Passa al contenuto principale

Current → 1.0 Gap Analysis

Research-project contract reconciliation — 2026-11-15​

Reconcile the prototype org.omi.research-project/1 writer with the canonical omi-research-project/0.1.0 schema. Cover source-preserving legacy conversion, unknown-resource retention, open-versus-add identity, project-scoped cache reload/undo, and failed-write binding safety. Direct cloud transfer, provider integrity/conflict receipts, archive resolution and public projection remain open gaps.

Status: proposed coordinated repair pending companion merges/CI; deployment and provider acceptance remain separate. No manuscript/container generation change or database migration.

Contract, compatibility and current implementation.

Text-box and research provenance — 2026-11-13​

Close the Studio-only schema divergence and unversioned text-box payload gap through canonical Draft definitions, an independent schema-generated profile validator, immutable source pin and shared golden. Close diagram provenance bypass through exact preview/confirmation, source/bibliography/derivation lifecycle and internal own-work declarations. This does not close full ContentCodec, export-fidelity or real release-acceptance gaps.

Status: proposed coordinated repair; merge/CI and deployment acceptance remain separate. No database migration, container-generation change or new review authority.

Profile, decision and acceptance scope.

Privacy/API boundary repair — 2026-11-08​

New unversioned privacy/account-copy routes and direct application→Prisma coupling were found in Studio a829afba6eb6e9a68da34a9bc21d0315101c7536; OMI 2491bc0f54b0a02cea6bcf5d1dcb98b969d7ac18 documented those URLs. Keep the features; apply targeted containment.

BoundaryProposed repairRequired evidence
Public transportAdditive v1, common legacy handlers/quota, deprecation signals, strict DTO/OpenAPI/error schemasDual-mount isolation, errors, quota bypass negatives and generated-spec parity
Identity persistenceInject account-data repository; keep explicit selected fields and optional-role fallback in adaptersSecret rejection, truncation, absent account, DB failure and login/projection regression
Authority and export scopeOperator-reviewed notice separate from identity/review authority; copy explicitly partialAnonymous notice, private download, selected native origin; deployment/operator acceptance

npm run test:privacy remains mandatory CI evidence under the existing release gates. Mocked persistence is not production/PostgreSQL or legal acceptance. This repair is proposed until merged and verified; other 1.0 blockers remain open.

Studio ADR-034; Studio PR #886.

PKP support and release evidence — 2026-11-07​

OJS 3.5.0-4/-5/-6 are in the Stable acceptance scope; OMP 3.5.0-4/-5/-6 remain Preview. All six named E02 jobs/archives are mandatory regression evidence. Add ojs-3.5.0-6 to the current v1 receipt (14 manual gates); preserve historical 13-gate receipts without treating them as approval for the expanded scope.

The release-policy mismatch is reconciled; current-candidate CI and real manual acceptance remain required, not inferred from earlier green runs. Current support decision.

Publication rendition boundary — 2026-11-04​

Correct the earlier frame-layout drift diagnosis: separate editable publication text is not inherently duplicate source authority. Inspected gaps are missing starting-revision/rendition binding, flattening and repeat append during text import, and incomplete rich-content export mapping. Existing history, overflow guards and extension round trips do not close those gaps. Visible author name/ORCID does not identify a document revision.

Inspected source, decision and acceptance work.

Outbound security and acceptance checkpoint — 2026-11-02​

Reclassify connection-time outbound transport from missing implementation to HARDEN: merged for WebDAV, PKP, configured publishing/AI, OIDC/ORCID/JWKS and Office callback paths. The host filter is partial and does not cover every outbound helper. Keep the remaining transport inventory, controlled-host negative tests and evidence reconciliation explicit.

Current evidence authority: reviewed security/acceptance checkpoint. Its dated status supersedes older claims below that #765/#774 are open or that connection-time pinning has no implementation. Historical findings remain historical; closed issues do not supply missing RC receipts.

Historical checkpoint: 2026-11-01. Earlier checkpoints are historical where superseded. Proposed repairs require merge/CI; real acceptance remains separate.

Session, integration and release evidence reconciliation — 2026-11-01​

Reviewable repairs: Studio #813 · Studio #814 · Studio #815 · Studio #816 · Studio #817. During the repair, Studio #812 merged the E02 collector. #817 reuses that existing latest-attempt/same-SHA/job/archive gate (including OMI-SPEC-320 conformance), adds candidate-only build/review and binds actual publication bytes/manual receipt; it does not introduce a competing workflow inventory.

Current componentStatusVerified problem → proposed repairResidual risk / evidence
EuroOfficeIsolatedFrame, embed hostHARDENAPI-origin remote code with same-origin sandbox → opaque origin plus HTTP CSP sandbox, exact window/channel checks, no API self networkHigh until matched API/native rollout and actual Office acceptance; real browser negative parent/storage/API access test
App.tsx, working store, recoveryREFACTORReact-owned tab manuscripts → application session manager, active-store adapter and runtime recovery fieldsMedium; switch already checkpointed. New external-open pending/selection/history regression, real store + memory ports, mobile E2E
nativeManuscriptFile.tsHARDENGlobal last-open path could overwrite another manuscript → per-manuscript runtime bindingsHigh data-integrity risk repaired; mocked native IO proves A/B output routing. Scoped-handle persistence remains separate
Extension routesSPLITVersioned payload on unversioned URL + route SQL → additive v1/legacy shared router, application service and repositoryMedium; HTTP tests for 401, strict manifest, owner isolation and parity; real PostgreSQL acceptance still separate
Container history serializerHARDENNested historical preview stripping changes legacy v1 digest subjects → preserve exact historical snapshots; declare implementation profileHigh integrity risk; actual-reader golden + independent ZIP/CRC/hash fixture. History may be larger; no silent v1 algorithm rewrite
PDF worker configurationHARDENProduction could silently execute locally → required socket and separate explicit local worker executorHigh until actual systemd/ACL/resource/egress evidence; renderer failure does not lock manuscripts
Release publication guardHARDENIssues/local markers alone insufficient → exact reviewed prior artifacts, hashes, same-SHA workflow evidence and mandatory manual receiptHigh promotion risk; foreign/missing/gate/hash negative fixtures. Actual host/device findings cannot be closed by CI

The companion repair is proposed pending merge/CI; this is not a declaration of deployment or 1.0 completion. OJS/OMP protocol and normative OMI schema stay unchanged. See ADR-004/005/006/007/014/015/016/017/018/019 and Studio ADR-030.

