Open-World: расширить бизнес-обзор факт-семействами
This commit is contained in:
@@ -15,8 +15,14 @@ If another document says `78%`, `87%`, `92%`, or `85%` for a module that is now
|
||||
- `Planner Autonomy Consolidation`: `100%` for the declared phase83 planner-brain slice, including catalog alignment, live-readiness gating, checked-source sanitation, and accepted mixed replay.
|
||||
- Active next module: broader `Open-World Bounded Autonomy Breadth` over unfamiliar 1C asks, while keeping Post-F and phase83 as regression gates.
|
||||
- Completed active slice: `Business Overview Evidence Fusion`, tracked in `22 - open_world_bounded_autonomy_breadth_2026-05-01.md`.
|
||||
- Current active slice: `Business Overview Catalog Route Fabric`: the route is reviewed in catalog/data-need/planner contracts, while fresh multi-probe runtime execution remains a pending bridge.
|
||||
- Active module progress: `~18% (Open-World Bounded Autonomy Breadth)`.
|
||||
- Completed active slice: `Business Overview Catalog Route Fabric`: the route is reviewed in catalog/data-need/planner contracts and exposes the stable `business_overview` route scope.
|
||||
- Completed active slice: `Business Overview Fresh Multi-Probe Runtime Bridge`: the reviewed route now executes incoming money flow, outgoing supplier payout, activity-window, net-spread, top-customer, and analyst-safe answer drafting, and has passed live semantic replay against the real assistant runtime.
|
||||
- Completed active slice: `Business Overview VAT/Tax Fact-Family Bridge`: explicit-period business overview can include confirmed VAT/tax position, while all-time follow-ups and negated VAT periods do not reuse stale tax scope.
|
||||
- Completed active slice: `Business Overview Debt-Position Fact-Family Bridge`: explicit-period business overview can include confirmed receivables/payables as-of-date debt position, while all-time follow-ups do not reuse stale debt snapshots and debt quality/aging remains unclaimed.
|
||||
- Completed active slice: `Business Overview Inventory-Position Fact-Family Bridge`: explicit-date business overview can include confirmed stock-on-hand inventory position, while all-time follow-ups do not reuse stale inventory snapshots and inventory liquidity/turnover remains unclaimed.
|
||||
- Completed active slice: `Business Overview Open-Settlement Quality Bridge`: explicit-period business overview can check open-contract settlement concentration on 60/62/76, while due-date aging/overdue debt remains unclaimed until a reviewed due-date route exists.
|
||||
- Next active slice: continue `Business Overview Fact-Family Expansion` into profit/margin and due-date debt aging where reviewed routes exist.
|
||||
- Active module progress: `~60% (Open-World Bounded Autonomy Breadth)`.
|
||||
|
||||
## Reporting Rule
|
||||
|
||||
@@ -53,7 +59,7 @@ The project is not yet a universal arbitrary-1C agent.
|
||||
|
||||
Remaining work belongs to the next breadth module:
|
||||
|
||||
- implement the fresh multi-probe `business_overview` runtime bridge behind the reviewed route-fabric contract;
|
||||
- extend `business_overview` beyond money-flow/activity, explicit-period VAT/tax, as-of-date debt position, open-settlement concentration, and as-of-date inventory position into separately proven profit/margin, due-date debt aging/overdue, and real inventory-liquidity evidence families;
|
||||
- broader dynamic schema traversal for unfamiliar 1C asks;
|
||||
- more primitive descriptors where live evidence proves a real gap;
|
||||
- more replay-backed domain packs that start from user business meaning, not from route convenience;
|
||||
|
||||
+147
-10
@@ -82,18 +82,123 @@ Implemented now:
|
||||
- the data-need graph recognizes broad company analysis as a bounded business-overview evidence need;
|
||||
- fresh business-overview probes require an organization scope instead of silently reusing stale context;
|
||||
- planner output can select `business_overview` as the catalog top match with structured alignment telemetry;
|
||||
- the pilot executor exposes `business_overview_route_template_v1` as an explicit scope, but returns an unsupported runtime boundary until the fresh multi-probe bridge is implemented.
