VSTD 1.2.0 release-candidate maintainer review

[TimeLordRaps]

Coordinate

• Primary release pull request: Prepare VSTD 1.2.0 and restore the public boundary #20
• Stacked completion pull request: Prepare the complete VSTD 1.2.0 release candidate #27
• Cumulative candidate head: edf591b40242a74e4b893e7938a0eb60a572c818
• Base: main@598c545be3833d6d81bb7e252ca5837f3bb2a449
• Package coordinate: 1.2.0 release candidate
• Scope: maintainer review only; this issue does not authorize merge, tag, or publication

Review seams

1. Verifier Standard (VSTD) claim/evidence binding and preservation of UNKNOWN, CONFLICTED, and NOT_ESTABLISHED.
2. Actor independence: distinct actors and execution seams must be evidence-bound, not inferred from serialized fields, references, repetition, or placement.
3. VSTD-4 and VSTD-Graph candidate/conformance ceilings, including VSTD-5 rejection of unearned readiness.
4. Frozen generic-run compatibility and the legacy, non-dispatching meaning of layer4_binding.
5. Supply Chain Integrity, Transparency, and Trust (SCITT) interoperability as adjacent signature/registration evidence rather than payload truth.
6. Governing zero-identity/zero-knowledge (ZIZK) artifact-first architecture; process-bound TRUST, ROT, and RUST; cryptographic enclosure for confidential witnesses; and the bounded RISC Zero and identity-disclosure mechanisms.
7. AGENTS.md, HUMANS.md, and TIME.md as non-overlapping maintainer controls.
8. Public presentation, GitHub Pages, security analysis, packaging, and exact-head release gates.

Required interpretation

• Identity or reputation alone cannot strengthen an artifact-bound result; checked identity may establish only the exact attribution, authorization, or separation proposition earned by its mechanism.
• Zero identity means zero identity-derived verdict weight, not anonymity. Architectural zero knowledge presumes no unevidenced proposition; when a witness is confidential, cryptographic zero knowledge may enclose the exact program, predicate, commitments, output, parameters, and verifier only through a named proof system.
• TRUST is mechanism-earned artifact support moving forward across checked transformations. ROT is typed time-indexed degradation of current admissibility without rewriting historical receipts. RUST is the inverse-TRUST memetic causal backtrace over recorded ancestor states; diagnostic reachability alone does not establish falsehood, guilt, responsibility, intervention-level causal localization, or automatic ancestor falsification. None is actor-tied trust or a scalar.
• Historical receipts remain immutable and readable. Current admissibility may change without rewriting historical status.
• Green repository checks are evidence about their named checks, not VSTD conformance, external adoption, or maintainer approval.

File-to-seam review matrix

Passing automation supplies evidence for the named checks below; it does not record the
maintainer disposition. Review each row against the pull-request diff and mark the
corresponding checklist only after the invariant and falsification boundary are acceptable.