Current reconciliation: 2026-10-31, below. Earlier checkpoints remain historical where superseded. Proposed repairs require merge and CI; local evidence is not deployment acceptance.

Draft translation, exchange and authority checkpoint — 2026-10-31​

Audited Studio f41dadf. Proposed repair evidence is not merged/deployed acceptance.

ComponentStatusVerified gapTarget / migration / risk
translationExport.ts → buildStructuredTranslationPlanHARDENHidden/editor-only annotation bodies could appear in both HTML proofs after review.Apply the shared public-note predicate before output/readiness; retain private state; negative tests in release-security. High privacy risk until merged.
translationActions.tsREFACTORNew use cases directly own Zustand mutation and auth-store lookup.Extract translationCommands.ts with synchronous atomic session port; retain the old action API as composition adapter. Plain-session and existing save/history/revert tests. Medium coupling risk; no history rewrite.
translation.ts local extension/path anchorsKEEP + HARDENTiptap paths/fingerprints are useful transitional locators, not stable cross-editor anchors.Declare experimental profile and operation; retain unknown-profile preservation; version any later semantic locator conversion. Medium compatibility risk.
ExportFormatsPanel.tsx report lifecycleREFACTORByte/report/failure/cancellation orchestration lived in React.ExportExchangeSession owns that lifecycle; preserve renderers/sidecars/delivery. Complete registry, streaming and uninstrumented flows remain open. Medium evidence inconsistency risk.
collaborationPolicy.ts, ticket/live serviceKEEP + HARDENNew per-message/15 s idle checks and 5 s deadline are absent from central baseline.Record role authority, read-only downgrade, disconnect/reconnect and single-node whole-manuscript scope. Preserve denial behavior and tests. Medium availability/authority risk.
180/340/350/core-review reports and CIKEEP + HARDENMultiple evidence types may be mistaken for interchangeable certification or RC approval.Explicit scope/version/subject/freshness and incomplete/indeterminate distinctions. Core shared code is not independent evidence; exact-artifact release gates remain. Medium governance risk.

No OJS/OMP protocol or DB migration is needed for this repair. Ordinary authoring remains unrestricted. See Studio ADR-029. The new core Drafts are review targets, not a silent change to the main writer's specification versions.

Security boundary reconciliation — 2026-10-30​

Implementation audited at Studio 540f67a. Earlier rows are historical where this checkpoint is more specific.

Current componentStatusCurrent issue / remaining gap1.0 target and next actionRisk
verifiedFederatedIdentity.ts, central/institutionAccess servicesKEEP + HARDENVerified matching OIDC/SAML email now gates ACTIVE allow-listed bootstrap; older grants are not retroactively checked. Operator guide still described any linked federation.Correct guide; private grant review, trusted-IdP reauthentication and tested second-owner recovery; service-level negative tests. Do not forge profile evidence or revoke last owner. ADR-012/017.Medium: provisioning lockout / retained historical grants
cookieOriginGuard.ts, requireSession.ts, app.tsKEEP + HARDENStrict Origin policy is implemented; older scripts may omit Origin. Existing unit tests mostly exercise a helper/static mount order.Retain exact PKP/native exceptions and independent downstream validation. Add real HTTP invalid-bearer/no-cookie/parser-order regressions; document approved browser/native/server-script rollout. ADR-015/017.Medium: client 403, widened exception risk
pdfImportAdmission.ts, pdfImportRoutes.tsKEEP + HARDENUser/process active and retained caps, 429/503 and retry header implemented. No upload reservation or distributed quota; registry lost on restart.Document user 1/process 2 active, user 12/process 100 retained, lazy 30-minute retention. Require coordinated admission before multi-instance support. ADR-017.Medium: aggregate memory/load and recovery
pdfWorkerEnvironment.ts, Poppler/Vivliostyle callsitesKEEP + HARDENAllowlist excludes inherited API secrets; still same OS user, HOME/file access and networking remain possible.Keep minimization and synthetic subprocess tests; deploy independent least-privilege worker/filesystem/resource limits (#774), verify connection-time egress (#765). No claim of sandbox completion. ADR-016/017.High until deployment isolation/egress evidence
ci.yml, 1.0-readiness.yml, security testsHARDENSecurity unit gate grew; no unit test proves real IdP, database concurrency or host isolation.Add server HTTP/bootstrap contract suite after server setup; E02 aggregates same candidate commit/digest plus actual host/PKP/native acceptance. ADR-019.High for unsupported RC promotion

The targeted repair is documentation and test coverage, not a schema/API mass migration. No database migration is added. Rollout/recovery runbook. OMI/PKP plugins need no matching protocol change for these contracts.

Baseline: Studio eca2cf45762116e1c00a89b3397840ed89109511; OMI e2f421707f1154c0ed30cb7750e0689a837e9b0a; OJS plugin 6c0842fc1ce6c4a1ea75f7b27660af8011be303b; OMP plugin 4f7a2c0a2f29f68e3ad05ac73ca7652bc6410ac7
Date: 2026-09-19

This is an implementation gap analysis. KEEP/HARDEN/REFACTOR classifications describe the migration strategy for the audited codebase; they do not change the lifecycle or conformance status of an OMI specification.

