Архитектура: централизовать navigation focus authority в continuity policy и сделать phase15 replay time-stable

2026-04-19 17:44:37 +03:00
parent 0e098cde38
commit 82cc5cfd1a
10 changed files with 199 additions and 92 deletions
@@ -23,7 +23,7 @@ This snapshot is based on:
 - live replay comparison between:
  - `address_truth_harness_phase12_wider_saved_session_pool_live_20260419_rerun16`
  - `address_truth_harness_phase14_counterparty_tail_resume_live_20260418_rerun2`
-  - `address_truth_harness_phase15_answer_inspection_followup_live_20260418_rerun8`
+  - `address_truth_harness_phase15_answer_inspection_followup_live_20260419_rerun11`
  - `address_truth_harness_phase16_multicompany_late_pivot_live_20260419_rerun10`
  - `address_truth_harness_phase17_clarification_resume_and_counterparty_tail_live_20260419_rerun5`
 - [10 - regression_breakpoint_analysis_2026-04-17.md](./10%20-%20regression_breakpoint_analysis_2026-04-17.md)
@@ -33,8 +33,8 @@ This snapshot is based on:

 Latest graph rebuild:

- `5371 nodes`
- `11523 edges`
+- `5372 nodes`
+- `11525 edges`
 - `135 communities`

 Most relevant current god nodes for turnaround `11`:
@@ -520,6 +520,12 @@ Latest phase17 clarification-resume evidence after the current replay hardening:
  - `decomposeStage` no longer hydrates inventory follow-up filters with the selected organization alias as a fake counterparty anchor, so hidden carryover state stays truthful;
 - targeted `assistantRoutePolicy`, `assistantAddressFollowupContext`, `addressImplicitOrganizationScope`, and `addressFollowupTemporalRegression` suites are green after the fix, and backend build stays green;
 - live replay `address_truth_harness_phase17_clarification_resume_and_counterparty_tail_live_20260419_rerun5` is accepted `10/10`, which is the current proof that clarification-resume, historical inventory continuation, and short counterparty-tail retarget are now semantically clean on a non-flagship saved-session path.
+- the next continuity-authority pass now centralizes navigation focus-state parsing for the hot selected-object path:
+  - `assistantContinuityPolicy` now owns `resolveNavigationSessionContextState(...)`, which extracts `date scope`, `organization`, `active focus object`, and `active result set id` from navigation state through one shared helper;
+  - `assistantTransitionPolicy` no longer reconstructs `session_context.active_focus_object` in multiple local branches for purchase-date VAT bridge detection and selected-item carryover setup;
+  - this matters because selected-item continuity, purchase-date VAT bridge, and navigation-driven focus-object hints now read one shared navigation-state interpretation instead of three local parsers inside the transition hot path;
+  - targeted `assistantContinuityPolicy` and `assistantTransitionPolicy` suites are green after the move (`37/37`), backend build is green, and graphify was rebuilt on the updated codebase;
+  - phase15 replay has also been made time-stable for the current snapshot step, and live replay `address_truth_harness_phase15_answer_inspection_followup_live_20260419_rerun11` is accepted `9/9`, which is the semantic proof that the selected-item / answer-inspection / VAT bridge contour still survives after the focus-authority convergence pass.

 ## Next Execution Slice (2026-04-19)

@@ -558,6 +564,34 @@ Current remaining heavy fronts before low-risk domain expansion:
 - complete the missing contour for counterparty shipped-goods / service extraction instead of relying on honest-but-limited document-list fallback;
 - keep answer shaping as secondary debt only where it materially affects acceptance, not as the primary architecture frontier.

+Execution framing for the next level:
+
+- the working target is now `90%+ pre-multidomain readiness` before controlled domain expansion begins;
+- the current honest level is `~78%`, so the remaining gap should be treated as four large iterations, not as a few cosmetic follow-ups.
+
+Current four-iteration plan:
+
+1. `Iteration 1 / Continuity authority completion`
+   Goal:
+   finish convergence toward one owner for `organization / date / root frame / focus object / clarification state` across the hot runtime path.
+   Expected gain:
+   `~+4%`.
+2. `Iteration 2 / Wider saved-session replay pool`
+   Goal:
+   widen replay breadth beyond the current flagship + phase14 + phase15 + phase16 + phase17 family set.
+   Expected gain:
+   `~+4%`.
+3. `Iteration 3 / Coordinator pressure reduction`
+   Goal:
+   reduce control-plane overload in `assistantService.ts`, `addressQueryService.ts`, and adjacent top-level orchestration seams.
+   Expected gain:
+   `~+2-3%`.
+4. `Iteration 4 / Critical domain-enablement gaps + final replay proof`
+   Goal:
+   close the remaining high-value business gaps such as `counterparty shipped goods/service` and confirm the result under replay until the system crosses the `90%` threshold honestly.
+   Expected gain:
+   `~+1-2%`.
+
 ## Ready Signal

 The project can leave the current breakpoint when:
@@ -26,7 +26,7 @@ Current confidence snapshot:
 - turnaround implementation progress: `~96%`
 - exit-from-danger-zone readiness: `~91%`
 - pre-multidomain readiness: `~78%`
- latest graph snapshot: `5371 nodes`, `11523 edges`, `135 communities`
+- latest graph snapshot: `5372 nodes`, `11525 edges`, `135 communities`

