Why: Register the U9 public contract proposal.
Evidence: CI success on 3a3904fe (governance).
Impact: Documentation only.
Refs: .task/feature-regulations-kb-ext20-proposal.md
Why: Task 10 closed four independent review rounds and needs a durable record.
Evidence: Backend a2f4978 SELF suite plus host REPLAY of tools/docgen tests and Dev build.
Impact: Documentation only; no product behavior change.
Refs: docs/records/2026-10-15-unified-document-renderer/09-implementation-plan.md Task 10
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: Task 9 now has a reproducible Renderer dependency, CLI implementation, independent review, one corrective pass and exact local/isolated-PostgreSQL evidence that require human assessment.
Evidence: The report fixes RDM 019d60d4 and Renderer 732e7e7, records all review findings and repairs, focused race, vet, make test/build and disposable PostgreSQL verification.
Impact: Documents the local RDM rendering transaction candidate without publishing profiles, migrating a real database, starting Task 10, pushing, merging or deploying.
Refs: Unified document renderer Task 9
Why: The three blocked Word bases now have safe derivative candidates implemented by CLI, independently reviewed, corrected, and verified without publishing profiles or changing source assets.
Evidence: RDM 48120ed and Renderer 732e7e7 bind a clean 44/44 compatibility receipt; the report records the review findings, one corrective pass, 20 safety tests, generator checks, and full build.
Impact: Replaces the prior 41-pass/3-blocked implementation snapshot with 44 safe-preflight draft bindings while retaining human visual and publication gates.
Refs: Unified document renderer Tasks 7-8
Why: PostgreSQL and authenticated product-matrix runs must remain tied to the exact RDM commit on which they executed.
Evidence: The report now distinguishes final-head SQLite and full-suite replay from the earlier 3f60265 PostgreSQL and 44/44 product runs.
Impact: Documentation-only evidence correction; no implementation or publication state changes.
Refs: Unified document renderer Tasks 7-8
Why: 8788 could not render docx because the sandbox hides host site-packages and there is no request path for extra packages. The proposal fixes the design before the implementation lands.
Evidence: SELF — proposal written from the cnv_7936c0a799ff transcript and the existing exec-deps-dir code path.
Impact: Documents only. Index lists the proposal as pending review, not approved.
Refs: docs/proposals/2026-10-15-layered-python-env.md
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: Tasks 7-8 need a reviewable record that separates draft profile generation, legacy RDM product-chain evidence, and actual renderer publication readiness.
Evidence: The report fixes Renderer 732e7e7 and RDM ec24a42, records the independent review and repair, and distinguishes 44/44 compilation from the 41-pass/3-blocked base-package safety preflight.
Impact: Human reviewers can decide corporate visual values and the three unsafe base-package derivatives without treating local SELF checks as publication, migration, or deployment.
Refs: Unified document renderer Tasks 7-8
Why: Fix Task 4-6 behavior, independent review findings, evidence, limits, and the next human review boundary in one durable record.
Evidence: SELF report content binds renderer 0efa07f8ce5cb4e1ff93886cf6769d111cba9872 and its full, race, vet, and diff verification.
Impact: Adds an implementation review record only; no runtime integration, migration, deployment, or publication.
Refs: unified-document-renderer Tasks 4-6
Why: Update the batch review after replacing the temporary profile interface with the immutable snapshot produced by Task 3.
Evidence: SELF contract test observed RED then GREEN; renderer full, race, vet, and diff checks pass at a08ae7ac81d4c648b350d890da97c91e64950769.
Impact: Updates the implementation review record only.
Refs: unified-document-renderer Tasks 2-3
Why: Give the first implementation batch a fixed human-review object with exact outputs, risks, review findings, fixes, evidence, and remaining boundaries.
Evidence: SELF relative-link and diff checks pass; the record cites exact RDM and renderer candidate SHAs and their fresh verification commands.
Impact: Adds an implementation review record only; no runtime, migration, deployment, or publication behavior changes.
