolled back; rejected

16.48. T48 Byzantine or dishonest resource-origin reports

Attacker objective

lie at source

Capability

owner source

Trust boundary

source honesty

Invariant at risk

limits of source honesty stated

Required verification

cross-source checks

Failure response

UNKNOWN or DISPUTED

Residual limitation

source dishonesty

Negative test

contradicting sources; DISPUTED

16.49. T49 Protected-local controller compromise

Attacker objective

subvert local state

Capability

local access

Trust boundary

controller

Invariant at risk

local trust stated

Required verification

detect inconsistency

Failure response

fail closed

Residual limitation

trusted base

Negative test

tampered local state; detected where possible

16.50. T50 Cryptographic algorithm or key compromise

Attacker objective

forge signatures

Capability

breaks keys

Trust boundary

signature layer

Invariant at risk

algorithm agility optional

Required verification

rotate and revoke

Failure response

revoke

Residual limitation

unknown break

Negative test

revoked key signature; rejected

16.51. T51 Denial of service against mandatory evaluators

Attacker objective

stall decisions

Capability

floods evaluators

Trust boundary

availability

Invariant at risk

unavailability never grants permission

Required verification

timeouts

Failure response

HOLD

Residual limitation

delay

Negative test

evaluator down; HOLD not approve

16.52. T52 Excessive reviewer or evidence requirements causing deadlock

Attacker objective

block progress

Capability

designs impossible rules

Trust boundary

policy

Invariant at risk

rules satisfiable or escalate

Required verification

deadlock detection

Failure response

escalate to human

Residual limitation

policy design

Negative test

unsatisfiable quorum; escalates

16.53. T53 Resource-owner ambiguity

Attacker objective

unclear owner

Capability

misconfigured

Trust boundary

owner resolution

Invariant at risk

actual owners resolved

Required verification

resolve ownership

Failure response

HOLD

Residual limitation

registry error

Negative test

two claimed owners; HOLD

16.54. T54 Time-source compromise

Attacker objective

extend authority

Capability

controls clock

Trust boundary

time

Invariant at risk

expiry uses protected time or events

Required verification

protected time source

Failure response

fail closed

Residual limitation

time dependency

Negative test

clock rewind; authority not extended

16.55. T55 Replay after checkpoint restoration

Attacker objective

reuse consumed state

Capability

restores checkpoint

Trust boundary

state restoration

Invariant at risk

consumption survives restore

Required verification

checkpoint lineage

Failure response

reject

Residual limitation

backup practices

Negative test

restored checkpoint reuses grant; refused

16.56. T56 Cross-jurisdiction evidence disclosure

Attacker objective

unlawful transfer

Capability

evaluator in other region

Trust boundary

evidence export

Invariant at risk

jurisdiction rules checked

Required verification

check constraints

Failure response

deny

Residual limitation

legal ambiguity

Negative test

evidence to forbidden region; denied

16.57. T57 Faulty recovery after a delayed valid receipt

Attacker objective

double effect

Capability

late receipt

Trust boundary

recovery

Invariant at risk

late receipt reconciled

Required verification

re-inquire

Failure response

reconcile

Residual limitation

timing

Negative test

receipt arrives after recovery; reconciled not repeated

16.58. T58 Privacy-violating diagnostic logs

Attacker objective

leak via logs

Capability

reads logs

Trust boundary

logging

Invariant at risk

logs minimized

Required verification

review logs

Failure response

redact

Residual limitation

residual info

Negative test

log with full evidence; flagged

16.59. T59 Incorrect use of a positive evaluator determination as an execution token

Attacker objective

treat approval as authority

Capability

integrator

Trust boundary

determination to gate

Invariant at risk

determination non-executable

Required verification

gate ignores determinations

Failure response

reject

Residual limitation

integration error

Negative test

gate receives determination alone; refuses

16.60. T60 Uncovered alternate native effectuation paths

Attacker objective

bypass the gate

Capability

finds other path

Trust boundary

coverage

Invariant at risk

claims limited to covered paths

Required verification

coverage inventory

Failure response

state limitation

Residual limitation

unknown paths

Negative test

effect via uncovered path; limitation stated

17. Interoperability Requirements

No new wire protocol is proposed where existing standards suffice. Candidate carriers, with architecture requirements kept separate from candidate schemas and normative requirements:

Table 4: Interoperability candidates
Mechanism Possible use Note
HTTP and TLS transport of requests and determinations does not supply appointment semantics
OAuth 2.0 and token exchange delegation of bounded evaluator access token is not an effect authorization
JWT and CWT with COSE/JOSE signed appointment and determination artifacts signature is not outcome truth
Workload identity evaluator and supervisor identity identity is not independence
Remote attestation (RATS) host and instance binding attests state, not truthfulness of evidence
Distributed tracing evidence lineage correlation traces are claims, not receipts
Transparency logs audit continuity inclusion is not truth
Message queues and agent protocols carrying requests delivery is not admission

18. Industry Profiles

Hypothetical and non-endorsed. These descriptions identify potential research contexts only and were not verified against vendor documentation by the editor; readers must verify against public product documentation. Established product functionality is separated from proposed integration, and the alignment is AI-generated and must be cross-checked against the original filings. It is not an endorsement, validation, adoption claim or infringement statement.

