{ "schema": "a2atraffic-gap-matrix-v1", "_comment": "The governance gap matrix this site publishes. Scores are the PAPER'S, re-read from the HTML on verified_at — not paraphrased from a summary. A previous summarising pass over this same paper produced a plausible-sounding taxonomy that was not the paper's, and it was published before being caught (ProjectAmp2/PORTFOLIO_AUDIT_REVIEW.md carries the retraction). Do not restate a score without re-opening Table III.", "verified_at": "2026-08-17", "paper": { "title": "Governance Gaps in Agent Interoperability Protocols: What MCP, A2A, and ACP Cannot Express", "authors": [ "Richard Kang", "Yudho Diponegoro" ], "submitted": "2026-06-30", "arxiv": "2606.31498", "url": "https://arxiv.org/abs/2606.31498", "html": "https://arxiv.org/html/2606.31498v1", "status": "preprint", "status_note": "arXiv preprint. NOT peer-reviewed. This site called it peer-reviewed nowhere, and must not.", "headline": "Voting and dissent preservation are universally absent across all five protocols.", "conclusion": "The authors conclude agent community governance is a missing architectural layer, not a missing feature of existing standards." }, "dimensions": [ { "id": "G1", "name": "Membership", "definition": "Protocol encodes admission, invitation, removal, and role assignment for community participants.", "a2a": "partial", "a2a_transports": "Agent Cards declare capabilities and can be registered in directories.", "a2a_cannot": "There is no protocol-native admission, invitation, or removal primitive. A card says what an agent claims it can do; nothing in A2A says whether it was allowed to join, or how it leaves.", "here": { "what": "box-and-box deontic rung (rung 3 of eight) and the delegatic OS-006 governance shim.", "rung": "in_tree", "detail": "The kernel answers 'is this permitted' with a certificate attached, over a floor that cannot be weakened. 118 enforced property-tested laws and 3 declared-open." }, "crosswalk": { "cells": [ "14", "15", "16" ], "claim": "Membership is admission, removal and role assignment. The authority graph answers all three structurally: delegation cannot cycle, a child cannot exceed its parent, and absence of a grant is refusal. That is a membership rule a protocol could carry but does not." } }, { "id": "G2", "name": "Deliberation", "definition": "Protocol encodes structured argument exchange with turn-taking, challenge, and response semantics.", "a2a": "absent", "a2a_transports": "Messages and tasks. An argument can be sent as text.", "a2a_cannot": "Nothing distinguishes an argument from a message, a challenge from a reply, or a turn from a retry. Structure has to be re-invented above the protocol by every party, which means no two parties agree on it.", "here": { "what": "deliberatic — Dung's argumentation framework extended to weighted bipolar (wBAF), graded semantics.", "rung": "spec", "detail": "Specified, no implementation. Its own page says so." }, "crosswalk": { "cells": [ "18", "20" ], "claim": "Both cells are missing. The table names convergence and seal integrity as the measurements a deliberation would need; neither exists. This dimension is as empty here as it is on the wire." } }, { "id": "G3", "name": "Voting", "definition": "Protocol encodes preference aggregation with quorum, rounds, and position resolution.", "a2a": "absent", "a2a_transports": "Nothing. A vote is an ordinary message.", "a2a_cannot": "No quorum, no round, no resolution rule. Two agents cannot disagree in a way the protocol can settle.", "here": { "what": "deliberatic argues explicitly against majority voting and specifies graded semantics instead; agentromatic covers bidding, debate and election.", "rung": "spec", "detail": "Specified, no implementation." }, "crosswalk": { "cells": [], "claim": "There is no cell for voting anywhere in the table. deliberatic argues against majority voting and specifies graded semantics instead, at spec rung — so the absence may be deliberate rather than an oversight, but an absent cell is an absent cell." } }, { "id": "G4", "name": "Dissent preservation", "definition": "Protocol ensures minority positions are retained in decision outputs, not silently dropped.", "a2a": "absent", "a2a_transports": "Nothing. A losing position is a message nobody reads again.", "a2a_cannot": "A decision transported over A2A carries its result and not its objections. The disagreement that produced it is unrecoverable from the wire.", "here": { "what": "deliberatic makes attacks first-class in the formal tuple (R-), and specifies that three vindicated dissents trigger constitutional review.", "rung": "spec", "detail": "Specified, no implementation." }, "crosswalk": { "cells": [ "19" ], "claim": "Dissent durability is the single cell in the table that names this, and it is sketched. It is also the strongest thing either census has to say about a dimension the paper finds absent in all five protocols." } }, { "id": "G5", "name": "Human escalation", "definition": "Protocol defines conditions and mechanisms for routing decisions to human authority.", "a2a": "absent", "a2a_transports": "Push notifications, and TASK_STATE_INPUT_REQUIRED / TASK_STATE_AUTH_REQUIRED as interrupted states.", "a2a_cannot": "Those are transport and status. Neither says under what condition a human MUST be reached, who counts as the authority, or what happens if nobody answers. An agent can ask; nothing obliges the escalation.", "here": { "what": "The RRABBIT mailbox. A message is 4 lanes x <=32 bits, so it carries an escalation signal and never prose. Mailbox