2026-10-10 implementation delta (conditional on PR review)​

The original audited baseline above remains dated 2026-09-19. The following newer paths were checked against Studio main bb6cca95ed54615879cc6e6cc58a706425cd4e16; linked fixes are open PRs, not merged evidence.

ComponentStatusVerified driftTarget and next gate
src/services/exportIdml.ts annotation projectionHARDEN, privacy blockerThe appendix included hidden semantic origin and editor-only annotations; anchored private markers could be rendered.Studio #654 limits public notes and adds a synthetic negative fixture. Audit other public/anonymous renderers before RC.
DOCX omiSourceFont mark → extractOmiInlineRuns → IDMLREFACTORImport fidelity currently rides in Tiptap JSON; IDML does consume the shared inline adapter, so a claim that the renderer independently parses the mark would be inaccurate.Studio #657 pins the transitional behavior. A06/B06 still require a portable AST mapping, explicit loss diagnostics and save/reopen fixtures.
src/services/studioServer.ts native selectionREFACTORShared service imports Tauri and owns localStorage plus token invalidation.Studio #656 isolates platform access. B10 secure storage and origin-scoped cache/auth tests remain.
main image publish and 1.0-readiness.ymlHARDENlatest is published after build/collaboration while readiness only runs for selected paths.Studio #655 uses explicit beta/commit tags and full-commit readiness trigger. E02 exact-commit RC aggregation, PKP and installer gates remain open.

1. Classification key​

StatusMeaning
KEEPResponsibility and implementation are fundamentally correct; normal maintenance is sufficient.
HARDENGood 1.0 foundation, but conformance, error handling, security, performance, or fixture coverage must improve.
REFACTORBehaviour should remain, but module ownership or dependency direction must change.
SPLITOne current unit mixes multiple owners, lifecycles, or responsibilities.
MERGEOne concept/contract is implemented in unnecessarily parallel forms.
REPLACEA narrow implementation choice cannot become the 1.0 contract. This does not imply rewriting the whole subsystem.
REMOVEEmpty, unused, or misleading code has no 1.0 role.
DEFERCode may remain experimental/preview but is outside the base stable 1.0 guarantee.

2. Aggregate assessment​

The estimate is based on files, call paths, tests, and contracts rather than a formal LOC measurement.

  • approximately 75–85% of working implementation and behaviour can remain;
  • approximately 45–55% is KEEP/HARDEN without major structural change;
  • approximately 25–35% remains after wrapping, relocation, or ownership separation;
  • genuine REPLACE/REMOVE work is approximately 5–10%.

Refactoring effort will be larger than the replacement percentage because many consumers currently know the serialized Tiptap content representation or mutate application state through the Zustand store.

3. OMI format and domain​

Current componentStatusMain gap1.0 target
OMI omi-manuscript-0.3.schema.json and fixturesHARDENDraft schema and mandatory conformance checks now exist; stable immutable release/provenance remains pendingimmutable released schema with checksum and independent evidence
Studio vendored 0.3 schemaHARDENexecutable Draft validator exists; no stable release pin claimexact frozen schema with checksum and generated-contract review
Studio OMI envelopeHARDENschema and omi.version now declare 0.3; portable ContentCodec/future policy remains incompletegenerated wire codec with explicit classification and stable compatibility policy
src/types/omi.ts aggregateSPLITstructure, identity, notes, references, assets, history, and app concerns share one unitacyclic core/omi/* modules
empty domain placeholdersREMOVEfiles imply ownership without implementationreal canonical modules or no files
OmiBlock.content: stringREFACTORserialized Tiptap JSON/legacy text acts as portable domain dataversioned OMI content AST behind a Tiptap codec
inline semanticsHARDENgood semantics but partly tied to PM node/mark vocabularycore semantic vocabulary plus adapter mapping
section/study/front-matter helpersKEEPpure and testedmove only when useful, preserve behaviour
identity/contributor modelsHARDENscholarly semantics usable; account/evidence link not consistently explicitaccount-independent agent/contribution domain
metadata/localized text/keywordsKEEPpure domain logiccore metadata module
citations/clustersHARDENgood model, but content lookup depends on editor representationeditor-neutral references and anchors
CSL renderingREFACTORoutput policy mixed with citation domainrenderer/application service
notes/note citations/rich textHARDENinherits portable-content codec problemportable note content through common codec
named anchors/xrefsHARDENstable IDs exist; editor positions must not become authoritystable semantic targets and deletion/tombstone rules
assetsHARDEN/SPLITasset metadata can be domain data; byte/path/storage concerns appear in import codeasset identity/role/digest in core; byte store port
tombstonesKEEPsuitable deletion/history primitivehistory and anchor integrity
proofingREFACTORfull before/after strings and UTF-16 offsets are fragile under structural editssemantic operations plus stable anchors
publication correctionREFACTORimplicit across proofing/history/buildexplicit correction event linked to revision/artifact
unknown extensionsREPLACEno demonstrated lossless forward policynamespaced raw extension bag plus quarantine rules

Required migration pattern​

The format/content migration must happen in this order:

  1. establish the codec and validation facade around current data;
  2. migrate readers/consumers behind the facade;
  3. add shadow validation and future-version classification;
  4. only then change the stable writer to the frozen OMI representation;
  5. preserve or explicitly report unknown/unsupported data.