Frontier AI platforms

Established: tool-call approval and human review steps exist in various platforms. Proposed: a gate placed at the tool-execution broker would reject an act whose appointment, round or determination is stale, and would require a separate owner-issued authorization before the call is released. Why it matters: model-approved text then cannot by itself cause an effect.

Enterprise and cloud

Established: policy decision and enforcement points, workload identity, audit logs. Proposed: an evaluator service whose appointment is derived from resource-owner consent and whose determinations are inputs to, not substitutes for, an existing PEP check at the resource API.

Financial

Established: dual control, maker-checker, reconciliation. Proposed: a reviewer appointed per transaction class, with minimal-disclosure outcome evidence, UNKNOWN on lost acknowledgments and compensation as a separately authorized act (Examples F, I, J, K).

Telecom

Established: orchestration and slice policies. Proposed: cross-domain evaluator for multi-operator changes, with owner-side predecessor checks and partial-outcome manifests when one domain does not commit.

Industrial

Established: safety interlocks and change management. Proposed: a persistent evaluator serving many maintenance projects with per-project mandates (Example B), with actuation only through the owner's existing controlled interface.

19. Worked Examples A-L

Each example gives initial authority, protected state, eligible evidence, determination, separate native authority and failure outcomes.

19.1. Example A: Two-company procurement

Initial authority: each company owns its purchasing and payment authority. Protected state: a committed, non-effective joint plan, frozen. Eligible evidence: inventory and price receipts from each owner. Determination: APPROVED_FOR_CONSIDERATION by an evaluator appointed by both. Native authority: each company's gate separately admits purchase and payment. Failures: stale price evidence yields HOLD; missing consent yields refusal; unknown payment outcome yields UNKNOWN.

19.2. Example B: Persistent industrial evaluator

Initial authority: one service, many projects, separate mandates. Protected state: per-project namespaces, rounds and validity. Evidence: sensor logs and work-order receipts. Determination: per-project, valid for a bounded interval. Native authority: plant controls admit changes. Failures: project A expiry gives no authority to B; platform compromise suspends all contexts.

19.3. Example C: Cross-domain evaluator committee

Initial authority: three owners set reviewer qualifications. Protected state: round with control-group map. Evidence: provenance-diverse reports. Determination: closed only with mandatory reviewer present and dissent preserved. Native authority: owners authorize separately. Failures: three accounts in one group count once; suppressed dissent rejects the round.

19.4. Example D: Communication freeze race

Initial authority: plan committed. Protected state: freeze epoch. Evidence: pending-work inventory. Determination: HOLD while a late message is unaccounted; snapshot invalidated. Native authority: unchanged, none granted. Failures: partial acknowledgment is not a freeze.

19.5. Example E: Failed agent substitution

Initial authority: old agent's consumed capability. Protected state: escrowed task data, source fenced. Evidence: neutrality screening. Determination: SUBSTITUTE recommended. Native authority: the substitute has no valid permission to continue consuming the old capability. Failures: credential inheritance denied.

19.6. Example F: Network partition

Initial authority: one authorized payment. Protected state: operation fingerprint and idempotency key. Evidence: none received. Determination: UNKNOWN. Native authority: retry needs new authorization. Failures: the coordinator must not repeat; a read-only status inquiry runs first.

19.7. Example G: Multi-organization membership change

Initial authority: epoch e with two owners. Protected state: electorate snapshot. Evidence: joiner consent; unresolved earlier operation. Determination: HOLD until the earlier operation settles. Native authority: local installation required. Failures: disconnected owner's veto not bypassed; activation withheld.

19.8. Example H: Leaderless cooperation

Initial authority: two independent owners. Protected state: predecessor manifest at each owner. Evidence: native receipt of operation one at owner two. Determination: convergence evidence only. Native authority: owner two admits operation two only if the receipt verifies. Failures: spoofed predecessor refused.

19.9. Example I: Minimal-disclosure outcome verification

Initial authority: bank controls records. Protected state: attribute commitments. Evidence: selective attributes for a stated purpose and audience. Determination: verified outcome attribute. Native authority: none gained. Failures: out-of-purpose request refused; metadata minimized.

19.10. Example J: Partial financial and inventory completion

Initial authority: two domain owners. Protected state: PREPARE records. Evidence: ledger committed, inventory not released. Determination: PARTIAL_EFFECT. Native authority: each domain separately. Failures: no global atomicity assumed; completion record not executable.

19.11. Example K: Reversible and irreversible compensation

Initial authority: a committed operation. Protected state: history record. Evidence: owner consent. Determination: compensation proposed. Native authority: compensation is a new native effect with its own receipt. Failures: irreversible effects cannot be recorded as rolled back.

19.12. Example L: Protected session termination

Initial authority: task apparently ended. Protected state: obligation inventory. Evidence: queued work and old grants found. Determination: closure withheld. Native authority: grants revoked or exhausted, then seals. Failures: DONE message alone is rejected.