Refs: unified-document-renderer Tasks 1-3
Why: Define the approved direction for one rule-driven Word renderer, all-template onboarding, human-readable traceability, and review-aware delivery before implementation starts.
Evidence: SELF documentation checks passed for relative links, balanced code fences, JSON parsing, manifest SHA256 coverage, and git diff whitespace.
Impact: Adds a 12-document review set and immutable review manifest; no runtime code, migration, deployment, or published rule changes.
Refs: feature/rdm-two-trees-latest-templates
Why: the duty page needs one record for the four domain rulings, and the earlier settings plan must show which duty-page sections it no longer owns.
Evidence: SELF review of the proposal text against the approved implementation plan.
Impact: r1 section 4 and the duty-page parts of P5 and P7 point at r2. Settings, members, and the permission matrix stay in r1.
Refs: docs/proposals/2026-10-15-rdm-duty-matrix-rework.md
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: The living proposal still said the human confirmation write path was blocked after the local chain had been implemented.
Evidence: Backend a4d6404, RDM 528dd0c, and Dev SELF receipt edb81a90 are cited in the new progress record.
Impact: The status now separates implemented, verified, merged, and deployed. Nothing is merged or deployed.
Refs: W4b human confirmation design r1
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: keep the compose proposal in the document set while it remains under review.
Evidence: this repository has no workflow; head 2a8deac89d contains origin/main. Backend CI is already green and merged.
Impact: proposal status stays pending review. Inclusion is not approval.
Refs: #107
Why: deploy needs a reviewed design for compiling selected plugins into the base binary before that behavior is treated as approved.
Evidence: the proposal records the throwaway compose build; implementation status stays unreviewed.
Impact: adds a pending proposal and an index link. Inclusion does not approve the design.
Refs: Dev .task/feature-svc-plugin-compose.md
Co-authored-by: Cursor <cursoragent@cursor.com>
Why: The proposal main record must reflect that trusted human confirmation lacks a public Backend proof and remains blocked pending review.
Evidence: Candidate design bbcff18 and exact file SHA256 are linked; W3 RED and Backend source gap remain separate from phase acceptance.
Impact: Updates status only, preserving the previously approved stage-four and stage-five objects.
Refs: Stage6 W4b design adjustment candidate r1; TT-D02/D03
Why: The pinned Backend extension exposes no trusted one-time HITL proof for RDM structure confirmation.
Evidence: Fixed RDM/Backend source identities, W3 RED receipt, and Dev READY precheck are cited; no W4b confirmation write was run.
Impact: Adds a review-only X3 verifier and RDM atomic-consumption design with permission, recovery, audit and replay checks.
Refs: Stage4 design r1 section 5.2; Stage5 W4; TT-D02/D03/T04
Why: The user approved the fixed TT-N03 and TT-R01 stage-five amendment for local stage-six implementation.
Evidence: User approval call_497a8a4c28bd4231a2cfd63d84375f2a for plan SHA256 10b91e3489ac8145eafec73759a019cb6d3b73fd50e7c74563bbbb6d18e5462c.
Impact: Update the current approval record without changing the approved plan bytes or permitting live migration, tag publication, or deployment.
Refs: TT-N03, TT-R01; Dev task card
Why: The approved stage-four design needs reviewable ownership, sequencing, and tests before new X3 product code can start.
Evidence: TT-N03 and TT-R01 stage-three approval; stage-four approval call_a6853c4232f34285a6cbae19dfbeeec5; document checker SELF receipt f4a4cdde2d3ad9b250742da764455a1304e1e26f07270da4f0d37ad6ed8b6c7e.
Impact: Record the stage-five amendment for human review; keep the original 28-item plan and approved implementation active.
Refs: TT-N03, TT-R01; work observation and risk stage-five amendment r1
Why: The authoritative target added DHF and manual merge queue records while this task saved approved two-tree amendments.
Evidence: origin/main=19aa34efe6482b78d779bab573b5397e6d2088b2 includes the Dev incoming gitlink 46506aec34.