4. Editor engine and application core​

Current componentStatusMain gap1.0 target
BlockEditor.tsxSPLITeditor config, serialization, dispatch, and React UI are mixedTiptap session/codec plus thin React host
OMI Tiptap extensionsHARDENrich semantic implementation; mapping is not a frozen contractadapter extensions with documented mapping fixtures
continuous manuscript projectionHARDENgood study projection but knows current content encodingcodec-injected OMI↔PM projection
progressive study mountingKEEPcorrect large-document direction with testsruntime service with performance budget
focus registryKEEPclear runtime ownershipeditor-session service
selection helpersHARDENuseful cross-section behaviour; sensitive to remount/browser/IMEruntime selection with stable OMI projection
clipboard/pasteHARDENsanitizer/structured copy useful; content contract coupledimporter-like adapter with diagnostics
useStudioStore.tsSPLITportable domain, revision, selection, UI, timers, persistence, and use cases coexistUI/session projection only
app action modulesREFACTORdirect Zustand state mutations and type escapesplatform-independent command handlers
multiple checkpoint timersMERGEpossible parallel scheduling/lifecycle racesone CheckpointScheduler per document session
alpha workspace stateREMOVE/DEFERactive production use not demonstratedisolate post-1.0 experiment or remove after usage proof

Create/open/save should migrate first because they define the format and persistence safety boundary.

5. History and persistence​

Current componentStatusMain gap1.0 target
versioning modelHARDENstrong immutable linear revision semantics; storage assumptions leak incore history semantics behind repository port
working stateKEEPuseful distinction from durable revisionapplication history session
revision integrity/digestHARDENneeds canonical serialization and algorithm/version metadatacross-runtime reproducible digests
full snapshot per revisionREFACTORO(revisions × document size), duplicated history storesRevisionRepository, with future content-addressed/delta optimization
last-session IndexedDB persistenceREFACTORmanuscript/history/UI/session authority mixedversioned recovery cache validated by canonical gate
native manuscript file serviceREPLACEmodule-global path and parse/cast without schema/version gatesession-owned opaque location, validated open, atomic save
OMI export serializerREFACTORspread/stringify without canonical validation; localStorage signature injectioncanonical serializer and separate evidence bundle
asset repositoryHARDENuseful IndexedDB/memory seedquota/integrity-aware byte-store port
secure OMI container readerKEEPstrong ZIP size/path/CRC/SHA/symlink/encryption/ambiguity controlspreserve parser; separate semantic decoding and add fuzz
container writerREFACTORuseful package exists but contract is pre-stable and mixes concernsOMI-SPEC-330 versioned writer
publication signature in localStorageREPLACEnot a safe evidence authoritysecure evidence repository bound to manifest/artifact digests

6. Import, export, and publication​

AreaStatusDirection
importer actions overallMERGEcommon Importer registry, diagnostics, fidelity, provenance
DOCX importHARDENwrap the existing substantial implementation; add synthetic corpus, cancellation, memory/loss fixtures
DOCX app actionsREFACTORimporter returns draft; application command commits it
PDF importHARDENkeep as explicit best-effort/loss-reporting importer
Office/HTML/table/image importHARDENretain parser/sanitizer; unify quotas, externalize assets, add hostile corpus
MusicXML/MIDIREFACTORfix outer routing defect for MIDI, then place both behind capability probe
RIS/BibTeX/CSL-JSONHARDENpreserve reference interchange, add conflict/encoding/provenance coverage
Zotero/MendeleyHARDENprovider clients become scoped ReferenceManagerConnector adapters
export services overallMERGEcommon renderer result; delivery separated
file deliveryKEEPbecomes ArtifactDeliveryPort
JATS renderer/validatorsHARDENstrong candidate; pin validator/profile receipts and canonical XML fixtures
HTML rendererHARDENexplicit active-content/remote-resource policy, CSP/a11y, deterministic package
DOCX rendererHARDENwrap existing logic and test structural XML
print PDF/VivliostyleHARDENpin renderer/font/resource environment; visual regression and timeout/sandbox
EPUB/LaTeXDEFERpreview until fidelity evidence is sufficient
IDML/XPress/MIF/ScribusDEFERpreview; do not block base 1.0
publication profile module augmentationREFACTORmove presentation policy out of manuscript domain
paragraph/publisher stylesHARDENpreserve rich system behind renderer-neutral intent/mapping
publication build/manifests/hashesKEEPstrong 1.0 evidence pipeline seed
signature verificationHARDENclarify evidence placement, key versioning/rotation, cross-runtime fixtures

The key principle is wrap, do not rewrite. Stable support is earned by fixture, fidelity, security, and acceptance evidence.

7. OJS, OMP, and peer review​

Current componentStatusDirection
parallel OJS/OMP clients/routesMERGEone Publishing System Connector contract plus profiles
Studio OJS integrationHARDENpreserve signed launch/scope/review/writeback implementation
Studio OMP integrationHARDENpreserve chapter confinement/native context; map to common contract
OJS PKP pluginHARDENshared fixtures plus supported-version Docker E2E
OMP PKP pluginHARDENsame common contract plus monograph-specific extensions
direct submission/writebackREFACTORtransactional outbox, idempotency, durable receipts, explicit failure UI
peer-review serviceSPLITdomain transitions, persistence, projection/visibility separated
anonymous manuscript projectionHARDENretain server allowlist; extend to asset metadata/SVG/provenance leak corpus
reviewer/author visibilityHARDENone common policy and negative-access matrix
review forms/attachments/recommendationsHARDENassignment-scoped identifiers and auditable transitions
PKP workflow authorityKEEPOJS/OMP remain authoritative for their own external workflow state

Anonymous/restricted views are server-generated projections. The client must never receive confidential data and then hide it.