20. Performance and Legacy Compatibility

The document makes no universal latency claim. Three paths are distinguished: a fast path (valid cached determination within validity), an evaluation path (new review), and a recovery path (partition, UNKNOWN, revocation). Legacy systems may adopt the architecture by placing the gate at an existing enforcement point, with the evaluator offline or batched.

Measurement plan: p50, p95, p99 and p99.9 latency; throughput; setup cost per appointment; reviewer count effect; revocation propagation time; partition recovery time; false acceptance and false denial rates; UNKNOWN frequency; and privacy leakage. No figures are asserted here. Source-reported test counts (see Section 11) come from local Python reference code only and are not performance or production evidence.

21. Prior Art and IETF Overlap

Each topic below is established prior art or practice in its own right. The applicant does not claim that any isolated primitive, signature, quorum, AI evaluator or distributed-transaction mechanism was invented by the applicant. The architecture's distinction, if any, lies in the particular cooperation of protected appointment, role independence, lifecycle separation, evidence eligibility, bounded determination, owner-specific authorization and native-effect admission. That statement is not a legal conclusion.

Related individual drafts by the same author: [REL-FOUND], [REL-CBEA], [REL-CAUSAL], [REL-IEV] (identifiers as supplied by the author; verify on the Datatracker). Others' work on agent delegation, handoff and requirements ([I-D.agentic-ai-usecases-requirements]) is acknowledged; the editor has not claimed a full survey.

21.2. Non-Normative Patent-Examination Methodology

  • Identify specific claim limitations and read them as written.
  • Identify the closest prior-art combinations.
  • State the objective technical problem.
  • Assess functional interdependence of the elements.
  • Ask whether there was a motivation to combine, and a reasonable expectation of success.
  • Distinguish what a skilled person could combine from what the prior art would have motivated them to combine.
  • Avoid hindsight.
  • Distinguish new terminology from actual technical differences.

This is not a legal conclusion on patentability.

22. Security Considerations

The architecture reduces the chance that an evaluation result is mistaken for authority, but it depends on correct enforcement at participating boundaries. Non-bypassable native-effect control applies only to actual participating protected enforcement paths. See Section 16 for 60 threat categories. Residual limits: a compromised trusted base, a dishonest resource owner, undetected correlated sources, algorithm compromise, and uncovered alternate effect paths. No implementation is assumed to guarantee globally atomic transactions, instantaneous revocation during disconnection, physically infallible source evidence, zero-latency evaluation or universal interoperability.

23. Privacy Considerations

Evidence is disclosed only under authorized purpose and audience constraints and with minimal retention. Metadata, logs and traces can leak information and are minimized. Cross-jurisdiction evidence transfer follows applicable constraints; the document gives no legal advice.

24. IANA Considerations

This document requests no IANA actions.

25. Notices

Human-Originated Invention

The underlying technical invention and patent disclosure are attributed to human inventor Sangam Kumar Das. AI-assisted preparation of the draft, terminology mapping, diagrams, industry comparisons or explanatory text does not independently establish AI inventorship.

AI-Assisted Research and Verification

AI-assisted material can contain mistakes, omissions, inaccurate product references or misinterpretations of technical claims. All assertions must be checked against the original filed patent documents, verified RFCs, primary standards documentation and official industry publications.

IETF Status

The document is an individual technical proposal and does not constitute an endorsed IETF standard, approved RFC, adopted working-group specification or IETF consensus position. No IETF-assigned identifier is claimed.

Industry Neutrality

References to OpenAI, Anthropic, Google, Microsoft, NVIDIA, AWS, Alibaba, Huawei, SAP, IBM, Nokia, Ericsson, financial companies or other organizations identify potential research contexts only. No adoption, endorsement, affiliation, copying, infringement or commercial implementation is asserted.

Technical Limits

A positive evaluator determination does not independently establish a real-world effect or confer authority for such an effect. Non-bypassable native-effect control applies only to actual participating protected enforcement paths.

IPR Requirements

The author must comply with applicable IETF BCP 78 and BCP 79 requirements. Mentioning the patent inside a draft does not replace any required formal IETF IPR disclosure. No FRAND, royalty-free or other licensing commitment is made or implied here.

26. Intellectual Property and Research Lineage

Current source disclosure: inventor Sangam Kumar Das; Indian Provisional Patent Application 202631131650; applicant-reported filing date 3 November 2026 (the source PDF is named with a 'NOV 4' label but contains a digital-signature date of 3 November; no 4 November filing is inferred). Supplied source: 1,905 pages, 27 embodiments, 540 numbered claims.

Lineage: PCT/IB2026/055615 filed 4 June 2026, published 16 July 2026 as WO 2026/150382 A2. Related Indian provisionals reported by the applicant: 202631134241 (8 November 2026), 202631136555 (15 November 2026). Later features are not assigned the June priority date automatically; entitlement to any priority date is a legal question outside this document.