|
||||
- the pilot executor exposes `business_overview_route_template_v1` as the stable runtime scope for the next bridge.
|
||||
|
||||
This is deliberately not a fake runtime success.
|
||||
This slice was deliberately only route fabric: it made the reviewed route visible without pretending fresh runtime evidence existed yet.
|
||||
|
||||
The assistant now has the route-fabric contract for the next slice, while the live business overview still uses the safe evidence-fusion bridge from Slice 1.
|
||||
The assistant now has the route-fabric contract used by Slice 3.
|
||||
|
||||
### Still Pending Runtime Slice
|
||||
## Slice 3 - Business Overview Fresh Multi-Probe Runtime Bridge
|
||||
|
||||
Promote this bridge into a real planner route:
|
||||
This slice makes the reviewed `business_overview` route execute as a bounded multi-probe MCP discovery bridge.
|
||||
|
||||
- run bounded fresh probes for year turnover, top customers, incoming/outgoing/net flow, debt, VAT, and inventory context where available;
|
||||
- return a layered analyst answer with exact evidence, bounded inference, unknowns, and recommended next probes.
|
||||
Implemented now:
|
||||
|
||||
- broad company-analysis turn meaning routes into `business overview evidence with bounded analyst interpretation` instead of being kept only in deterministic living chat;
|
||||
- the bridge runs fresh scoped probes for incoming customer money flow, outgoing supplier payouts, and activity-window evidence;
|
||||
- the pilot derives checked incoming total, checked outgoing total, net cash-flow spread, net direction, top confirmed customer, and activity window;
|
||||
- the answer adapter turns those derived facts into a layered analyst-safe draft: confirmed facts, bounded interpretation, unknowns, and next probes;
|
||||
- response policy can replace the old deterministic broad summary only when the discovery candidate has grounded text, while clarification candidates still preserve the safe deterministic answer;
|
||||
- user-facing answers must not expose `business_overview_route_template_v1`, MCP primitive names, raw planner/debug labels, or profit/margin claims.
|
||||
|
||||
Live semantic replay is now accepted for this slice:
|
||||
|
||||
- `address_truth_harness_phase84_business_overview_multi_probe_bridge_live_20260503_runtime_bridge2` passed `3/3`;
|
||||
- step 1 proves explicit company business overview for `ООО Альтернатива Плюс`;
|
||||
- step 2 proves an exact `Группа СВК` 2020 net-flow follow-up after the company overview without stale organization-scope contamination;
|
||||
- step 3 proves returning to `ООО Альтернатива Плюс` business overview after a counterparty pivot without treating net cash-flow as profit.
|
||||
|
||||
## Slice 4 - Business Overview VAT/Tax Fact-Family Bridge
|
||||
|
||||
This slice adds the first separately checked fact family beyond money-flow/activity.
|
||||
|
||||
Implemented now:
|
||||
|
||||
- explicit-period business overview can select the reviewed VAT/tax recipe and execute the tax probe beside the money-flow/activity probes;
|
||||
- the pilot derives sales VAT, purchase/deduction VAT, net VAT direction, and the checked tax period only from confirmed VAT rows;
|
||||
- the answer adapter can surface VAT/tax position as a confirmed line without treating net cash-flow as profit or margin;
|
||||
- broad all-time business overview does not silently reuse a prior or negated VAT period;
|
||||
- wording such as `do not carry VAT for 2020` / `не тащи НДС за 2020` is treated as a temporal exclusion, not as the active period for the current turn;
|
||||
- organization extraction strips trailing all-time/business-overview clauses so the company scope remains the company name, not the whole user sentence.