8. Identity, backend, and API​

Current componentStatusDirection
identity Prisma schemaHARDENformalize as account/auth/institution authority
duplicate main-DB user/session/identity conceptsSPLITreduce to StudioPrincipal projection/workflow references
principal bridgeHARDENadd idempotent reconcile/outbox/repair evidence
local password authHARDENversioned password-hash policy and rehash-on-login; document actual implementation
ORCID/Google/Microsoft/OIDCHARDENcommon provider state/nonce/PKCE/redirect/linking policy
institution/adminHARDENpolicy service backed by identity authority and versioned API
author/publication signatureREFACTORseparate evidence from portable manuscript/account state
duplicated provider credential locationsMERGEone encrypted CredentialRepository with provider metadata
native bearer token in localStorageREPLACEOS-backed SecureStorage; browser uses HttpOnly session
mixed /api and /integrations routesREFACTORadditive versioned /api/v1, legacy facades during migration
direct route→Prisma business mutationsSPLITroute → auth/policy → application use case → repository
Zod request validationHARDENmake Zod/OpenAPI schemas one transport contract source
shutdown lifecycleHARDENclose both DB clients, workers, and renderers deterministically

9. Integrations, storage, and platforms​

Current componentStatusDirection
client integration contracts/registryHARDENone shared capability schema; eliminate naming drift
server integration executionREFACTORserver-issued scoped ExecutionGrant, minimized payload
DeepL/AIHARDENexplicit purpose, retention, consent/audit; suggestion by default
arbitrary extension runtimeDEFERstable 1.0 supports reviewed built-in connectors, not a general executable sandbox
cloud-storage provider abstractionKEEPremote object-storage port
WebDAV/NextcloudHARDENredirect-aware SSRF, ETag conflict, recovery corpus
Google Drive/OneDrive/DropboxHARDENcommon receipts, scope inventory, refresh/revoke/retry
SharePoint-specific providerDEFERnot a base 1.0 blocker
synchronized local cloud folderKEEPSyncFolderAdapter with atomic/conflict semantics
device/platform detectionHARDENcapability-driven composition instead of scattered detection
Tauri capabilities/configKEEPdesktop adapter; add secure-storage/share/open-with ports
Android SAF/update/distributionHARDENpersisted URI permission, process death, save/reopen/update evidence
iOS/iPadOSDEFERshared core with Apple adapters; stable only after evidence
system font discoveryREFACTORplatform resolver produces versioned resource manifest for renderer

10. CI, testing, and documentation​

Current evidenceStatus1.0 action
Studio unit/pretest suiteKEEPpreserve as fast characterization gate; add boundary tests
frontend lint/buildHARDENadd bundle/performance budget and real lazy loading
server typecheck/buildKEEPadd both-schema migration checks
publication release suitesKEEPmake exact tool/artifact versions part of evidence
Playwright E2EHARDENexpand runtime critical journeys, recovery, accessibility
PKP Docker workflowHARDENreusable exact OJS/OMP matrix; no RC path filtering
desktop/Android workflowsHARDENsigned artifact install/open/save/update smoke
iOS workflowDEFERpreview evidence; security/privacy/data-loss remain blocking defect classes
1.0 readiness workflowHARDENone exact-commit aggregate release-candidate workflow
OMI website CIREFACTORmake file-format conformance mandatory where schema/spec changes
parser security testsHARDENshared hostile-input corpus and fuzz budget
accessibilityREFACTORrelease-blocking critical WCAG 2.2 AA journeys
recoveryREFACTORkill-during-save, corrupt file, recovery, outbox retry, updater rollback
governance/status pagesHARDENevidence-linked status to prevent prose drifting behind implementation
private manuscript policyKEEPonly synthetic/public/test data in Git/CI; content-free logs

11. Important documented mismatches​

  1. OMI-SPEC-320 0.2 vs Studio 0.1. The specification expresses the intended interoperable direction and Studio expresses current behaviour. Missing 0.2 integration is a blocker; neither side should be silently treated as already converged.
  2. Container status. Some governance/status prose trails the working Studio container implementation. Derive and test the stable contract from proven behaviour rather than inventing another container.
  3. Identity documentation. Some prose describes future or different identity/password assumptions while the audited code already has a two-database identity system. The actual authority and hash policy must be formalized.
  4. Review status. Server-side anonymous projection, visibility serialization, and PKP writeback are ahead of some status documents. The remaining gap is common contracts and adversarial/leak evidence, not a missing subsystem.
  5. Meaning of green CI. Normal green CI is useful health evidence but not full 1.0 release evidence because of path filters and incomplete conformance aggregation.

12. Explicit non-rewrite list​

Do not rewrite:

  • Tiptap and the current semantic editor extensions;
  • linear immutable revision semantics;
  • the secure OMI container ZIP parser;
  • DOCX/JATS/HTML processors;
  • publication-build manifest/hash foundations;
  • cloud provider and SSRF helpers;
  • Tauri/Android shells and release workflows;
  • OJS/OMP authorization and object-scope controls.

Place these behind stable boundaries and strengthen their evidence.

13. P0/P1 gaps before architecture freeze​

PriorityGapFreeze condition
P0empty Studio schema plus 0.1/0.2 mismatchcanonical schema pinned; open/save validation and quarantine operate
P0Tiptap JSON is the portable content contractOMI-SPEC-100 content grammar plus codec migration plan/fixtures accepted
P0container contract not aligned with OMI-SPEC-330manifest/path/security/version profile frozen
P0review/AI confidential scope partly client-controlledserver-issued grants plus leak/negative tests
P0native token in localStoragesecure-storage migration complete on stable native platforms
P1Zustand/application ownership mixedapplication facade and create/open/save use cases active
P1apparent dual identity authorityidentity DB authority plus StudioPrincipal projection/reconciliation established
P1parallel OJS/OMP contracts/writebackcommon connector DTO plus outbox/idempotency implemented
P1path-filtered release evidenceexact-RC aggregate workflow available

These gaps define the minimum architecture work that should precede a stable schema/API freeze.

14. Website assurance supplement — 2026-09-23​

The following rows describe the proposed implementation supplement, not the original 2026-09-19 audit snapshot.