Public research records: https://doi.org/10.5281/zenodo.24274259 and https://doi.org/10.5281/zenodo.24076231; Technical Disclosure Commons records 12542, 12934 and 12973. These were not all fetched by the editor; verify. An IETF IPR disclosure (reference 7610) exists for earlier drafts of the author; whether it covers this draft is not established. IPR will be disclosed at the appropriate time as per IETF guidelines.

27. Normative References

[RFC2119]
Bradner, S., "Key words for use in RFCs to Indicate Requirement Levels", RFC 2119, BCP 14, , <https://www.rfc-editor.org/info/rfc2119>.
[RFC8174]
Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 2119 Key Words", RFC 8174, BCP 14, , <https://www.rfc-editor.org/info/rfc8174>.

28. Informative References

[I-D.agentic-ai-usecases-requirements]
Reddy, T., Sarker, Z., and K. Yao, "Agentic AI Use Cases and Requirements", Work in Progress, Internet-Draft, draft-agentic-ai-usecases-requirements, , <https://datatracker.ietf.org/doc/draft-agentic-ai-usecases-requirements/>.
[IN-131650]
Das, S., "Indian Provisional Patent Application 202631131650", Date as reported by the applicant; not a certified registry record., , <https://ipindia.gov.in/>.
[PCT-055615]
Das, S., "PCT/IB2026/055615 (WO 2026/150382 A2)", Filed 4 June 2026; published 16 July 2026. Verify on PATENTSCOPE., , <https://patentscope.wipo.int/>.
[REL-CAUSAL]
Das, S., "Causal Diagnostics for Semantic Drift in Autonomous Agent Communications", Identifier as supplied by the author; verify on the Datatracker. Not fetched., Work in Progress, Internet-Draft, draft-das-causal-semantic-handoff-diagnostics, , <https://datatracker.ietf.org/>.
[REL-CBEA]
Das, S., "Consequence-Bound Execution Authority", Identifier as supplied by the author; verify on the Datatracker. Not fetched., Work in Progress, Internet-Draft, draft-das-consequence-bound-execution-authority, , <https://datatracker.ietf.org/>.
[REL-FOUND]
Das, S., "Execution Finality Foundations", Identifier as supplied by the author; verify on the Datatracker. Not fetched., Work in Progress, Internet-Draft, draft-das-execution-finality-foundations, , <https://datatracker.ietf.org/>.
[REL-IEV]
Das, S., "Receipt-Gated Independent Effect Verification (title as supplied; verify)", Identifier as supplied by the author; verify on the Datatracker. Not fetched., Work in Progress, Internet-Draft, draft-das-receipt-gated-iev, , <https://datatracker.ietf.org/>.
[TDC-12542]
Das, S., "Technical Disclosure Commons record 12542 (PCT file; title not verified)", Not fetched; verify., , <https://www.tdcommons.org/dpubs_series/12542/>.

Appendix A. Appendix A: E01-E27 Profiles and Claim Crosswalk

The complete profiles for E01-E27 are in Section 12 and Section 13; the crosswalk of all 60 independent claims (mapped to the embodiment of each) and the indexed dependent claims 61-540 is in Section 11. Independent claim titles are listed in each profile's Claim references field.

Appendix B. Appendix B: Diagrams and State Machines

27 diagrams appear in the body; their captions are listed here for reference.

Appointment state machine: proposed -> owner-approved -> active -> (expired | revoked); evaluation round: open -> admitting -> (closed | disputed); determination: valid -> (stale | withdrawn). Transitions to active require all mandatory owner consents; any lifecycle can end without ending the others (see Section 6).

Appendix C. Appendix C: Message and Evidence Examples

Illustrative records are in Section 14. Example evidence chain: a resource-origin receipt (EvidenceRecord, class native_receipt) is admitted into a closed EvaluationRound, which yields a non-executable EvaluatorDetermination; owner policy separately issues a bounded authorization; the native gate admits or withholds the act and emits a NativeOutcomeReceipt.

Appendix D. Appendix D: Negative Security Tests

The 60 negative-test categories, each with a concrete test, are in Section 16 (T01-T60). Test results reported in the source are local reference-code tests and are not reproduced as evidence of security.

Appendix E. Appendix E: Examiner Keyword and Phrase Index

Counts: 212 keywords and 115 phrases. The index serves searches on WIPO PATENTSCOPE, EPO Espacenet, Google Patents, IEEE, ACM, arXiv, the IETF Datatracker and general technical literature. Presence of a term does not imply novelty or infringement. Cross-references give the embodiment (E) and, through Section 11, the independent claims.