|
||||
|
||||
Live semantic replay is accepted for this slice:
|
||||
|
||||
- `address_truth_harness_phase85_business_overview_tax_family_live_20260504_taxfamily2` passed `2/2`;
|
||||
- step 1 proves explicit 2020 business overview may include a confirmed VAT/tax position;
|
||||
- step 2 proves all-time follow-up over the same company does not reuse the 2020 VAT/tax position from the prior turn or from the negated wording;
|
||||
- the accepted debug path shows `explicit_date_scope=null`, `tax_position=null`, and `pilot_business_overview_tax_probe_skipped_without_explicit_period` for the all-time follow-up.
|
||||
|
||||
## Slice 5 - Business Overview Debt-Position Fact-Family Bridge
|
||||
|
||||
This slice adds a second separately checked fact family, but deliberately stops short of debt quality.
|
||||
|
||||
Implemented now:
|
||||
|
||||
- explicit-period business overview can derive an as-of-date from the user-visible period and execute reviewed receivables/payables snapshot recipes beside money-flow/activity and optional VAT/tax probes;
|
||||
- the pilot derives receivables, payables, net debt-position amount, net direction, and top debt-side counterparties only from confirmed 1C balance rows;
|
||||
- the answer adapter can surface the debt-position snapshot as a confirmed line without treating it as overdue debt, debt aging, credit quality, profit, or margin;
|
||||
- all-time business overview does not reuse the prior as-of-date debt snapshot and instead keeps debt position as an unknown fact family until the user gives a new explicit date;
|
||||
- raw organization extraction now strips trailing explicit period clauses such as `за 2020 год` / `на 2020-12-31`, so the company scope remains the company name and the period remains a separate temporal axis.
|
||||
|
||||
Live semantic replay is accepted for this slice:
|
||||
|
||||
- `address_truth_harness_phase86_business_overview_debt_position_live_20260504_debt2` passed `2/2`;
|
||||
- step 1 proves explicit 2020 business overview may include a confirmed receivables/payables debt-position snapshot on `2020-12-31`;
|
||||
- step 2 proves an all-time follow-up over the same company does not reuse the `2020-12-31` debt snapshot as current or all-time debt position;
|
||||
- the accepted debug path shows `organization_scope=ООО Альтернатива Плюс`, `explicit_date_scope=2020`, `debt_position.as_of_date=2020-12-31` for the explicit-period step, and `explicit_date_scope=null`, `debt_position=null`, `pilot_business_overview_debt_probe_skipped_without_explicit_as_of_date` for the all-time follow-up.
|
||||
|
||||
## Slice 6 - Business Overview Inventory-Position Fact-Family Bridge
|
||||
|
||||
This slice adds a third separately checked fact family, but deliberately stops short of warehouse liquidity or turnover.
|
||||
|
||||
Implemented now:
|
||||
|
||||
- explicit-date business overview can derive an as-of date and execute reviewed inventory on-hand and optional purchase-date aging probes beside the existing money-flow/activity, VAT/tax, and debt-position probes;
|
||||
- the pilot derives inventory on-hand rows, rows with amount, rows with quantity, total stock amount, total quantity, top stock items, and optional purchase-date aging signal only from confirmed 1C inventory rows;
|
||||
- the answer adapter can surface the inventory-position snapshot as a confirmed line without treating it as turnover, obsolescence, liquidation value, full inventory health, profit, or margin;
|
||||
- if the current business-overview slice has no incoming/outgoing money rows, the answer no longer emits a fake `net 0` cash-flow interpretation;
|
||||
- business-overview headlines now list only the fact families that were actually confirmed in the current turn, so a stock/debt/tax snapshot does not claim money-flow or activity when those rows were absent;
|
||||
- all-time business overview does not reuse a prior as-of-date inventory snapshot and instead keeps stock/inventory position as an unknown fact family until the user gives a new explicit date.
|
||||
|
||||
Live semantic replay is accepted for this slice:
|
||||
|
||||
- `address_truth_harness_phase87_business_overview_inventory_position_live_20260504_inventory2` passed `2/2`;
|
||||
- step 1 proves explicit-date business overview may include a confirmed warehouse stock snapshot on `2026-04-16`;
|
||||
- step 2 proves an all-time follow-up over the same company does not reuse the `2026-04-16` stock snapshot as current or all-time warehouse position;
|
||||
- the accepted user-facing answer confirms stock amount `716 418,33` rub. over `11` rows and keeps warehouse liquidity/turnover as unconfirmed.