Current componentStatusProblem1.0 targetTransitionRisk
webPublicationArtifact.ts, NewsletterPublishingPanel.tsxREFACTORCommitted artifact/hash and visible assurance exist; checkpoint/evidence/build orchestration still starts in React/Zustand.PrepareWebPublication application use case.Preserve renderer/UI, add facade and characterization tests before moving callers.medium
editorialDecisionService.ts, peerReviewRoutes.ts, Prisma editorial decisionsHARDENNative review acceptance adds an independent security authority requiring lifecycle/transaction proof.Immutable editor acceptance bound to exact revision/state digest; exclude OJS/OMP assignments.Role matrix, changed-revision, race/revocation and confidential-data tests.high
publishing/webPublication.ts, web routes, outbox migrationHARDENGrant, idempotency and transport receipts need real database/receiver recovery evidence.ArtifactDeliveryPort with target-version binding and exact payload receipts.PostgreSQL migration, WordPress/generic receiver, timeout/crash/reconcile and tamper corpus; retain Preview.high

The required visual seal is fixed across languages (OMI · PEER REVIEW · VERIFIED / NOT VERIFIED); explanation remains localized. Existing renderer logic is retained. No new general publishing platform or rewrite is required.

2026-10-09 evidence delta​

Current componentStatusProblem / evidence1.0 targetNext stepRisk
Studio omiPortableFormat.ts, vendored 0.2 schema and fixturesHARDENDraft schema validation and an edit/save extension fixture now exist; the original empty-schema gap is closed. The specification is still Draft and the observed round trip covers a finite corpus.Freeze only after normative requirement mapping, compatibility decision and independent evidence.Pin immutable schema provenance/checksum in CI; broaden lossless/future-version gates.high
ResearchModuleInsertPanel.tsxREFACTORMain previously read local module JSON and recursively flattened arbitrary fields into a Tiptap block without preview or consent.Section scaffold only until typed, authorized contribution exists.Review Studio PR #630; add a separate allow-listed, source-attributed projection with exact-content/source preview, consent, privacy fixtures and missing-source rejection.high
Module disciplineWorkspace.ts / statistical IndexedDB workspacesSPLITModule state is stored in browser storage; availability/retention and JSON export semantics differ from a portable OMI manuscript.Versioned module project records behind a workspace repository with explicit save/recovery outcomes and retention policy.Characterize existing imports/exports first, then introduce a port and migrate one module at a time.medium-high

These deltas update the 2026-09-19 snapshot; they do not retrospectively change the original classifications for that baseline.

External source attribution delta​

ComponentStatusProblem1.0 targetStepRisk
VisualInsertPanel / officeImport provenanceHARDENMachine filename/part metadata existed, but a file/clipboard object could be inserted without a reviewed visible source.Per-object source prefill, edit, confirm and portable rendering.Review Studio PR #632; verify export fidelity.medium
DOCX/PDF, OJS/OMP, HTML/JATS, reference-manager and asset ingressSPLITDifferent owners and source granularity; no proven universal object-source gate.Explicit source per document/record/object as appropriate, without rewriting valid OMI imports.Characterize each path, add missing-source and privacy fixtures, migrate incrementally.high

Source rule coverage (2026-10-09)​

IngressCurrent PRRemaining 1.0 risk
File/clipboard visual, including HTML visual elements#632Adjacency alone does not create a semantic bibliographic link; test every renderer.
DOCX/PDF conversion#634One document-level source is not a substitute for distinct citations inside the manuscript; avoid private file paths.
OJS/OMP editable intake#636 after #634PKP E2E and anonymity evidence; no reviewer/file identity leak.
RIS/BibTeX/CSL JSON#635Bibliographic record is itself the source; file lineage is not a work identifier.
Research workspaces#637 after #631Statistical IndexedDB, confidentiality, source validation and structured link remain.
JATS manuscript importno implementationDo not claim support; adopt gate if implemented later.

The coverage tracker is Studio #633.

2026-10-11 merged-state reconciliation and remaining gates​

ComponentStatusObserved gapTarget and transitionRisk
IDML annotation appendixHARDEN#654 merged; format-specific guard is presentKeep implementation and privacy regression; do not infer other formats from itmedium
HTML/JATS/DOCX and LaTeX/MIF/SLA/XTG annotationsREFACTORPublic output can disclose hidden or workflow notes#662 adds one non-mutating publication view and eight-format synthetic fixture; merge only after CIhigh
EPUBHARDENInline private markers can leak; note-body fidelity remains limited#662 filters anchors; explicit fidelity limitation remainsmedium
Readiness and main image channelsHARDEN#655 merged; per-commit readiness is no RC promotion aggregateRetain continuous beta builds; E02 must bind evidence and artifacts to one commit/digesthigh
Native origin-selection adapterHARDEN#656 merged; token invalidation alone is not secure storage/cache isolationB10/D01 secure storage and origin-scoped account/cache acceptance remain openhigh
DOCX source-font adapterKEEP / HARDEN#657 characterizes interim compatibility behaviorKeep tested code; ADR-004 codec/AST freeze still pendingmedium

2026-10-12 current implementation checkpoint​

This checkpoint supersedes earlier dated implementation-state descriptions; it does not retrospectively change their audit evidence or declare architecture freeze. Studio main was verified at 6242dbae4e6e7be3f9bff9470f64647554f363d8; OMI at bc932272d0cbe18b8e45f758fbdb0d6076344bbf.