E.1. Evaluator architecture

  • Location-Independent Evaluator (E04,E17,E21)
  • Neutral Evaluation Service (E04,E21)
  • Independent Evaluator (E04,E17)
  • Evaluation Function (E04)
  • Policy-Bounded Evaluator (E04,E06)
  • Trusted Local Evaluator (E04,E17)
  • Human-Assisted Evaluator (E04,E17)
  • Protected Decision Service (E04)
  • Single-Component Evaluator (E04)
  • Evaluator Role (E04)
  • Evaluator Software Identity (E17)
  • Decision Service Isolation (E04,E05)
  • Neutral Evaluator (E04,E13)
  • Review Function (E17)
  • Consequence Evaluator (E04)
  • Non-Effective Decision (E04)
  • Typed Determination (E04,E21)
  • Recommendation Versus Permission (E04,E07)
  • Evaluation Mandate (E04,E24)
  • Evaluator Roster (E21)

E.2. Evaluator authority

  • Protected Evaluator Appointment (E21,E24)
  • Task-Bound Mandate (E20,E21)
  • Appointment Completeness (E20,E21)
  • Authority Source Resolution (E20,E21)
  • Evaluator Role Qualification (E21)
  • Owner-Approved Review (E21)
  • Mandate Revocation (E24)
  • Scope-Limited Determination (E04,E21)
  • Appointment Expiry (E24)
  • Use-Count Limit (E24)
  • Authority Lease (E24)
  • Authority Generation (E24)
  • Owner-Derived Authority (E20,E21)
  • Mandatory Owner Resolution (E20)
  • Self-Appointment Prevention (E20)
  • Joint Supervisor Appointment (E20)
  • Supervisor Host Binding (E20)
  • Per-Call Authorization (E20)
  • Owner Withdrawal (E20,E24)
  • Non-Delegable Condition (E22)

E.3. Lifecycle

  • Ephemeral Evaluator (E21,E24)
  • Persistent Evaluator (E21,E24)
  • Hybrid Evaluator (E21,E24)
  • Evaluation Instance Lifetime (E24)
  • Appointment Lifetime (E24)
  • Review-Round Lifetime (E17,E21)
  • Determination Validity (E17,E24)
  • Evidence Retention (E19,E26)
  • Cross-Project Authority Isolation (E21,E24)
  • Runtime Retirement (E19,E24)
  • Stale Authority Reuse Prohibition (E24)
  • Context Migration (E14,E24)
  • Protected Continuation (E01,E16)
  • Round Closure (E17)
  • Closure Tombstone (E19)
  • Authority Retirement (E19)
  • Non-Reopenable Session (E19)
  • Determination Staleness (E15,E24)
  • Appointment Revocation (E24)
  • Platform Compromise Handling (E24)

E.4. Governance

  • Multi-Organization Governance (E20,E22,E25)
  • Mandatory Approver (E20)
  • Principal-Level Quorum (E25)
  • Independent Control Group (E17,E21)
  • Mandatory Veto (E06,E25)
  • Reviewer Independence (E17,E21)
  • Collusion-Resistant Review (E17)
  • Membership Epoch (E25)
  • Organizational Consent (E20,E25)
  • Threshold Governance (E25)
  • Eligible Electorate Snapshot (E25)
  • Joiner Consent (E25)
  • Old-Epoch Fencing (E25)
  • Membership Reconstitution (E25)
  • Principal-Independent Voting (E25)
  • Credential-Count Inflation (E25)
  • Reviewer Qualification (E17,E21)
  • Mandatory Reviewer Completeness (E17)
  • Dissent Preservation (E17,E21)
  • Dispute Containment (E21)

E.5. Evidence

  • Evidence Admission (E04,E17)
  • Source Role (E04,E16)
  • Evidence Provenance (E16,E17)
  • Source Authentication (E16,E26)
  • Outcome Truth (E15,E16)
  • Native Effect Receipt (E15,E27)
  • Correlated Evidence (E17)
  • Source Diversity (E17,E21)
  • Evidence Commitment (E26)
  • Selective Disclosure (E26)
  • Evidence Lineage (E26)
  • Freshness (E04,E15)
  • Source-Origin Receipt (E15,E27)
  • Resource-Origin Evidence (E15,E16)
  • Evidence Custodian (E26)
  • Evidence Audience (E26)
  • Attribute Commitment (E26)
  • Inclusion Proof (E26)
  • Contradictory Evidence (E17)
  • Unresolved Evidence (E17,E21)

E.6. Agent workflow

  • Multi-Agent Negotiation (E01)
  • Joint Plan Commitment (E01)
  • Communication Freeze (E02)
  • Distributed Quiescence (E02)
  • Protected Next Actor (E07)
  • Single-Use Next-Move Passport (E11)
  • Renegotiation (E08)
  • Task-State Migration (E14)
  • Pending-Work Escrow (E14)
  • Plan Revision (E08)
  • Corrective Patch (E08)
  • Role Reconfiguration (E09)
  • Bounded Delegation (E09)
  • Capability Transition (E09)
  • Substitute Agent (E13)
  • Single-Turn Authority (E10)
  • Automatic Refreezing (E10)
  • Inter-Phase Hold (E10)
  • Concurrent Move Arbitration (E07)
  • Causal Message Containment (E03)