|
||||
|
||||
## Slice 7 - Business Overview Open-Settlement Quality Bridge
|
||||
|
||||
This slice widens the debt family from a simple receivables/payables position into a bounded quality signal, but deliberately stops short of due-date aging or confirmed overdue debt.
|
||||
|
||||
Implemented now:
|
||||
|
||||
- explicit-period business overview can execute the reviewed `open_contracts_confirmed_as_of_date` balance recipe beside money-flow/activity, VAT/tax, debt-position, and inventory probes;
|
||||
- the pilot derives gross open settlement amount, unique open-contract count, unique counterparty count, top open contracts, top counterparties, and concentration percentages only from confirmed 1C balance rows on 60/62/76;
|
||||
- the answer adapter can surface this as `качество открытых расчетов` without treating concentration as contractual due-date aging, confirmed overdue debt, credit risk, profit, or margin;
|
||||
- all-time business overview keeps open-settlement quality unknown unless the current turn has a fresh explicit as-of date;
|
||||
- the remaining hard boundary is still due-date/overdue aging: open contracts prove concentration of open balances, not payment-term delinquency.
|
||||
|
||||
Live semantic replay is accepted for this slice:
|
||||
|
||||
- `npm.cmd test -- assistantMcpDiscoveryPilotExecutor.test.ts`: passed `31/31`;
|
||||
- `npm.cmd test -- assistantMcpDiscoveryAnswerAdapter.test.ts`: passed `34/34` with `1` skipped;
|
||||
- `npm.cmd test -- assistantMcp`: passed `305/305` with `9` skipped;
|
||||
- `npm.cmd run build`: passed;
|
||||
- `address_truth_harness_phase88_business_overview_open_settlement_quality_live_20260504_openquality4` passed `2/2`;
|
||||
- step 1 proves explicit 2020 business overview may include confirmed open-settlement concentration: `35 472 380,36` rub. gross open contract balances, `8` contracts, `11` counterparties, and top contract share `54.2%`;
|
||||
- step 2 proves an all-time follow-up does not reuse the `2020-12-31` open-contract/debt snapshot as current overdue debt or all-time debt quality;
|
||||
- the accepted user-facing answer keeps due-date aging and overdue debt as unconfirmed because open contracts prove concentration of balances, not payment-term delinquency.
|
||||
|
||||
### Still Pending Breadth Slices
|
||||
|
||||
Grow this bridge beyond the first confirmed signal bundle:
|
||||
|
||||
- add separate evidence families for profit/margin and due-date debt aging/overdue quality where reviewed routes exist;
|
||||
- extend inventory evidence from as-of-date stock position into real turnover/liquidity only when reviewed sales velocity, aging, or obsolescence evidence exists;
|
||||
- extend debt evidence from as-of-date position/open-settlement concentration into overdue aging only when reviewed due-date or aging evidence exists;
|
||||
- extend VAT/tax beyond explicit-period tax position only when the requested tax fact is provable and the period is explicit;
|
||||
- keep Post-F stale-scope and phase83 catalog-alignment canaries green while widening the route.
|
||||
|
||||
## Acceptance Signals
|
||||
|
||||
@@ -112,11 +217,43 @@ Initial local validation:
|
||||
- `npm.cmd test -- assistantTurnMeaningPolicy.test.ts assistantLivingChatRuntimeAdapter.test.ts`: passed `20/20`.
|
||||
- `npm.cmd test -- assistantTurnMeaningPolicy.test.ts assistantLivingChatRuntimeAdapter.test.ts assistantRoutePolicy.test.ts assistantMcpDiscoveryResponsePolicy.test.ts`: passed `56/56`.
|
||||
- `npm.cmd run build`: passed.
|
||||
- graphify rebuild: `5977 nodes`, `12983 edges`, `137 communities`.