Impact: Keep both documentation histories; approved two-tree object bytes remain intact.
Refs: Dev task-sync and stage 4 X3 design
Why: The approved work-source requirements need a governed path from real messages, retrieved material, and analysis into project evidence and timely clarification.
Evidence: Stage-three amendment SHA256 552a9fdea996bb547c185949b5072a51efafdce1339a9eaafd49537dfedd91b4 and user approval call_b517d0f07af54aa188a01ddb100485b9; independent Backend and RDM scoped design reviews closed the identified privacy, timing, and atomicity gaps; document hashes and links checked.
Impact: Save an unapproved stage-four design only; no new public interface, migration, model run, or product behavior.
Refs: TT-N03, TT-R01; work observation and risk design r1
Why: Preserve the owner’s Stage 5 decision and implementation method on the reviewed plan object.
Evidence: Owner approved plan 28c4ceb in the task conversation; git diff --check and relative links passed (SELF).
Impact: Documentation status only; Dev implementation remains subject to Stage 6 and 7 review.
Refs: Agent-System-Documents #105; Agent-System-Dev #518
Why: The approved Stage 4 design needs a reviewable Stage 5 implementation and test sequence.
Evidence: git diff --check, relative-link validation, M2-1 through M2-6 coverage, and placeholder scan passed (SELF).
Impact: Documentation only; no Dev queue activation or task takeover.
Refs: Agent-System-Documents #104; Agent-System-Dev #518
Why: The first trial reduced final CI churn, while later Dev merges exposed gaps in shared window records that must be addressed before broadening manual coordination.
Evidence: SELF Gitea API snapshot for PR #539–#548, current #518 comments, protection and open PR states; fixed first-trial report and current project rules reviewed.
Impact: Add a stage-4 design and evidence snapshot only; no queue activation, platform protection change, production task takeover, or workflow edit.
Refs: Dev issue #518; Documents 2026-10-14 merge queue trial report
Why: The user approved adding source processing and timely ambiguity clarification to the goal.
Evidence: Stage-two amendment SHA256 2627342f7d6e94137d5e7418d995ce57218093f6512f67fb7b221f01067da04d and chat approval call_aea78d4156fe4ca89f4b9b6883b2b23f; document hashes and 158 local links checked.
Impact: Add two unapproved stage-three requirement IDs without changing the original 28 or authorizing the public observation implementation.
Refs: TT-N03 and TT-R01; stage 3 amendment r1
Why: The expanded user requirement needs explicit normal-work source intake and timely clarification criteria.
Evidence: Original .tmp data-flow and dialog sources were matched by SHA; fixed Backend public surfaces and current requirements were checked; local links and approved document hashes passed.
Impact: Add an unapproved stage-two scope evaluation and two candidate requirement IDs; prior approvals and product code remain untouched.
Refs: TT-N01, TT-G03, proposed TT-N03 and TT-R01; stage 2 amendment r1
Why: The authorized small tool stream again lacked an observed DONE marker, leaving transport and provider causes unresolved.
Evidence: U3 r0.5 terminal receipt 6b013965 and independent assessment f35d1c17; local links and diff checks passed.
Impact: Design only; the missing-DONE tool gate, runtime behavior, and paid resource boundary remain unchanged.
Refs: Agent-System-Dev feature/fast-slow-u2-u3-plan; Documents PR #95
Why: Preserve the authorized one-shot diagnostic and its limits in the durable validation record.
Evidence: Dev run 3136; SELF terminal receipt 6b013965; independent result assessment f35d1c17.
Impact: Documentation only; tool execution gate and A01 quality status stay unchanged.
Refs: Agent-System-Dev task feature/fast-slow-u2-u3-plan
Why: the progress note still listed Documents #102, RDM #365 and Dev
#540 as WIP and said the stage-8 review was under rectification, but all
three PRs merged on 2026-10-14 and the Dev task card is archived; the
note must state the terminal state instead of stale WIP/GATE caveats.