Seam	Primary review surfaces	Reject the candidate if	Current evidence	Disposition
1. Claim/evidence binding	standard/LADDER.md; docs/CLAIMS_AND_LIMITS.md; src/verifier/core/checker.py; src/verifier/core/run.py; src/verifier/core/run_validation.py; tests/test_assurance_flow_invariants.py; tests/test_generic_run.py	A declaration, digest, field name, or missing mechanism strengthens a claim.	Full suite, generic-run adversarial tests, presentation gate	PENDING MAINTAINER REVIEW
2. Actor independence	src/verifier/core/run.py; src/verifier/core/run_reproduction.py; src/verifier/core/run_validation.py; generic-run schemas and examples; tests/test_generic_run.py; tests/test_assurance_flow_invariants.py	References, repeated runs, matching results, or placement emerge as evidence-bound distinct actors or execution seams.	Independence falsification and same-result rerun tests	PENDING MAINTAINER REVIEW
3. VSTD-4 and Graph ceilings	standard/LADDER.md; src/verifier/core/depth.py; src/verifier/data/graph_level.py; tests/test_vstd4_depth.py; tests/test_graph_level.py	Caller-supplied ratings or structural candidates become established conformance, or VSTD-5 accepts unearned readiness.	Depth, duplicate-path, cycle, conflict, and candidate-label tests	PENDING MAINTAINER REVIEW
4. Historical layer4_binding	standard/WIRE_IDENTIFIERS.md; src/verifier/core/run.py; src/verifier/core/run_support.py; src/verifier/core/run_validation.py; generic-run receipt schema; tests/test_generic_run.py; tests/test_vstd3_backward_compatibility.py	Historical bytes or reads break, the name drives VSTD-4 dispatch, or current writing silently changes frozen meaning.	Compatibility fixtures, strict-profile validation, packaged specifications	PENDING MAINTAINER REVIEW
5. SCITT interoperability	docs/standards/VSTD_SCITT_CROSSWALK.md; src/verifier/interoperability/scitt/adapter.py; SCITT examples; tests/test_scitt_interop.py; tests/test_scitt_crypto_example.py	Signature or registration success manufactures payload truth, VSTD correctness, or actor independence.	Required real COSE path plus adversarial composition tests	PENDING MAINTAINER REVIEW
6. ZIZK TRUST/ROT/RUST architecture	README.md; docs/ARCHITECTURE.md; standard/LADDER.md; examples/zizk_artifact_first/; experiments/artifact_first_mechanisms/; tests/test_zizk_artifact_first.py	Governing architecture is buried as a side experiment; actor identity becomes TRUST; ROT rewrites historical truth; RUST loses its memetic backtrace semantics; or diagnostic ancestry becomes guilt, falsehood, or causal localization.	Tracked proof artifacts, manifest binding, supported-Linux offline verification, ZIZK tests	PENDING MAINTAINER REVIEW
7. Maintainer controls	AGENTS.md; HUMANS.md; TIME.md; scripts/check_time_status.py; tests/test_presentation_surface.py	Agent rules, human reasoning, and live repository contradictions overlap, or publication proceeds with live TIME contradictions.	TIME.md clear, release-status and presentation tests	PENDING MAINTAINER REVIEW
8. Release and public presentation	.github/workflows/; scripts/release_artifacts.py; scripts/check_release_boundary.py; docs/API_STABILITY.md; docs/; scripts/build_reference.py; tests/test_public_api.py; tests/test_release_artifacts.py; tests/test_presentation_surface.py	Repository checks are described as VSTD conformance, exact artifacts are not reproducible, private boundaries leak, or first-use/public claims exceed implementation.	Exact-head local build, 476-member boundary scan, deterministic CycloneDX SBOM, installed-wheel smoke, branch-coverage artifact, hosted run and CodeQL	PENDING MAINTAINER REVIEW

Review order for the stacked candidate

1. Review pull request Prepare the complete VSTD 1.2.0 release candidate #27 first: 60 files, 2,530 additions, and 1,232 deletions relative to Prepare VSTD 1.2.0 and restore the public boundary #20.
2. Review pull request Prepare VSTD 1.2.0 and restore the public boundary #20 for the underlying release semantics: 275 files, 25,486 additions, and 5,222 deletions relative to main.
3. Do not record final disposition against Prepare VSTD 1.2.0 and restore the public boundary #20 until Prepare the complete VSTD 1.2.0 release candidate #27 has landed into its branch and the resulting Prepare VSTD 1.2.0 and restore the public boundary #20 head has rerun all protected checks.
4. Treat the cumulative comparison main...codex/post-1.2-professionalization as the current complete candidate: 290 files, 27,127 additions, and 5,567 deletions.

A four-hour pass can be a risk-prioritized maintainer disposition over the eight seams; it is not an exhaustive line-by-line review of approximately 32,000 changed lines.

Exact-head evidence

• Cumulative head has a good GNU Privacy Guard signature from key F5537E7240663768250B315091A5B5158391B78C.
• Main-branch protection requires signed commits; all 48 cumulative candidate commits are GitHub-verified with reason valid.
• Local full suite: 405 passed.
• At the cumulative head, current reports and examples name VSTD-1, VSTD-2, and VSTD-Graph-1; removed 0.x specification filenames are absent from every live GitHub branch, while issued 0.x wire identifiers and tagged release bytes remain frozen historical compatibility evidence.
• The v0.1.0 and v0.2.0 GitHub release pages carry additive historical-coordinate notices directing readers to the current identifier map; their tags and attached artifact bytes were not changed.
• On 2026-08-27, in the documented Ubuntu 24.04 Linux environment under Windows Subsystem for Linux 2, direct network-offline RISC Zero 3.0.6 verification at the cumulative head accepted the exact tracked receipt and public envelope against the pinned image identifier without the private witness.
• Artifact-first manifest and repository-bound artifacts verify at sha256:756425630fcb8552cd0892c1ba64a3150fbb3c021b0d930b9cf4f5e86bf6b4fe.
• Exact-commit package build, manifest verification, CycloneDX 1.6 Software Bill of Materials validation, Twine validation, installed-wheel smoke, Pages, presentation, and release-boundary checks pass; the boundary scan covered 476 text members across 5 artifacts.
• All 15 workflow jobs passed on the cumulative head in run 33090567593; the pull request reports 16 checks including the adjacent GitHub CodeQL result.
• CodeQL passed on the cumulative pull-request merge ref; GitHub currently reports 0 open code-scanning, Dependabot, and secret-scanning alerts.
• GitHub reports pull requests Prepare VSTD 1.2.0 and restore the public boundary #20 and Prepare the complete VSTD 1.2.0 release candidate #27 as clean and mergeable; Prepare VSTD 1.2.0 and restore the public boundary #20 does not yet contain Prepare the complete VSTD 1.2.0 release candidate #27, and no review approval is currently recorded.
• TIME.md is Status: CLEAR.
• Version 1.2.0 remains intentionally UNRELEASED; publication requires a later explicit metadata finalization and tag decision.

