Research

Research

The research behind the protocol. Open access, Apache 2.0, cite freely. Full bibliography on ORCID.

Papers

9 papers, each with a DOI. Pick a square.
  • Authority and delegation The Agent Social Contract, Monotonic Narrowing, Faceted Authority Attenuation, Physics-Enforced Delegation
  • Governance and accountability Governance in the Medium, Cognitive Attestation, Behavioral Derivation Rights
  • Evidence and experiments The Evidence-Safety Gap, Plausibly Wrong
  1. The foundational paper. Defines the contract between agents, owners and observers, and why governance is a substrate problem rather than a policy problem.

    DOI 10.5281/zenodo.18749779
  2. The invariant that makes delegation chains safe. Authority can narrow as it descends, never widen. Mirrored at SSRN abstract 6415678.

    DOI 10.5281/zenodo.18932404
  3. Multi-facet authority decay across delegation hops, with closed-form attenuation per facet.

    DOI 10.5281/zenodo.19260073
  4. Cryptographic primitives that make scope violation physically impossible rather than merely policy-prohibited.

    DOI 10.5281/zenodo.19478584
  5. Governance lives in the substrate that carries actions, not in the policies layered on top. This is why the gateway exists.

    DOI 10.5281/zenodo.19582550
  6. Signed declarations of feature-level model computation. Accountability for AI decisions without exposing model internals.

    DOI 10.5281/zenodo.19646276
  7. How an agent earns the right to act based on observed behaviour. Derivation, not declaration.

    DOI 10.5281/zenodo.19476002
  8. Why standard ML safety arguments do not transfer to agentic systems, and what evidentiary structures are needed instead.

    DOI 10.5281/zenodo.19914628
  9. A controlled isolation of the retention rule in short-lived LLM agent populations. Peer-voted retention kept a worse shared memory than random selection in 4 of 5 replicates. Pre-registered, with kill conditions and an artifact record.

    DOI 10.5281/zenodo.21208555

Cited and referenced in independent work

25 works by other authors.
  • IETF Internet-Draft
  • arXiv
  • Zenodo
  • Other
  1. The rttp and iqa URI Schemes (S. Li, IETF Internet-Draft, October 2026): cites draft-pidlisnyi-aps-04 as a normative reference. draft-li-rttp-iqa-addressing

  2. Agent Delegation Receipts (R. Nelson, IETF Internet-Draft, October 2026): references Action Receipts v1.1 and disclaims verified wire-format interoperability with it. draft-nelson-agent-delegation-receipts

  3. Action Reference (P. Etcheverry and K. Ives, IETF Internet-Draft, October 2026): lists the APS SDK's external action-ref helper as an implementation, covering that helper only. draft-etcheverry-action-ref

  4. MUSUBI Settlement (T. Oga, IETF Internet-Draft, October 2026): lists the APS settle core checker as an implementation by others, not counted as independent. draft-oga-musubi-settlement

  5. Canonical Payload Binding (S. Mih and A. Sokolov, IETF Internet-Draft): credits Tymofii Pidlisnyi (Agent Passport System) as a contributor for the action reference pattern. draft-mih-sokolov-scitt-payload-binding

  6. Agent Execution Evidence: A SCITT Profile (S. Thaha and S. Shamsudeen, IETF Internet-Draft, October 2026): cites draft-pidlisnyi-aps-04 and says the profile could serve as its evidence layer. draft-thaha-scitt-agent-execution-evidence

  7. Authoritative Resolution of Persistent Digital Agent Identities (E. Tănase, AIDAXIA, IETF Internet-Draft): cites draft-pidlisnyi-aps-04 in its sections on delegator authority and unresolved verification, and in related work. draft-tanase-ain-authoritative-resolution

  8. An External Verifier Contract for Agent Authorization Decisions (V. P. Kondoju, Bolyra, IETF Internet-Draft, September 2026): cites draft-pidlisnyi-aps-03 and describes wrapping an APS decision engine as a verifier. draft-kondoju-evc

  9. Witness Answers (T. Oga, IETF Internet-Draft, November 2026): cites the APS draft for stating what a verified record does and does not show. By a collaborator, not counted as independent. draft-oga-audit-witness-answer

  10. Notarized Agents (J. Figuera, arXiv:2606.04193): references APS.

  11. The Civilization Framework (G. Liu, arXiv:2609.03425): cites draft-pidlisnyi-aps-03 for authority attenuation.

  12. Revocation Is Not Complete at Receipt (L. Liu, H. Yu and D. Zhang, arXiv:2610.15991): cites draft-pidlisnyi-aps-02 in related work.

  13. VERA: Authority-Preserving Edge Revocation (L. Liu, H. Yu and X. Jiang, arXiv:2608.30091): uses APS as a baseline and reports that its subtree cascade over-revokes agents with a second authorizing parent.

  14. Authorization Propagation in Agent2Agent Systems (G. Donno, K. Mehmood and K. Kralevska, arXiv:2611.07050): compares APS with other multi-hop authorization approaches, citing draft-pidlisnyi-aps-03.

  15. Composable Trust Infrastructure for the Agent Internet (F. Moore, 10.5281/zenodo.19263547): names APS delegation governance as one of four layers.

  16. Verifiable Agent Trust Envelope (Poke-nushi, 10.5281/zenodo.19839769): lists APS among its closest adjacent work, with a comparison document.

  17. The Governability Gap in Agentic AI (M. Agnihotri, 10.5281/zenodo.21376745): cites the APS draft for its reversibility scheme.

  18. PDR in Production, section 7.6 (10.5281/zenodo.19323172): references the APS reputation model and verdict separation.

  19. Agent Authority Manifest (AAM) (G. Arora, Institute for Technology Stewardship, public consultation draft, November 2026, 10.5281/zenodo.24104750): cites draft-pidlisnyi-aps-04 as the closest related work and says AAM should reuse its restriction mechanics.

  20. The Linked-Witness Mechanism (P. Le Fol du Taillis, SPOC, preprint, September 2026): cites APS among the passport proposals whose signed envelope pattern it follows. spoc-linked-witness-mechanism.pdf

  21. Emilia Protocol interoperability survey: records draft-pidlisnyi-aps-03 as one of two of 25 drafts complete on fields. emilia-protocol

  22. Íñigo Medina (Instituto Tramontana, April 2026): catalogued "Governance in the Medium" with a summary and critique.

  23. Who Authorized This? (S. Shilina, Humanode, October 2026): names the Agent Passport System draft of 28 September (draft-pidlisnyi-aps-04) as treating agent identity, principal, delegated authority, policy approval, execution and observed result as distinct things. blog.humanode.io

  24. The Missing Layer: Why Your CBOM Doesn't Know About Your AI Agents (All Things on Security, August 2026): names the Agent Passport System among the agent identity systems that sign with Ed25519, in an argument that agent signing keys belong in a cryptographic bill of materials for post-quantum migration. allthingsonsecurity.com

  25. SI Passport (E. D. H. Wang, SI Organization, Technical Disclosure Commons, October 2026): a defensive publication that lists the Agent Passport System Internet-Draft among known prior art, with its delegation chains, narrowing scopes, cascading revocation and signed receipts. tdcommons.org

Federal submissions

Filed as public comment. Filing is not endorsement.
  • NIST CAISI

    Comment filed with the NIST Center for AI Standards and Innovation. Acknowledged by Drew Keller.

  • NIST AI 800-2

    Comment on the framework, covering agent identity and delegation primitives.

  • NCCoE

    Concept paper submitted to the National Cybersecurity Center of Excellence.