|
||||
- graphify rebuild at Slice 2 boundary: `5977 nodes`, `12983 edges`, `137 communities`.
|
||||
|
||||
Business-overview route-fabric validation:
|
||||
|
||||
- `npm.cmd test -- assistantMcpCatalogIndex.test.ts assistantMcpDiscoveryDataNeedGraph.test.ts assistantMcpDiscoveryPlanner.test.ts assistantMcpDiscoveryPilotExecutor.test.ts`: passed `102/102`.
|
||||
- fresh multi-probe runtime execution remains intentionally pending behind `business_overview_route_template_v1`.
|
||||
|
||||
Graphify must be rebuilt after this code/doc slice before commit.
|
||||
Business-overview fresh runtime bridge validation:
|
||||
|
||||
- `npm.cmd test -- assistantMcpDiscoveryRuntimeEntryPoint.test.ts assistantMcpDiscoveryPilotExecutor.test.ts assistantMcpDiscoveryAnswerAdapter.test.ts assistantMcpDiscoveryTurnInputAdapter.test.ts assistantMcpDiscoveryResponsePolicy.test.ts assistantMcpDiscoveryPlanner.test.ts`: passed `211/211` with `9` skipped.
|
||||
- `npm.cmd test -- assistantMcp`: passed `296/296` with `9` skipped.
|
||||
- `npm.cmd run build`: passed.
|
||||
- live replay `address_truth_harness_phase84_business_overview_multi_probe_bridge_live_20260503_runtime_bridge2`: accepted `3/3` with `catalog_alignment_ok=true`, `human_answer_quality_ok=true`, and no internal route/debug terms in the user-facing answer.
|
||||
|
||||
Business-overview VAT/tax fact-family validation:
|
||||
|
||||
- `npm.cmd test -- assistantMcpDiscoveryTurnInputAdapter.test.ts`: passed `76/76` with `6` skipped.
|
||||
- `npm.cmd test -- assistantMcp`: passed `300/300` with `9` skipped.
|
||||
- `npm.cmd run build`: passed.
|
||||
- live replay `address_truth_harness_phase85_business_overview_tax_family_live_20260504_taxfamily2`: accepted `2/2` and proved that a negated 2020 VAT period is not reused as an all-time tax position.
|
||||
|
||||
Business-overview debt-position fact-family validation:
|
||||
|
||||
- `npm.cmd test -- assistantMcpDiscoveryPilotExecutor.test.ts`: passed `30/30`.
|
||||
- `npm.cmd test -- assistantMcpDiscoveryAnswerAdapter.test.ts`: passed `33/33` with `1` skipped.
|
||||
- `npm.cmd test -- assistantMcpDiscoveryTurnInputAdapter.test.ts`: passed `77/77` with `6` skipped.
|
||||
- `npm.cmd test -- assistantMcp`: passed `303/303` with `9` skipped.
|
||||
- `npm.cmd run build`: passed.
|
||||
- live replay `address_truth_harness_phase86_business_overview_debt_position_live_20260504_debt2`: accepted `2/2` and proved that explicit debt-position snapshots do not leak into all-time follow-ups.
|
||||
|
||||
Business-overview inventory-position fact-family validation:
|
||||
|
||||
- `npm.cmd test -- assistantMcpDiscoveryPilotExecutor.test.ts`: passed `31/31`.
|
||||
- `npm.cmd test -- assistantMcpDiscoveryAnswerAdapter.test.ts`: passed `34/34` with `1` skipped.
|
||||
- `npm.cmd test -- assistantMcp`: passed `305/305` with `9` skipped.
|
||||
- `npm.cmd run build`: passed.
|
||||
- live replay `address_truth_harness_phase87_business_overview_inventory_position_live_20260504_inventory2`: accepted `2/2` and proved that explicit inventory-position snapshots do not leak into all-time follow-ups.
|
||||
|
||||
Graphify rebuild after Slice 6 code/doc sync: `6001 nodes`, `13058 edges`, `140 communities`.