E.7. Recovery

  • Unknown Outcome (E11,E18)
  • Distributed Partition (E18)
  • Durable Fencing (E18)
  • Split-Brain Prevention (E18)
  • Migration Epoch (E18)
  • Partial Completion (E27)
  • Authoritative Reconciliation (E18,E27)
  • Protected Compensation (E27)
  • Non-Reopenable Session Closure (E19)
  • Idempotency Key (E18)
  • Native-Effect Fingerprint (E18)
  • Stale Leader Fencing (E18)
  • Checkpoint Lineage (E18)
  • Lost Acknowledgment (E18)
  • Source-Status Inquiry (E18)
  • Retry Authorization (E18)
  • Recovery Authorization (E18)
  • PARTIAL_EFFECT (E27)
  • Reversible Versus Irreversible Effect (E27)
  • Successor Activation (E24)

E.8. Native execution

  • Effectuation Boundary (E12)
  • Finality Sink (E12)
  • Native Resource Admission (E12,E27)
  • Protected Local Commit (E27)
  • Owner-Controlled Effect (E12)
  • PREPARE/COMMIT (E27)
  • Predecessor Manifest (E23,E27)
  • Consequence-Bound Authorization (E12)
  • Pending Act Reconstruction (E12)
  • Alternate-Path Closure (E12)
  • External-Dispatch Outbox (E12)
  • Exact-State Authorization (E12)
  • One-Use Permission Consumption (E11,E27)
  • Composite Completion Record (E27)
  • Completeness Roster (E27)
  • Native Predecessor Verification (E27)
  • Resource-Side Admission (E12)
  • Point of Effect (E12)
  • Owner-Specific Commit (E27)
  • Non-Executable Completion (E27)

E.9. Privacy and cross-domain policy

  • Purpose-Bound Evidence (E26)
  • Audience-Bound Verification (E26)
  • Minimal Disclosure (E26)
  • Trust Compatibility (E22)
  • Policy-Semantic Compatibility (E22)
  • Federated Governance (E22)
  • Sovereign Resource Authority (E22)
  • Constraint Intersection (E22)
  • Policy Generation (E22)
  • Drift Containment (E22)
  • Metadata Leakage (E26)
  • Disclosure Audit (E26)
  • Evidence Redaction (E26)
  • Source Sequence Anchor (E26)
  • Joint Authority Formation (E22)
  • Leaderless Coordination (E23)
  • Causal Dependency Evidence (E23)
  • Remote-Status Lease (E23)
  • Owner-Side Admission (E23)
  • Local Supervision (E23)

E.10. Risk and simulation

  • Virtual Consequence Reconstruction (E05)
  • Non-Effective Simulation (E05)
  • Digital Twin Evaluation (E05)
  • Tail Exposure (E06)
  • Benefit-Loss Evaluation (E06)
  • Cumulative Exposure (E06)
  • Domain Threshold (E06)
  • Calibration Reassessment (E06)
  • Uncertainty-Aware Evaluation (E05)
  • Counterfactual Operational State (E05)
  • Risk-Constrained Actor Selection (E04,E07)
  • Adverse-Outcome Feedback (E06)

E.11. Industry terminology

  • Policy Decision Point (E04,E12)
  • Policy Enforcement Point (E12)
  • Workflow Orchestration (E15)
  • Saga Transaction (E27)
  • Two-Phase Commit (E27)
  • Fencing Token (E18)
  • Byzantine Fault Tolerance (E17,E25)
  • Remote Attestation (E20)
  • Workload Identity (E20)
  • Human-in-the-Loop Approval (E04,E17)
  • Zero Trust (E12)
  • Quorum Approval (E25)
  • Audit Log (E26)
  • Distributed Tracing (E03)
  • Agent-to-Agent Protocol (E01)
  • Model Context Protocol (E12)
  • Token Exchange (E11)
  • Proof-of-Possession Token (E11)
  • Lease (E24)
  • Idempotent Operation (E18)

E.12. Search Phrases

Boolean phrases (quoted terms in practice); each lists the embodiments.