  • Studio now writes/validates OMI-SPEC-320@0.3.0 Draft with schema and omi.version, and retains the separately versioned container/history layers. Do not add a redundant schemaVersion field or promise pre-stable 0.2 migration. Stable compatibility obligations still begin with the first frozen stable generation.
  • OMI-SPEC-225@0.1.0 defines sourceAttributions: independently identified prov:wasDerivedFrom links to bibliographic records with explicit machine-inferred / author-declared / author-confirmed status. These are distinct from citation occurrences, technical import provenance, rights and authorship evidence. The model exists; universal ingress/export coverage does not.
  • Studio PR #674 proposes atomic source-relationship cleanup on block-subtree deletion, retaining unrelated relationships, bibliography records and prior revision snapshots. It also proposes explicit HTML/JATS source-attribution-not-preserved warnings. Until merged and verified, these are repairs under review, not main guarantees. The warnings disclose loss; they do not establish a lossless mapping or generate a citation/licence claim.
  • Mandatory regression evidence for this repair is actual insertion → checkpoint → subtree deletion → checkpoint → OMI save/reopen → historical revert, plus HTML/JATS per-relationship loss diagnostics and non-mutation. Fixtures must be synthetic/public. Broader deletion/merge paths, additional exporters, anonymous projection and OMI-container source-graph round trips remain separately assessed gates.
  • Preserve the working editor, snapshot history and renderers. Portable AST/ContentCodec, immutable schema provenance, unknown-field/future-version handling, secure native storage and exact-commit RC evidence remain prerequisites; 0.3 adoption is not 1.0 readiness.
Current componentStatusProblemTarget / actionRisk
sourceAttributions + visual insertionHARDENstructured relationship exists; not every ingress is coveredpreserve OMI-SPEC-225 status and explicit source identity; extend coverage only with fixturesmedium
stageRemoveBlockHARDENcurrent main leaves a dangling relationship after deletion#674 atomic cleanup, retain prior history; gate delete/save/revert lifecyclehigh
HTML/JATS source graphHARDENmain silently omits unsupported relationships#674 per-relation warning; future documented mapping and reimport testsmedium–high
Original 0.1/0.2 mismatch observationsKEEP as historyobsolete if read as current stateuse this checkpoint and revised format rows for current planningmedium

2026-10-13 — ODT Preview trust and fidelity checkpoint​

The earlier #674 source-lifecycle repair is merged at 5825c15a1bc3860a3371eaa1c20111bfe4b74ea5; its prior proposal checkpoint is historical.

ODT import/export entered Studio main in #675 (commit 5417e19cb26550508b66c780107eea87382b5f2f). It adds an interoperability format, not a new OMI schema, account authority or durable working-state format. ODT is Preview, pending independent ODF/office fidelity and real-document acceptance; it must not be promoted merely because the ZIP can be opened.

Studio #678 proposes targeted containment of two reproduced gaps. The import reader had trusted declared ZIP sizes and fully materialized DEFLATE output before checking it. The exporter omitted sourceAttributions without loss reporting. Treat #678 as a proposal until merge and verification, not a main guarantee.

The proposed boundary retains the 200 MiB input-package limit and enforces 25 MiB per expanded part, 100 MiB aggregate expanded data, and 4096 entries. Actual streamed bytes trigger cancellation before complete output materialization. Local/central headers, bounds, CRC32 and duplicate/unsafe/overlapping entries are checked; encryption, symlinks, ZIP64 and multidisk are explicitly unsupported. Successful repeated reads share the cached budget. CRC is integrity checking, not authenticity.

ODT uses the shared source-attribution-not-preserved diagnostic per unsupported relationship, alongside existing visible delivery warnings. A source label is not a semantic derivation link; no citation, licence or authorship claim is fabricated. Portable OMI retains the authoritative graph/history.

AreaDecision / release gate
ADR-008 / ADR-017 — importHARDEN the existing ODT adapter with real output-byte limits and ZIP integrity. Forged STORE/DEFLATE lengths, aggregate/entry quotas, CRC/header corruption, duplicate paths, truncation and ordinary data-descriptor compatibility are mandatory synthetic regression cases. Preview does not waive security gates.
ADR-008 / ADR-009 / ADR-023 — exportReport loss of every unsupported source relationship, with stable code and object target; preserve original portable state and keep diagnostic messages content-free. A later target mapping needs independent export/reimport proof.
ADR-019 — capabilityODT remains Preview with documented supported subset and losses. ODF schema/independent office acceptance and real-document fidelity gate Stable promotion.
PR planMerge/review #678 as the bounded incremental repair after #675; retain the working import projection and delivery paths. Then prove independent ODF validity and cross-office fidelity before any Stable claim. No beta-format migration or renderer rewrite is justified.

This checkpoint supplements the earlier format matrix and source-graph gates. The original schema/API/ContentCodec/native security/exact-commit RC prerequisites remain in force.

Architecture repair checkpoint — 2026-10-15: workspace mode and passkeys​

ADR-001/002/005: simple writing is a Studio workspace preference keyed by manuscript ID, persisted locally outside portable manuscript state and revision snapshots. New documents do not write editorMode; the Studio vendored 0.3 schema must match the OMI source artifact. Existing unknown fields remain preserved under the general round-trip policy; this repair does not silently delete an imported extension or reinterpret it as authorization. Moving a file to another installation does not transfer local writing-mode preferences. The simple visual editor remains available.

ADR-012/015/016/017: passkeys authenticate the existing local account; they do not establish scholarly identity or publication evidence. /api/v1/auth/passkeys is the client contract. /api/auth/passkeys remains a compatibility facade over the same handlers and rate limits. HTTP routes own transport/status/cookies, the passkey service owns WebAuthn use cases, and the repository adapter owns identity-database access. Challenges remain expiring and single-use; verification still binds RP, web origin, user handle and user verification. Credential material never enters OMI.

Support is web-origin only: platform-authenticator availability is checked behind a capability adapter. Tauri origins do not advertise this flow. Native promotion requires a verified browser/native handoff and contract tests; do not widen accepted origins to make native UI appear supported. PostgreSQL migration, real authenticator registration/login/revocation, negative origin/RP/signature cases and native handoff evidence remain release requirements, not claims established by mocked route tests.

Validation for the repair includes per-document workspace isolation, unchanged portable bytes/history, schema parity, native-origin capability rejection and v1/legacy route equivalence with replay rejection. This checkpoint records a proposed repair pending merge and CI, not a completed 1.0 freeze.