|
||||
|
||||
Graphify rebuild after Slice 7 code/doc sync: `6008 nodes`, `13078 edges`, `138 communities`.
|
||||
|
||||
@@ -41,7 +41,7 @@ This package answers the next question:
|
||||
21. [21 - current_status_canon_2026-05-01.md](./21%20-%20current_status_canon_2026-05-01.md)
|
||||
22. [22 - open_world_bounded_autonomy_breadth_2026-05-01.md](./22%20-%20open_world_bounded_autonomy_breadth_2026-05-01.md)
|
||||
|
||||
## Current Status Snapshot (2026-05-01)
|
||||
## Current Status Snapshot (2026-05-04)
|
||||
|
||||
This package is no longer planning-only.
|
||||
|
||||
@@ -51,7 +51,12 @@ Status canon for planning:
|
||||
- Planner Autonomy Consolidation is closed at `100%` for the declared phase83 planner-brain slice.
|
||||
- The active next module is now `Open-World Bounded Autonomy Breadth` over unfamiliar 1C asks, with Post-F and phase83 retained as semantic canaries.
|
||||
- The first active slice is `Business Overview Evidence Fusion`: broad company-analysis wording now produces a richer evidence-grounded business overview from confirmed MCP/session facts instead of a thin generic summary.
|
||||
- The current follow-up slice is `Business Overview Catalog Route Fabric`: `business_overview` is now a reviewed catalog/data-need/planner chain, while fresh multi-probe execution remains honestly bounded behind an unsupported runtime scope until the next bridge is implemented.
|
||||
- The current completed slice is `Business Overview Fresh Multi-Probe Runtime Bridge`: `business_overview` is now a reviewed catalog/data-need/planner chain and a live-replay accepted runtime bridge over incoming money flow, outgoing supplier payouts, activity-window evidence, net-spread, top customer, and analyst-safe answer drafting.
|
||||
- The current completed breadth slice is `Business Overview VAT/Tax Fact-Family Bridge`: explicit-period business overview can include confirmed VAT/tax position, while all-time follow-ups and negated VAT period wording do not reuse stale tax scope.
|
||||
- The current completed breadth slice is `Business Overview Debt-Position Fact-Family Bridge`: explicit-period business overview can include confirmed receivables/payables as-of-date debt position, while all-time follow-ups do not reuse stale debt snapshots and debt quality/aging remains unclaimed.
|
||||
- The current completed breadth slice is `Business Overview Inventory-Position Fact-Family Bridge`: explicit-date business overview can include confirmed stock-on-hand inventory position, while all-time follow-ups do not reuse stale inventory snapshots and inventory liquidity/turnover remains unclaimed.
|
||||
- The current completed breadth slice is `Business Overview Open-Settlement Quality Bridge`: explicit-period business overview can check open-contract settlement concentration, while due-date aging and confirmed overdue debt remain outside the answer until a reviewed due-date route exists.
|
||||
- The next active breadth slice continues `Business Overview Fact-Family Expansion` into profit/margin and due-date debt aging, then broader unfamiliar 1C route breadth without relaxing truth boundaries.
|
||||
- The short source of truth for status wording is [21 - current_status_canon_2026-05-01.md](./21%20-%20current_status_canon_2026-05-01.md).