  1. location independent evaluator AND task appointment (E04,E21)
  2. ephemeral evaluator AND protected mandate (E21,E24)
  3. persistent evaluator AND project scoped authority (E21,E24)
  4. hybrid evaluator AND task-specific context (E21,E24)
  5. evaluator determination AND execution authorization (E04,E12)
  6. evaluator decision validity AND appointment lifetime (E17,E24)
  7. neutral evaluator AND administrative independence (E04,E21)
  8. single component evaluator AND independence (E04,E17)
  9. multi agent negotiation AND protected freeze (E01,E02)
  10. joint plan AND non-effective AND evaluator (E01,E04)
  11. distributed quiescence AND evaluation snapshot (E02,E03)
  12. mandatory reviewer AND dissent AND authorization (E17,E21)
  13. independent control groups AND threshold governance (E17,E25)
  14. cross organization supervisor AND appointment (E20)
  15. agent substitution AND pending work escrow (E13,E14)
  16. source authenticated outcome AND independent evaluator (E16,E21)
  17. unknown outcome AND durable fencing (E18)
  18. leaderless coordination AND native effect (E23)
  19. cross domain policy semantic compatibility (E22)
  20. resource owner AND predecessor receipt (E23,E27)
  21. partial effect AND authorized compensation (E27)
  22. non executable determination AND native authorization (E04,E12)
  23. purpose bound selective disclosure AND outcome evidence (E26)
  24. protected session closure AND domain seal (E19)
  25. membership reconstitution AND owner veto (E25)
  26. evaluation authority AND effectuation authority (E04,E12)
  27. evaluator AND appointment AND revocation AND lease (E24)
  28. review round AND closure AND dissent (E17)
  29. evidence admission AND source role AND task binding (E04,E17)
  30. correlated evidence AND source independence (E17)
  31. agent self-appointment AND evaluator (E20)
  32. mandatory owner AND consent AND supervisor (E20)
  33. host instance binding AND supervisor installation (E20)
  34. next-move passport AND single use AND actor bound (E11)
  35. bearer token AND non-bearer AND continuation (E11)
  36. single turn authority AND refreezing (E10)
  37. selective communication resumption AND hold (E10)
  38. inter-phase hold AND evaluator (E10)
  39. corrective patch AND committed plan AND recommitment (E08)
  40. role reconfiguration AND capability generation (E09)
  41. neutral substitute AND credential inheritance (E13)
  42. pending work escrow AND source fencing (E14)
  43. destination acceptance AND task migration (E14)
  44. completion receipt AND phase advancement (E15)
  45. actual outcome AND feedback AND continuation (E16)
  46. untrusted feedback AND instruction containment (E16)
  47. lost acknowledgment AND resource authoritative status (E18)
  48. stale leader AND resource side fencing (E18)
  49. checkpoint lineage AND migrated agent (E18)
  50. session termination AND native obligation inventory (E19)
  51. non-reopenable tombstone AND session (E19)
  52. independent outcome evaluation AND cross organization (E21)
  53. dispute appeal AND new round isolation (E21)
  54. policy drift AND owner drift AND containment (E22)
  55. constraint intersection AND joint authority (E22)
  56. causal receipt AND remote status lease (E23)
  57. predecessor manifest AND owner installed (E23,E27)
  58. supervisor lease AND revocation convergence (E24)
  59. successor activation AND partition (E24)
  60. monotonic revocation serial AND tombstone (E24)
  61. principal independent voting AND agent count (E25)
  62. membership epoch AND old epoch fencing (E25)
  63. selective evidence commitment AND source authenticated (E26)
  64. audience bound release AND confidential outcome (E26)
  65. minimal retention AND audit continuity (E26)
  66. prepare commit AND native domain gate (E27)
  67. composite completion AND no composite authority (E27)
  68. compensation AND new native effect (E27)
  69. partial composite effect AND controlled resume (E27)
  70. virtual consequence AND non-effective scenario (E05)
  71. simulation AND not proof of execution (E05)
  72. tail exposure AND domain threshold AND veto (E06)
  73. risk benefit loss AND organization policy (E06)
  74. next actor selection AND bounded instruction (E07)
  75. concurrent next moves AND arbitration (E07)
  76. communication edge AND pending work AND freeze (E02)
  77. late operation disposition AND admission fencing (E02)
  78. causal message containment AND snapshot validity (E03)
  79. negotiation state integrity AND evaluator (E03)
  80. finality sink AND evaluator determination (E12)
  81. alternate path closure AND external dispatch outbox (E12)
  82. pending act reconstruction AND exact state (E12)
  83. evaluator integrity AND provenance AND review (E17)
  84. reviewer qualification AND control group (E17,E21)
  85. evidence retention AND determination validity (E19,E24)
  86. cross project isolation AND evaluator (E21,E24)
  87. platform compromise AND evaluator contexts (E24)
  88. context migration AND revalidation (E14,E24)
  89. agent communication AND evaluation freeze AND quiescent (E02)
  90. policy semantic mismatch AND refusal (E22)
  91. non-delegable condition AND joint authority (E22)
  92. owner local admission AND convergence evidence (E23)
  93. dependency spoofing AND resource owner (E23)
  94. supervisor revocation AND disconnected domain (E24)
  95. bounded freshness AND revocation (E24)
  96. eligible electorate AND snapshot (E25)
  97. joiner consent AND mandatory veto (E25)
  98. metadata leakage AND evidence exchange (E26)
  99. inclusion proof AND source commitment (E26)
  100. irreversible effect AND compensation record (E27)
  101. native predecessor effect AND manifest (E27)
  102. UNKNOWN AND not retried AND status inquiry (E18,E27)
  103. partial effect AND honest reconciliation (E27)
  104. HOLD AND UNKNOWN AND DISPUTED AND evaluator (E04,E17,E21)
  105. agent AND tool AND evaluator AND approval AND gate (E04,E12)
  106. multi-agent AND cross-organizational AND evaluator (E01,E21)
  107. agent protocol AND evaluator appointment (E20,E21)
  108. policy decision point AND evaluator lifecycle (E04,E24)
  109. workflow engine AND review round AND closure (E15,E17)
  110. saga AND compensation AND owner authorization (E27)
  111. fencing token AND supervisor generation (E18,E24)
  112. quorum AND organization AND agent count (E25)
  113. human reviewer AND committee AND evaluator (E04,E17)
  114. agent evaluation framework AND appointment (E04,E21)
  115. audit log AND selective disclosure AND audience (E26)