Evidence: merge SHAs 9062445 (Documents), 57d2214 (RDM), 11aa76a (Dev)
per the archived card .task/archive/feature-dhf-template-pipeline-plan.md
(Dev #542); this change is documentation-only and does not touch the
frozen r0.1 body or its recorded SHA256.
Impact: readers of the plan see the merged terminal state and the
remaining U1/U2-U6 exits; no requirement or scope change.
Refs: Dev task feature/dhf-plan-progress-note
Why: The review found a provider-label leak and incomplete browser evidence for the candidate.
Evidence: Sentinel RED/GREEN, twelve exact-pin browser cases, and isolated PG/Redis receipts.
Impact: Distinguishes the reviewed old object from the fixed zero-paid candidate; model quality stays NOT_RUN.
Refs: U3 r0.5 independent review
Why: The merged index predates the approved r0.5 implementation and standalone CI applicability check.
Evidence: Dev task card and isolated component/combination receipts; Documents origin/main merged at 9062445.
Impact: Corrects review status and marks absent Frontend/Documents workflows N/A.
Refs: U3 r0.5 implementation plan Task 5
Why: The task needs an auditable distinction between the new mechanism checks and the failed r0.4 model sample.
Evidence: Dev task receipts for component tests, browser replay, isolated PostgreSQL, and composed build.
Impact: Documents the candidate and leaves r0.5 model quality as NOT_RUN.
Refs: U3 r0.5 implementation plan Task 5
Why: Keep the verified RDM candidate and independent replay available beyond temporary reviewer files.
Evidence: Report, original receipt, and replay script were copied byte for byte; the archive and exact pins were validated before focused race replay.
Impact: Documents only; the review covers P2, while time repair and performance acceptance remain separate gates.
Refs: RDM 57a8d7caedf72d9be4bdca179ae2a93e456ba009; stage 6 W4a
Why: RDM PR 365 is now merged and the unique U1 content contract is available for W3.
Evidence: Fresh remote main and API metadata agree on 57d2214feb0ac4a534fa58623fff7d02f93a860a; 44 assets remain needs-review in the unchanged catalog.
Impact: Clear the U1 merge prerequisite while retaining read-isolation, time repair, and template publication gates.
Refs: stage 6 W1, W3, W4a; dependency receipt dc68957a
Why: Preserve the merged DHF plan and governance records before refreshing the Dev combination.
Evidence: origin/main was fetched as 9062445db8082976a124e54a0000007b6391d6cf; approved two-tree records retain their hashes.
Impact: Integrate upstream documents without replacing this task or its approvals.
Refs: stage 6 dependency coordination
Why: The user approved the fixed stage 6 amendment for durable domain times and honest legacy gaps.
Evidence: Chat call_6ce04b5a3ca3496487ab2714ed208965 item 0 approves document SHA256 d78849440e6346482fc09925d402790a7ad257ef76c85275610046a845a45f37; original bytes retained.
Impact: Permit isolated implementation and tests after read isolation; real database migration and deployment remain outside this approval.
Refs: TT-D03, TT-A02, TT-A04; stage 6 time-contract amendment r1
Why: Fixed replay RED and source review show omitted domain timestamps and clock rollback can change recovered project state.
Evidence: RDM 8f80e55 contains three deterministic time RED cases; document hashes and links were checked through Dev evidence.
Impact: Save an unapproved schema and time-contract amendment plus current component and DHF dependency status; production time semantics remain unchanged.
Refs: TT-D03, TT-A02, TT-A04; stage 6 W4a and approved read-isolation revision
Why: Delegated RDM renames currently lose the real operator before the Backend project write, leaving an account-disable race in the Backend database.
Evidence: Read-only source review identified both HTTP lifecycle and command rename paths; the proposed contract and validation plan are pending human stage 4-5 review.
Impact: Adds a narrow request-bound projection capability proposal and two-account final transaction plan. No product code, production state, or approval status changes.
Refs: Approved U3 r1 and A1-A4 amendment; Dev feature/iam-u3-account-directory-recovery task card.