|
||||
|
||||
It now documents a turnaround that is already operational in code, already materially past the acute regression breakpoint, and already moved through bounded MCP autonomy, Post-F hardening, inventory breadth proof, and the declared Planner Autonomy slice:
|
||||
@@ -86,7 +91,7 @@ It now documents a turnaround that is already operational in code, already mater
|
||||
- explicit document/movement data-need now scores over ambiguous carried metadata surfaces without forcing neutral follow-ups into a lane;
|
||||
- lifecycle now behaves as a bounded activity-window inference chain with an explicit legal-fact boundary instead of an unqualified age answer;
|
||||
- current-turn value-flow aggregate questions can override narrower supported exact routes when the user asks for totals/net/payment amounts;
|
||||
- broad business evaluation remains in the deterministic living-chat bridge instead of being displaced by generic metadata discovery;
|
||||
- broad business evaluation remained guarded during phase83 and is now carried forward in the next breadth module as a reviewed `business_overview` discovery route instead of being displaced by generic metadata discovery;
|
||||
- inventory stock snapshot, supplier overlap, purchase provenance, and sale trace are now reviewed catalog chain templates; generic free-form inventory execution remains forbidden, and evidence must pass through reviewed exact recipe bridges;
|
||||
- runtime bridge and answer adapter now keep unsupported inventory route templates behind an explicit user-facing boundary instead of letting template planning look like confirmed stock/supplier/purchase/sale evidence;
|
||||
- inventory catalog templates now bridge through existing exact inventory recipes (`41.01` scoped stock, supplier overlap, purchase provenance, and sale trace) inside the bounded MCP discovery pilot, while missing selected-item anchors still clarify instead of guessing;
|
||||
@@ -115,12 +120,12 @@ Current honest status:
|
||||
- exit-from-danger-zone readiness: `~97%`
|
||||
- pre-multidomain readiness: `~90%`
|
||||
- bounded-autonomy foundation readiness: `~89%`
|
||||
- open-world bounded-autonomy readiness: `~86%`
|
||||
- active Open-World Bounded Autonomy Breadth progress: `~18%`, with business-overview evidence fusion and the reviewed `business_overview` catalog/data-need/planner route-fabric slice locally tested; fresh multi-probe runtime execution is still pending
|
||||
- open-world bounded-autonomy readiness: `~87%`
|
||||
- active Open-World Bounded Autonomy Breadth progress: `~60%`, with business-overview evidence fusion, the reviewed `business_overview` catalog/data-need/planner route-fabric slice, the fresh multi-probe runtime bridge, the explicit-period VAT/tax fact-family bridge, the explicit-period debt-position bridge, the explicit-date inventory-position bridge, and the open-settlement quality bridge accepted by live semantic replay; profit/margin, due-date debt aging/overdue, and real inventory-liquidity expansion are still pending
|
||||
- Post-F semantic integrity module progress: `~99%` operationally closed, with remaining risk now treated as next-slice discovery rather than an open blocker inside the closed slice
|
||||
- active inventory-stock breadth slice progress: `100%` for the declared scenario pack, not for arbitrary inventory questions
|
||||
- Planner Autonomy Consolidation progress: `100%` for the declared module, with catalog-fabric, value-flow arbitration, lifecycle bounded inference, broad-evaluation bridge, inventory catalog templates, inventory runtime-boundary honesty, exact inventory recipe bridging, unambiguous metadata-surface lane inference, catalog chain-template scoring, structured chain-match contract exposure, runtime/debug propagation, subject-aware bidirectional comparison arbitration, structured catalog-alignment verdicts, representative alignment regression guard, catalog-alignment reason-code telemetry, explicit `alignment_status` propagation, truth-harness/acceptance-matrix surfacing, soft divergence warning, `catalog_alignment_ok` acceptance invariant, step-level expected catalog-alignment assertions, phase66 and phase32 spec alignment expectations, AGENT source-catalog surfacing, generated phase83 mixed planner-brain replay spec, checked-source user-facing error sanitation, surface-grounded catalog promotion, and guarded live phase83 acceptance validated. Broader unfamiliar 1C asks are now next-module breadth work rather than an open blocker inside this declared slice