E.13. Boolean Search Strings

  • ("evaluator" AND "appointment") AND ("task" OR "mandate") AND ("agent")
  • ("determination" AND "not executable") AND ("native" OR "resource owner")
  • ("review round") AND ("dissent") AND ("control group" OR "independent")
  • ("freeze" AND "quiescence") AND ("agent" OR "agents") AND ("evaluation")
  • ("unknown outcome" OR "lost acknowledgment") AND ("fencing") AND ("recovery")
  • ("supervisor" AND "lease") AND ("revocation") AND ("successor")
  • ("prepare" AND "commit") AND ("native domain") AND ("partial effect")
  • ("selective disclosure") AND ("purpose") AND ("audience") AND ("outcome evidence")

Appendix F. Appendix F: Industry Integration Mapping

See Section 18. The mapping is hypothetical, non-endorsed, AI-generated and unverified against vendor documentation; cross-check against the original filings before drawing any conclusion. It is not an endorsement, validation or infringement statement.

Gate placement

Where a tool broker, API gateway, payment rail adapter, network controller or actuator interface already enforces access, a gate that additionally requires a current owner-issued authorization could sit there.

What is checked

appointment, round, determination validity, policy generation, actor, operation and single-use state.

What is rejected

stale, replayed, wrong-actor or determination-only requests.

Why it matters

a model or reviewer approval cannot itself cause an effect.

Appendix G. Appendix G: Patent and Research Cross-References

See Section 26, [IN-131650], [PCT-055615], [TDC-12542]. Related drafts: [REL-FOUND], [REL-CBEA], [REL-CAUSAL], [REL-IEV]. This document is distinguishable from them: causal diagnosis locates drift; execution finality governs admission of an effect; this document governs evaluator appointment, evidence eligibility, lifecycles, typed determinations and cross-domain evaluation governance, and uses E12, E15, E18 and E27 only in relation to those existing drafts.

Appendix H. Appendix H: Source Verification and Limitations

H.1. Source Verification

  • Source: 1,905-page PDF; page and embodiment map built from the extracted text; independent claims 1-60 and dependent ranges checked for gaps.
  • Filing date: the screenshot of the IP India eFiling paid-forms page shows reference 202631131650, C.B.R. No. 16058, CBR date 03/11/2026. This corroborates the applicant-reported date but is a payment record, not the official acknowledgment, which the editor has not seen.
  • The source file name says 'NOV 4'; it contains a digital-signature date of 3 November; no 4 November filing is inferred.
  • The Technical_Disclosure_1.json file was not available to the editor.

H.2. Quality Checklist

  1. Actual source is the 1,905-page evaluator specification: partly verified (page count and embodiment map checked; read selectively).
  2. All 27 embodiments represented: yes.
  3. 60 independent claims mapped: yes, titles checked against the filing.
  4. Dependent-claim index covers 61-540: yes, by embodiment range.
  5. Three deployment types preserved: yes.
  6. Six lifecycles explained: yes.
  7. Location independence not confused with control independence: yes.
  8. Single-component evaluators not excluded: yes.
  9. Neutrality not assumed from third-party hosting: yes.
  10. Owner-derived appointment vs self-appointment: yes.
  11. Vetoes, quorum, independence policy-specific: yes.
  12. Control groups not account counts: yes.
  13. Mandatory dissent preserved: yes.
  14. Origin vs outcome truth: yes.
  15. Evaluator decision vs owner authorization: yes.
  16. Native admission independently enforced: yes.
  17. Freeze accounts for in-flight operations: yes.
  18. Simulation not proof of execution: yes.
  19. Replacement agents inherit no stale authority: yes.
  20. Unknown outcomes not retried: yes.
  21. Source-side fencing and revocation modeled: yes.
  22. Membership epoch/owner conditions: yes.
  23. Leaderless cooperation not global authority: yes.
  24. Partial effects honest: yes.
  25. Compensation separately authorized: yes.
  26. Closure needs native obligation reconciliation: yes.
  27. Privacy-sensitive evidence purpose/audience bound: yes.
  28. Cryptographic vs protected-local distinguished: yes.
  29. Post-quantum optional: yes.
  30. IETF and industry references verified: NO, partly unverified (see Appendix H).
  31. Existing RFC functions and prior art credited: yes.
  32. No invented performance figures: yes.
  33. No company adoption inferred: yes.
  34. IETF IPR requirements addressed: partly; formal disclosure pending.
  35. Source and filing metadata reconciled: partly; acknowledgment not seen.
  36. Every diagram technically meaningful: reviewed by editor.
  37. Security, Privacy, IANA, References complete: yes.
  38. Search terms cover inventor and industry language: yes.
  39. Distinguishable from existing DAS drafts: yes.
  40. Understandable without patent documentation: intended; reader feedback needed.

Author's Address

Sangam Kumar Das
Independent Inventor
Balasore 756001
Odisha
India