 ## What Is Already True

@@ -139,7 +139,13 @@ The system should not be considered ready for the next level until all of the fo

 ## Recommended Next Execution Sequence

-### Pass 18. Continuity authority completion
+The current planning assumption is:
+
+- `90%+ pre-multidomain readiness` is the practical threshold for starting controlled domain expansion;
+- the project is currently around `78%`;
+- therefore the remaining gap should be treated as roughly four large iterations rather than as a handful of cosmetic fixes.
+
+### Iteration 1 / Pass 18. Continuity authority completion

 Goal:

@@ -149,7 +155,11 @@ Target:

 - transition / route / clarification should consume one continuity snapshot before making divergent decisions.

-### Pass 19. Wider saved-session acceptance pool
+Expected readiness gain:
+
+- `~+4%` toward pre-multidomain readiness.
+
+### Iteration 2 / Pass 19. Wider saved-session acceptance pool

 Goal:

@@ -159,17 +169,11 @@ Target:

 - several saved sessions covering inventory, VAT, counterparty, payables/receivables, meta interrupts, and cross-domain pivots beyond phase12 / phase16 / phase17.

-### Pass 20. Human answer shaping cleanup
+Expected readiness gain:

-Goal:
+- `~+4%` toward pre-multidomain readiness.

- remove the remaining mechanical, template-heavy feel from long exact answers.
-
-Target:
-
- product-quality business answers on already-correct truth paths.
-
-### Pass 21. Coordinator pressure reduction
+### Iteration 3 / Pass 20. Coordinator pressure reduction

 Goal:

@@ -177,7 +181,30 @@ Goal:

 Target:

- less policy/service glue concentrated in `assistantService.ts` and adjacent god-modules.
+- less policy/service glue concentrated in `assistantService.ts`, `addressQueryService.ts`, and adjacent coordinator-heavy seams.
+
+Expected readiness gain:
+
+- `~+2-3%` toward pre-multidomain readiness.
+
+### Iteration 4 / Pass 21. Critical domain-enablement gaps and final acceptance proof
+
+Goal:
+
+- close the remaining high-value domain gaps that still block truthful expansion claims and then prove the result on replay.
+
+Target:
+
+- missing business contours such as `counterparty shipped goods/service`, plus the final replay confirmation that the system is over the `90%` threshold.
+
+Expected readiness gain:
+
+- `~+1-2%` toward pre-multidomain readiness.
+
+Secondary cleanup after those four iterations:
+
+- human answer shaping cleanup where it materially affects acceptance;
+- further quality polish that should not be confused with the core readiness gate.

 ## Final Statement

@@ -48,7 +48,7 @@ Current honest status:
 - turnaround implementation progress: `~96%`
 - exit-from-danger-zone readiness: `~91%`
 - pre-multidomain readiness: `~78%`
- graph snapshot after latest rebuild: `5371 nodes`, `11523 edges`, `135 communities`
+- graph snapshot after latest rebuild: `5372 nodes`, `11525 edges`, `135 communities`
 - current breakpoint:
  - the validated hot paths are no longer structurally broken;
  - flagship continuity collapse is no longer the primary risk;
@@ -65,7 +65,7 @@ Latest live proof now includes:

 - `address_truth_harness_phase12_wider_saved_session_pool_live_20260419_rerun16` accepted `20/20`
 - `address_truth_harness_phase14_counterparty_tail_resume_live_20260418_rerun2` accepted `10/10`
- `address_truth_harness_phase15_answer_inspection_followup_live_20260418_rerun8` accepted `9/9`
+- `address_truth_harness_phase15_answer_inspection_followup_live_20260419_rerun11` accepted `9/9`
 - `address_truth_harness_phase16_multicompany_late_pivot_live_20260419_rerun10` accepted
 - `address_truth_harness_phase17_clarification_resume_and_counterparty_tail_live_20260419_rerun5` accepted `10/10`

@@ -74,6 +74,7 @@ Current architectural reading:
 - the system is already materially past the dangerous regression breakpoint;
 - it is now safe for continued architecture hardening and controlled domain-by-domain enablement under replay gates;
 - it is now materially closer to pre-multidomain stability, but still not safe to declare broad low-risk multi-domain expansion.
+- the practical next target is now `90%+ pre-multidomain readiness`, and the remaining gap should be treated as four large architecture iterations rather than as cosmetic cleanup.

 For the detailed audit, current percentages, and remaining debt, read:

@@ -50,12 +50,12 @@
      ],
      "expected_recipe": "address_inventory_on_hand_as_of_date_v1",
      "required_filters": {
-        "as_of_date": "2026-04-18",
+        "as_of_date": "{{runtime.today_iso}}",
        "organization": "ООО Альтернатива Плюс"
      },
      "required_direct_answer_patterns_any": [
        "(?i)на складе|остат",
-        "18\\.04\\.2026"
+        "{{runtime.today_dot_regex}}"
      ],
      "criticality": "critical",
      "semantic_tags": [