Why: The approved plan remained an earlier review snapshot and contradicted the index and task card after U1 implementation began.
Evidence: The pre-decision r0.1 file hash is f99e1424; the additive dated note preserves that text and records owner stage 6/7 decisions, four package reviews and remaining gates.
Impact: Documentation status only; no asset or product behavior change.
Refs: docs/proposals/2026-10-14-dhf-template-generation-plan.md
Why: Shared project read pointers can modify cached state without an event or transaction.
Evidence: User approved the immutable r1 design and its alias, publication-order, and performance verification scope.
Impact: Allows isolated W4a state-copy and sealed-log implementation; historical time semantics remain a separate issue.
Refs: feature/rdm-two-trees-latest-templates W4a
Why: Shared read pointers still violate project snapshot isolation, while context and audit candidates now have a tested local combination.
Evidence: Fixed RDM RED, source-reviewed read-isolation options, and clean Backend combination SELF receipts are linked with precise limits.
Impact: Preserves a pending W4a design and local probe report without claiming stage-six acceptance or component release.
Refs: feature/rdm-two-trees-latest-templates W4a W6 W9
Why: The approved r0.5 design needs an auditable zero-paid execution sequence after the failed r0.4 model sample.
Evidence: The approved design object is Documents 0d473fe; current source seams, file ownership, relative links, and whitespace were checked before drafting this plan.
Impact: Add five test-first work units and explicit stage, budget, privacy, and stop gates. This commit contains no product implementation or model call.
Refs: U3 r0.5 design; Dev task feature-fast-slow-u2-u3-plan
Why: The authorized r0.4 A01 exposed an unobserved stream terminator and an unsupported numeric relationship in the immediate reply.
Evidence: Independent A01 assessment and fixed-head Backend/Dev CI are linked from the design and task record; relative Markdown links and git diff --check pass.
Impact: Propose content-free terminal diagnostics and a source-safe host acknowledgement while retaining strict tool rejection and the original A01 quality gate. This document grants no implementation or model budget.
Refs: U3 r0.4 validation; Dev task feature-fast-slow-u2-u3-plan
Why: Explicit low audit IDs can bypass the old default function and disappear across fixed-cutoff pages.
Evidence: User approved the immutable r1 correction; the exact isolated PostgreSQL RED receipt is preserved byte-for-byte.
Impact: Authorizes only isolated all-writer guard and offline-copy testing; live migration remains outside scope.
Refs: feature/rdm-two-trees-latest-templates W9
Why: An isolated PostgreSQL test demonstrates that explicit low IDs bypass the current committed-prefix guard and disappear across pages.
Evidence: The test-only RED and reviewed database/migration tradeoffs are bound to the documented candidate and exact SHA256.
Impact: Adds a human-review design for all online audit writes and exclusive offline copy; no runtime change or migration.
Refs: feature/rdm-two-trees-latest-templates W9
Why: independent review found in-flight write, API-key admission, lock-order, and cross-instance cursor gaps in the U3 candidate.
Evidence: SELF source review of Backend 4053c4d against approved U3 r1; proposal records triggers, scope, and replay plan.
Impact: request stage-four and stage-five re-review before further affected implementation; no production action.
Refs: IAM U3 r1; Dev policy/delivery.md
Why: prepare the U3 review amendment against the current documentation baseline.
Evidence: origin/main=2d35cd49c203820d24d7c38acead06d37793c1f1.
Impact: integrate upstream documentation without changing the approved U3 r1 text.
Refs: IAM U3 r1
Why: The single authorized real-model trial produced no persisted checkpoint or final result and exposed an unsupported fast-channel inference.
Evidence: Independent read-only assessment SHA 2ac373ed66d47aaa9cdb699d3bf77eb8a515d46956ee0321b70cc0230d79b667; restricted 20-file source-verified archive manifest SHA 881991dcab17ad2539f8b24332f962e3505ab2f248662f58283fe08daa179077.
Impact: Preserve the FAIL decision, returned-versus-persisted Usage discrepancy, resource stop, and next design gate without rewriting the sample.