External integration repair checkpoint — 2026-10-18​

ComponentClassificationGap and targetChange/evidenceRemaining risk
EuroOffice callbackHARDEN / SPLITUnconditional updates and route-owned use cases → revision CAS, receipt and service/repository ownershipImmutable force/conflict recovery, prior-head backup, row lock; duplicate/stale/concurrent testsReal provider force-save/final lifecycle and PostgreSQL gate must pass
DOCX downloadHARDENFull arrayBuffer plus magic-only check → actual streamed quota and bounded ZIP/OPC/XML integrityMissing/forged-length, malformed/CRC/expanded-quota fixturesNot full OOXML schema or rendering conformance
Integration API/platformREFACTORNew unversioned surfaces and duplicated native token reads → additive v1 plus shared transportLegacy facades retained; origin-switch token testExisting general native secure-storage work remains open
Paid recognitionHARDENModule toggle/consent alone → expiring grant and durable digest-bound intentUnique request key, ACCEPTED/UNKNOWN status, no automatic paid retryProvider cannot guarantee exactly-once billing; unknown outcomes require external reconciliation

The separate ordinary-DOCX storage is intentional and does not change OMI schema/container semantics. Historical simple-writing-mode checkpoint text is superseded by its subsequent removal; no portable editorMode field is reintroduced. Repairs are proposed until merged and migration/CI evidence is available.

2026-10-21 — Federated discovery repair checkpoint (ADR-026)​

The new server directory is an operational discovery authority, separate from account, manuscript, review and publication authorities. DNS proves domain control only. Heartbeat online denotes a recent authenticated heartbeat, not service health, compatibility, capacity or trustworthiness. Browser selection opens the chosen origin; native selection retains the existing origin/token boundary. Discovery never proxies application traffic or transfers accounts/documents automatically.

The repair introduces /api/v1/server-registry; the old /api/server-registry is a shared-handler facade. The route delegates to application service, repository and GitHub audit adapters. Registration needs a private claimant capability distinct from public DNS data, expires in 24 hours, and uses database CAS for PENDING → VERIFYING → ACTIVE. Unknown audit outcomes stay quarantined without automatic issue resubmission. ACTIVE credentials survive the additive migration; old PENDING entries must re-register. Rotation/revocation/heartbeat updates compare current state and token atomically.

This is a proposed repair checkpoint, not a completed 1.0 freeze. Mandatory evidence includes synthetic negative/dual-mount tests, actual PostgreSQL migration/concurrency/lifecycle tests, plus controlled external DNS/GitHub acceptance and operational recovery. Deploy migration → API → clients/CLI/website; keep discovery Preview until acceptance evidence exists. Origin-scoped account/cache ownership and native SecureStorage remain separate open work.

ComponentStatusGap / targetNext stepRisk
Registry routes / service / repositorySPLIT / HARDENPrivate proof, v1 and CAS replace route-owned verificationMerge repair, migration and dual-mount testshigh
Registry audit and operational recoveryHARDENQuarantine unknown side effects; no blind replayOperator DNS/GitHub/recovery acceptance; future durable audit toolingmedium

2026-10-22 — Document import origin clarification (ADR-023)​

This checkpoint supersedes the requirement to anchor whole-manuscript own-work declarations to imported content. OMI-SPEC-225@0.2.0 permits own-work imports without bibliographic attribution. An explicit own-work choice for DOCX, ODT or PDF stores the version-1 https://openmanuscript.org/extensions/manuscript-import-origin/1 document extension: sourceFormat, importedAt, and assertion: author-declared:self-authored-outside-omi. It is self-attestation, separate from a citation and verified authorship. No filename, path, URL, account identity, section ID, block ID or anchor is stored. Other import paths retain their existing per-object policy.

Every document part remains editable, replaceable, movable and deletable. The declaration is document metadata and creates no content relationship; replacing all imported content does not produce dangling anchors. It records a historical import declaration, not authorship of later edits. Working OMI, history snapshots and core-snapshot interchange preserve it via the existing namespaced extension contract, without a schema-version change. Public publication projections omit the extension without mutating the working document; anonymous server review projections remain allow-listed and exclude it.

Validation for the repair: DOCX/ODT/PDF own-work intake, deletion/replacement of imported content and titles, save/reopen, history-free core-snapshot round-trip, revision snapshot retention, public export omission and anonymous review filtering. Keep the existing editor, import parsers and revision engine. No database migration or legacy-marker conversion is required. Deployment status remains proposed until the coordinated Studio repair is merged and CI passes.

2026-10-25 — Publisher profile import repair (ADR-009)​

External file, URL and catalog profiles must pass a shared nested contract validator before activation. Validate required rule groups, enums, finite bounded layout values, citation style identifiers, protection shape, attached style structure and JSON complexity; reject prototype keys. Invalid input must not enter storage or change the active profile. The import boundary records SHA-256 over the exact received bytes for all three paths; catalog imports additionally compare the independently supplied catalog digest. A digest identifies content, not publisher identity. Imported protection metadata remains an unverified read-only declaration, never an authorization credential.

Generic scholarly and humanities profiles with independent note/bibliography defaults advance to 0.2.0. Retain explicit 0.1.0 definitions; resolve known versions exactly and reject unavailable versions instead of silently selecting newer behavior. Existing publication-build manifests already hash the resolved profile; retain that digest independently from local download provenance. Changes to render-affecting profile defaults require a new profile version. No application release bump or OMI-schema migration is needed for this repair.

Release evidence: negative nested-input fixtures, checksum mismatch and malformed UTF-8 rejection, deterministic byte hashes, old/new profile resolution, existing publication-build and renderer regression suites. The validator remains an application contract for the alpha profile model, not a newly frozen normative OMI schema. Further limits for streamed network reads and independently authenticated publisher signatures remain hardening work; checksum comparison alone does not establish trust. This checkpoint is proposed until the coordinated Studio PR is merged and CI passes.