|
||||
- graph snapshot after latest rebuild: `5977 nodes`, `12983 edges`, `137 communities`
|
||||
- graph snapshot after latest rebuild: `6008 nodes`, `13078 edges`, `138 communities`
|
||||
- current regression-gate breakpoint:
|
||||
- the validated hot paths are no longer structurally broken;
|
||||
- flagship continuity collapse is no longer the primary risk;
|
||||
@@ -160,7 +165,13 @@ Latest live proof now includes:
|
||||
- lifecycle/value-flow Planner Autonomy response gate accepted: `address_truth_harness_phase19_mcp_discovery_response_gate_planner_lifecycle_rerun4` accepted `8/8`, proving bounded lifecycle inference, current-turn value-flow aggregate arbitration, and sanitized evidence wording
|
||||
- broad-evaluation bridge continuity accepted: `address_truth_harness_phase21_net_followup_after_broad_eval_planner_lifecycle_rerun2` accepted `3/3` and `address_truth_harness_phase22_broad_business_evaluation_bridge_planner_lifecycle_rerun2` accepted `3/3`
|
||||
- latest local Planner Autonomy slice accepted: full MCP-discovery suite passed `268/268` with `9` skipped; broad MCP/living-chat/route/meaning slice passed `305/305` with `9` skipped; build passed
|
||||
- business-overview route-fabric slice accepted locally: catalog/data-need/planner/pilot boundary slice passed `102/102`; the pilot executor exposes `business_overview_route_template_v1` as an explicit unsupported scope until the fresh multi-probe bridge exists
|
||||
- business-overview route-fabric slice accepted locally: catalog/data-need/planner/pilot boundary slice passed `102/102`, proving the reviewed `business_overview` chain and stable route scope
|
||||
- business-overview fresh multi-probe runtime bridge accepted locally: targeted runtime-entry/pilot/answer/turn-input/response-policy/planner slice passed `211/211` with `9` skipped; full MCP-discovery suite passed `296/296` with `9` skipped; build passed
|
||||
- business-overview fresh multi-probe runtime bridge accepted live: `address_truth_harness_phase84_business_overview_multi_probe_bridge_live_20260503_runtime_bridge2` accepted `3/3`, proving explicit company overview, exact counterparty net-flow after the company overview, and explicit company overview after a counterparty pivot with `catalog_alignment_ok=true`, `human_answer_quality_ok=true`, and no internal route/debug leakage in the user-facing answer
|
||||
- business-overview VAT/tax fact-family bridge accepted live: `address_truth_harness_phase85_business_overview_tax_family_live_20260504_taxfamily2` accepted `2/2`, proving explicit-period VAT/tax position in the company overview and all-time follow-up protection against stale or negated VAT-period reuse
|
||||
- business-overview debt-position fact-family bridge accepted live: `address_truth_harness_phase86_business_overview_debt_position_live_20260504_debt2` accepted `2/2`, proving explicit-period receivables/payables as-of-date debt position and all-time follow-up protection against stale debt snapshot reuse
|
||||
- business-overview inventory-position fact-family bridge accepted live: `address_truth_harness_phase87_business_overview_inventory_position_live_20260504_inventory2` accepted `2/2`, proving explicit-date stock-on-hand position and all-time follow-up protection against stale inventory snapshot reuse
|
||||
- business-overview open-settlement quality bridge accepted live: `address_truth_harness_phase88_business_overview_open_settlement_quality_live_20260504_openquality4` accepted `2/2`, proving explicit-period open-contract concentration and all-time follow-up protection against stale open-contract/debt-quality reuse
|
||||
- inventory template lift accepted locally: catalog/data-need/planner/turn-input slice passed `139/139` with `6` skipped; full MCP-discovery slice passed `276/276` with `9` skipped; build passed; graphify stayed at `5912 nodes`, `12833 edges`, `138 communities`
|
||||
- inventory runtime-boundary hardening accepted locally: runtime-bridge/answer-adapter/pilot-executor slice passed `68/68` with `1` skipped; full MCP-discovery slice passed `277/277` with `9` skipped; build passed; graphify rebuilt to `5913 nodes`, `12837 edges`, `138 communities`
|
||||
- inventory exact-runtime bridge accepted locally: runtime-bridge/answer-adapter/pilot-executor slice passed `70/70` with `1` skipped; full MCP-discovery slice passed `279/279` with `9` skipped; build passed; graphify rebuilt to `5930 nodes`, `12884 edges`, `135 communities`
|
||||
|
||||
Reference in New Issue
Block a user