Falsification condition

Request changes if declarations, names, hashes, repetition, graph multiplicity, actor reputation, placement, SCITT registration, or satisfiability can increase assurance without a named mechanism validating the exact proposition and evidence binding; if mechanism separation becomes actor independence; if recorded ancestry becomes causal proof; if UNKNOWN or CONFLICTED is cleaned; if a candidate becomes conformance; or if any exact-head evidence above fails.

Maintainer disposition

• Select an exact private conduct-reporting route and replace the nonexistent
  "private GitHub channel" instruction in CODE_OF_CONDUCT.md; do not repurpose public
  issues or silently treat security advisories as the general conduct mechanism.
• Decide whether normative releases require an approving reviewer independent of the
  pull-request author, then align branch protection and governance text with that actual
  control.
• Review the eight seams above and the corresponding pull-request diff.
• Record APPROVED here when the normative, compatibility, security, and release boundaries are acceptable, or record exact requested changes.
• Merge only after that disposition is recorded.

[TimeLordRaps]

Review reopened: ZIZK architecture and proof-artifact boundary

Do not approve or merge pull request #20 yet. Signed head 842c26e177b677e8556be8b492386cd7d1ceac37 records two new TIME.md contradictions:

1. The normative artifact-first/zero-actor-trust orientation in standard/LADDER.md is presented in the README, roadmap, and experiment index as if ZIZK itself were an optional side experiment. Experimental status belongs only to unfinished transfer/event/localization mechanics and optional proof implementations.
2. The recorded real RISC Zero proof receipt and public envelope are absent from the repository. Their hashes are published and the exact files still exist in the isolated local proof worktree, but a public consumer cannot retrieve and verify that recorded proof. The private witness remains correctly excluded.

Required before disposition: separate governing architecture from experimental mechanisms in the public information hierarchy, publish and bind the exact non-secret proof artifacts or narrow the recorded claim, rerun verification/presentation/release gates, and return TIME.md to Status: CLEAR through a signed resolution commit.

[TimeLordRaps]

Prior ZIZK blocker resolution at predecessor candidate head

This comment preserves the evidence that resolved two earlier blockers at signed predecessor
candidate 039ba96517ebebe733e3f728aa52989233f8b010:

1. Zero-identity/zero-knowledge (ZIZK) artifact-first TRUST was presented as governing
   architecture. Experimental status was limited to unfinished serialization, TRUST
   transfer, RUST localization, trichotomy, and optional-backend mechanisms.
2. The exact non-secret RISC Zero receipt, public envelope, and self-test result were
   tracked, digest-bound by the artifact-first manifest, and accepted by the documented
   supported-Linux network-offline verifier path. At that predecessor head, the manifest
   digest was sha256:20a9060d8244ed1af59d0ea2e058dc412b08c5b4851c4abcbf7994059e2b093f.

At that predecessor head, RUST was stated as a memetic causal backtrace over recorded
ancestor states while remaining diagnostic reachability rather than guilt, responsibility,
automatic ancestor falsification, or intervention-level causal localization.

The predecessor exact-head hosted run
33036499955
passed all 14 workflow jobs; CodeQL reported 0 results across 50 rules; TIME.md was
Status: CLEAR.

This comment is predecessor-review context, not current-candidate evidence. The issue body
now records the cumulative candidate at
edf591b40242a74e4b893e7938a0eb60a572c818, where pull request #27 formally defines
TRUST, ROT, and RUST and distinguishes architectural zero knowledge from its optional
cryptographic zero-knowledge enclosure.

This resolved those two recorded blockers only. It did not record maintainer approval,
a merge decision, release authorization, external review, or publication. The review
matrix and remaining human decisions in the issue body still govern disposition.

[TimeLordRaps]

VSTD 1.2.0 has now been published from the reviewed release sequence: #27 supplied the cumulative professionalization tree, #28 finalized the actual publication metadata, and #29 recovered the immutable release publication. The versioned coordinate is https://github.com/TimeLordRaps/verifier/releases/tag/v1.2.0.

This release-candidate review issue is therefore superseded and is being closed. Closure records lifecycle state only: it does not retroactively convert every unchecked item in this historical candidate checklist into an independently established approval. Any post-release defect or falsification should be filed against the exact released or current commit using the dedicated issue forms.