Refs: U3 r0.4 plan; Dev task feature-fast-slow-u2-u3-plan
Why: A validated sensitive-text persistence defect requires a reviewed X1 source reauthorization contract.
Evidence: User approved the immutable r1 design at its recorded SHA256; the Backend RED remains preserved.
Impact: Allows isolated W6 implementation while retaining separate migration, tag and acceptance gates.
Refs: feature/rdm-two-trees-latest-templates W6
Why: Persisting goal-sensitive model prompts in long-lived context receipts violates the approved content-minimization boundary.
Evidence: A fixed Backend HEAD RED reproduces stored workflow output; source and migration constraints were reviewed against current code.
Impact: Adds a reviewable X1 interface and data correction proposal; implementation stays paused until human approval.
Refs: feature/rdm-two-trees-latest-templates W6
Why: W2 transaction consistency and PostgreSQL audit pagination need their reviewed dependency and data-order boundaries preserved.
Evidence: User approved the two immutable r1 objects at their recorded SHA256 values; local link and whitespace checks passed.
Impact: Main record now authorizes isolated W4a and W9 work while retaining separate live migration gates.
Refs: feature/rdm-two-trees-latest-templates stage 6
Why: PostgreSQL sequence IDs can become visible out of order and undermine fixed audit pagination.
Evidence: Isolated PostgreSQL late-commit reproduction and sequence cache behavior were checked against official documentation.
Impact: Adds a reviewable proposal with lock, sequence, migration, rollback, and throughput boundaries; no runtime code changes.
Refs: feature/rdm-two-trees-latest-templates W9
Why: The approved U6 plan requires a reviewable account of three real serial deliveries, CI costs, recovery paths, and manual adoption limits.
Evidence: Dev #489/#488/#486 fixed-head Actions runs 3038/3050/3067 and main runs 3040/3053/3068; component PRs and Issue #518 preserve raw SELF and independent REPLAY.
Impact: Document observed behavior and untested automatic coordination boundaries without changing product code or declaring rollout authority.
Refs: Dev issue #518; merge queue implementation plan U6
Why: Preserve the reviewed 44-template scope, design, implementation plan, and exact stage decisions in the shared project record.
Evidence: Approved stage hashes and local links verified; W0 inventory has 44 records and current implementation limits are explicit.
Impact: Adds traceable planning, source receipts, audit-link approval, and a pending W2/W4 sequence adjustment without changing published assets.
Refs: feature/rdm-two-trees-latest-templates
Why: Preserve approved proposal status updates while retaining the U3 r0.4 design and implementation plan.
Evidence: The two lines of work touch disjoint document paths; merge status is clean and the parent Dev composition will be revalidated.
Impact: Documentation only; no model or runtime behavior change.
Refs: Dev U3 r0.4 Task 6 synchronization
Why: Keep the planning index aligned with the approved scope and actual U1 results.
Evidence: The RDM branch saved the 44-asset catalog and E1 receipts; exact candidate fidelity remains under human review.
Impact: Documentation status only; no template is published and no runtime behavior changes here.
Refs: DHF plan r0.1; RDM feature/dhf-template-assets@6af4497
Why: Keep the owner-reviewed scope and asset evidence available for implementation.
Evidence: Plan review and SELF document/source receipts are recorded in the Dev task card.
Impact: Documentation and asset inventory only; no product behavior changes.
Refs: DHF plan r0.1; owner approval to begin implementation
Why: turn the approved recovery design into reviewable cross-component implementation and validation steps.
Evidence: six tasks and thirty steps mapped to the spec; review-focus tests, local links and diff checks pass.
Impact: plan and validation status only; code and paid model execution remain NOT_RUN.
Refs: docs/proposals/2026-10-14-fast-slow-u3-r04-reliable-first-checkpoint-implementation-plan.md; docs/reports/2026-10-13-fast-slow-u2-u3-validation.md
Why: refine the approved IAM scope into enforceable account, directory and recovery boundaries.