is the only role with no ports, so 'beside the road' is forced by the algebra rather than chosen. Unread is a projection, never a stored count, because otherwise reading a message mutates the world.", "rung": "live_local", "detail": "node test/mail.mjs -> 68 passed, 0 failed (run 2026-08-17)." }, "crosswalk": { "cells": [ "24" ], "claim": "Escalation monotonicity is missing, and it is filed under Tier / Budget (OS-005, model tier adaptation) — escalating to a bigger model, not to a person. Read strictly, the table has no human-escalation cell at all; the working artifact is the RRABBIT mailbox, which is live_local and not in this table." } }, { "id": "G6", "name": "Audit and replay", "definition": "Protocol produces tamper-evident event logs enabling deterministic reconstruction of the decision process.", "a2a": "absent", "a2a_transports": "A Traceability extension exists.", "a2a_cannot": "The paper scores this absent despite that extension, because it does not define tamper-evident logs or replay semantics. You can observe the traffic; you cannot prove the log was not edited, and you cannot re-run it.", "here": { "what": "REPLAY — typed refusals, no silent repair. A step whose road is gone is refused by name, never remapped.", "rung": "live_local", "detail": "Five typed refusal codes: TRACK_EMPTY, TRACK_ROAD_GONE, TRACK_ROAD_CLOSED, TRACK_WINDOW_GONE, TRACK_BUSY (RRABBIT/docs/TRACKS_HANDOFF.md:135). An internal review said six; the file says five, and the file wins." }, "crosswalk": { "cells": [ "17", "33" ], "claim": "Append-only audit is proved: history cannot be rewritten, which is half of what the dimension asks. The other half — deterministic reconstruction — maps to non-replay in Security & Federation, and that is missing. So this dimension is genuinely split, and the page should not round it up." } } ], "protocols": [ { "id": "mcp", "label": "MCP v1.1", "scores": { "G1": "absent", "G2": "absent", "G3": "absent", "G4": "absent", "G5": "absent", "G6": "partial" }, "coverage": "1/12" }, { "id": "a2a", "label": "A2A v1.0.1", "scores": { "G1": "partial", "G2": "absent", "G3": "absent", "G4": "absent", "G5": "absent", "G6": "absent" }, "coverage": "1/12", "focus": true }, { "id": "acp", "label": "ACP", "scores": { "G1": "partial", "G2": "partial", "G3": "absent", "G4": "absent", "G5": "absent", "G6": "absent" }, "coverage": "2/12" }, { "id": "anp", "label": "ANP", "scores": { "G1": "absent", "G2": "absent", "G3": "absent", "G4": "absent", "G5": "absent", "G6": "absent" }, "coverage": "0/12" }, { "id": "erc8004", "label": "ERC-8004", "scores": { "G1": "partial", "G2": "absent", "G3": "absent", "G4": "absent", "G5": "absent", "G6": "partial" }, "coverage": "2/12" } ], "scoring_note": "Two points per dimension, twelve available. full = 2, partial = 1, absent = 0. Best coverage of any protocol assessed is 2/12.", "_version_note": "The paper assesses A2A v1.0.1. The current release is 1.0 per the spec site; 1.0.1 is a patch within that line. Nothing in the v1.0 changes re-derived in protocol.json adds a governance primitive, so the scores are not stale on that account — but they are the paper's reading of one version, and this site does not claim to have re-scored the protocol itself.", "version_drift": { "_comment": "Stated on the page. A matrix that quietly scores superseded versions is the same defect as a stale metric, and MCP shipped a major revision 28 days after this paper was submitted.", "note": "This paper was submitted 2026-06-30. MCP shipped its largest revision since launch on 2026-07-28 — the version string is literally \"2026-07-28\"; the C# SDK calls its implementation v2.0 and much of the press calls it MCP 2. The paper therefore scores MCP v1.1, which is superseded.", "rescored": false, "why_the_row_still_stands": "Read against the 2026-07-28 release notes, that revision is about transport and deployment shape, not community governance: a stateless request/response core, an Extensions framework, Tasks moved out to the io.modelcontextprotocol/tasks extension, MCP Apps, Enterprise Managed Authorization, Multi Round-Trip Requests, and authorization hardening (RFC 9207 issuer validation, Dynamic Client Registration replaced by Client ID Metadata Documents). Roots, Sampling, Logging and the legacy HTTP+SSE transport are deprecated on a twelve-month window. Nothing in it adds membership, deliberation, voting, dissent preservation, human escalation or tamper-evident replay. So the G1-G6 row is very unlikely to have moved — but this site has NOT re-scored it, and 'unlikely to have moved' is not a measurement.", "source": "https://blog.modelcontextprotocol.io/posts/2026-07-28/" }, "crosswalk_note": { "_comment": "Printed on the page verbatim. The mapping below is the only thing on this site that is neither a measurement nor a quotation.", "kind": "judgement", "whose": "Claude (Opus 5), session 2026-08-17, reviewing both sources side by side.", "statement": "Which invariant answers which governance dimension is a reading, not a measurement. Nobody has published a mapping between these two vocabularies, the two were built independently and for different purposes, and no test enforces the correspondence. DOCTRINE.md rule 1: where a condition cannot be settled by a command, say that it is a judgement and whose. This is one, and it is mine.", "what_is_not_a_judgement": "The A2A scores are the paper's. The invariant statuses are extracted from the canonical table by tools/sync-invariants.mjs and re-derived on every build. Only the arrows between them are argued." } }