Evidence: SELF document checks and independent document reviews; business tests and CI are NOT_RUN.
Impact: proposed contract and implementation plan only; preserve the approved global plan and all runtime data.
Refs: Dev feature/iam-u3-account-directory-recovery; IAM r1 U3 and IAM-D/IAM-C
Why: the approved r0.3 A01 sample failed before any useful background checkpoint.
Evidence: preserved A01 receipt, SQLite and provider records; offline SSE and retrieval replay; local links and diff checks pass.
Impact: design and validation record only; no behavior, budget, or model route changes.
Refs: docs/proposals/2026-10-14-fast-slow-u3-r04-reliable-first-checkpoint-design.md; docs/reports/2026-10-13-fast-slow-u2-u3-validation.md
Why: old r2 and r3 draft-stage paragraphs contradicted the retained approval records and current user takeover authorization.
Evidence: independent Documents review identified the ambiguity; targeted diff and link checks are captured outside the candidate.
Impact: label historical planning limits without changing genuine NOT_RUN results, deployment boundaries, or the newer login recovery implementation record.
Refs: Documents PR #89; Dev PR #489; independent review /tmp/mq-trial489-docs-independent-review-20261014.json
Why: retain the approved recovery and account additions while preserving the newer login recovery record and all current documentation.
Evidence: diff checks and affected links are captured by the evidence tool; historical CI remains scoped to its original object and is not current admission.
Impact: merge the current document target into the existing status task, preserve r3 records, and apply the user mandate that agents execute merges.
Refs: Documents PR #89; Dev PR #489; merge queue Issue #518; user continuation authorization on 2026-10-14
Why: Reviewers need the exact successful combination and a clear distinction between prepared evaluation code and paid authorization.
Evidence: Dev 6f91ca9 make test/build SELF receipt; Backend 5f83454 label regression RED→GREEN.
Impact: Documentation only; real-model quality remains NOT_RUN and six PRs stay drafts.
Refs: Dev task feature/fast-slow-u2-u3-plan; Documents PR #95
Why: Repeated independent semantic review required a conservative final source and analysis boundary.
Evidence: Backend 5f53f80 independent replay found no remaining Critical/Important; task-owned PostgreSQL and HTTP browser receipts identify separate fixed SHAs.
Impact: Code mechanism review is PASS, while real-model quality, exact Dev combination and platform gates remain pending.
Refs: Dev task feature/fast-slow-u2-u3-plan; Documents PR #95
Why: The first r0.3 code review found three unsafe inference paths despite green mechanism tests.
Evidence: Reviewer's read-only replay, Backend RED→GREEN counterexamples, isolated PostgreSQL and task-owned HTTP browser receipts are identified separately.
Impact: Keep the old review failure and new fix candidate traceable; real-model quality stays NOT_RUN.
Refs: Dev task feature/fast-slow-u2-u3-plan; Documents PR #95
Why: The new evidence-bound checkpoint candidate needs a reviewable account of what the local and browser gates prove.
Evidence: Tool-imported Dev combination, isolated PostgreSQL and real HTTP fake-stream browser receipts; previous knip failure retained.
Impact: Documentation only; r0.2 model quality remains FAIL and r0.3 paid A01 remains NOT_RUN.
Refs: Dev task feature/fast-slow-u2-u3-plan; Documents PR #95
Why: Preserve delegated project settings while keeping physical Host mutations owner-scoped.
Evidence: User explicitly approved the independently reviewed amendment and repair plan; original amendment digest recorded in the contract.
Impact: Clarify actual-actor authorization and audit, single-target owner projection, durable recovery and masked metadata; no provider implementation or deployment claim.
Refs: IAM I1 host projection amendment r1; Documents PR 98; Dev policy/delivery.md affected stages 4-5.
Why: The user approved zero-paid execution of the written plan.
Evidence: Approval response on 2026-10-13; git diff --cached --check passed.
Impact: Document status only; new paid A01 remains separately gated.
Refs: Documents PR #95; Dev task feature/fast-slow-u2-u3-plan
weichen
