Planner Autonomy: выбрать comparison chain для явного bidirectional graph

docs/ARCH/11 - architecture_turnaround/20 - planner_autonomy_consolidation_2026-05-01.md

@@ -57,7 +57,7 @@ The planner now instantiates selected evidence chains from this catalog for the
 
 The follow-up consolidation step moved the value-flow planner seams onto the same catalog fabric:
 
-- bidirectional incoming-vs-outgoing comparison now instantiates `value_flow_comparison`;
+- bidirectional incoming-vs-outgoing comparison now instantiates `value_flow_comparison`, including explicit-counterparty comparison graphs rather than only subjectless organization-scope graphs;
 - ranked revenue/payment questions now instantiate `value_flow_ranking`;
 - organization-scoped open totals now instantiate `value_flow` with subjectless primitives but catalog-owned axes and evidence-gate semantics;
 - heuristic fallback routes for value-flow, lifecycle, metadata, movement, document, entity, and unclassified metadata inspection now also use catalog chain templates.
@@ -213,6 +213,13 @@ Latest validation after runtime/debug propagation of structured chain matches:
 - `npm.cmd run build`: passed
 - graphify rebuild: `5940 nodes`, `12909 edges`, `137 communities`
 
+Latest validation after subject-aware bidirectional comparison arbitration:
+
+- targeted planner tests: passed, `36 passed`
+- full MCP-discovery suite: passed, `282 passed`, `9 skipped`
+- `npm.cmd run build`: passed
+- graphify rebuild: `5940 nodes`, `12909 edges`, `137 communities`
+
 ## Next Step
 
 The next safe step is still to re-run live replay once the 1C side is actively polling the proxy. In parallel, local-only consolidation can continue by letting selected downstream arbitration consume `catalog_chain_template_matches` where this can be done without changing exact runtime behavior.

docs/ARCH/11 - architecture_turnaround/README.md

@@ -81,6 +81,7 @@ It now documents a turnaround that is already operational in code, already mater
   - 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;
   - unambiguous metadata surfaces can now infer the next reviewed lane from `Document.*`, `Register.*`, or `Catalog.*` objects even before upstream labels `downstream_route_family`, while mixed surfaces still do not guess;
   - catalog index now scores reviewed chain templates directly from fact/action/axis/comparison/ranking needs, and planner/runtime/debug surfaces expose ranked catalog chain matches through the structured `catalog_chain_template_matches` contract path instead of relying only on reason-code strings;
+  - explicit-counterparty incoming-vs-outgoing data-need graphs now select the reviewed `value_flow_comparison` chain instead of falling back to generic `value_flow`;
   - live map sync: [20 - planner_autonomy_consolidation_2026-05-01.md](./20%20-%20planner_autonomy_consolidation_2026-05-01.md)
 
 Current honest status:
@@ -92,7 +93,7 @@ Current honest status:
 - open-world bounded-autonomy readiness: `~85%`
 - 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: `~82%` 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, and runtime/debug propagation validated locally, but live replay for the new bridge is currently blocked by missing active 1C polling and broader unfamiliar 1C asks still need replay-backed growth
+- Planner Autonomy Consolidation progress: `~83%` 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, and subject-aware bidirectional comparison arbitration validated locally, but live replay for the new bridge is currently blocked by missing active 1C polling and broader unfamiliar 1C asks still need replay-backed growth
 - graph snapshot after latest rebuild: `5940 nodes`, `12909 edges`, `137 communities`
 - current breakpoint:
   - the validated hot paths are no longer structurally broken;
@@ -141,6 +142,7 @@ Latest live proof now includes:
 - catalog chain-template scoring accepted locally: catalog/planner slice passed `54/54`; full MCP-discovery slice passed `282/282` with `9` skipped; build passed; graphify rebuilt to `5938 nodes`, `12903 edges`, `139 communities`
 - structured chain-template planner contract accepted locally: planner slice passed `36/36`; full MCP-discovery slice passed `282/282` with `9` skipped; build passed; graphify rebuilt to `5939 nodes`, `12906 edges`, `138 communities`
 - structured chain-template runtime/debug propagation accepted locally: runtime/debug slice passed `18/18`; full MCP-discovery slice passed `282/282` with `9` skipped; build passed; graphify rebuilt to `5940 nodes`, `12909 edges`, `137 communities`
+- subject-aware bidirectional comparison arbitration accepted locally: planner slice passed `36/36`; full MCP-discovery slice passed `282/282` with `9` skipped; build passed; graphify rebuilt to `5940 nodes`, `12909 edges`, `137 communities`
 
 Current architectural reading:
 

llm_normalizer/backend/dist/services/assistantMcpDiscoveryPlanner.js

@@ -496,16 +496,19 @@ function recipeFor(input) {
         });
     }
     if (graphFactFamily === "value_flow") {
-        if (dataNeedGraph?.comparison_need === "incoming_vs_outgoing" && !hasSubjectCandidates(dataNeedGraph)) {
+        if (dataNeedGraph?.comparison_need === "incoming_vs_outgoing") {
             pushUnique(axes, "amount");
             pushUnique(axes, "coverage_target");
             if (requestedAggregationAxis === "month" || graphAggregation === "by_month") {
                 pushUnique(axes, "calendar_month");
             }
             const template = (0, assistantMcpCatalogIndex_1.getAssistantMcpCatalogChainTemplate)("value_flow_comparison");
+            const fallbackPrimitives = hasSubjectCandidates(dataNeedGraph)
+                ? ["resolve_entity_reference", ...template.fallback_primitives]
+                : template.fallback_primitives;
             const primitiveSelection = selectPrimitivesFromGraphAndCatalog({
                 dataNeedGraph,
-                fallbackPrimitives: template.fallback_primitives,
+                fallbackPrimitives,
                 requiredAxes: axes,
                 metadataSurface: input.metadataSurface,
                 actionFamily: action,

llm_normalizer/backend/src/services/assistantMcpDiscoveryPlanner.ts

@@ -703,16 +703,19 @@ function recipeFor(input: AssistantMcpDiscoveryPlannerInput): PlannerRecipe {
   }
 
   if (graphFactFamily === "value_flow") {
-    if (dataNeedGraph?.comparison_need === "incoming_vs_outgoing" && !hasSubjectCandidates(dataNeedGraph)) {
+    if (dataNeedGraph?.comparison_need === "incoming_vs_outgoing") {
       pushUnique(axes, "amount");
       pushUnique(axes, "coverage_target");
       if (requestedAggregationAxis === "month" || graphAggregation === "by_month") {
         pushUnique(axes, "calendar_month");
       }
       const template = getAssistantMcpCatalogChainTemplate("value_flow_comparison");
+      const fallbackPrimitives = hasSubjectCandidates(dataNeedGraph)
+        ? (["resolve_entity_reference", ...template.fallback_primitives] as AssistantMcpDiscoveryPrimitive[])
+        : template.fallback_primitives;
       const primitiveSelection = selectPrimitivesFromGraphAndCatalog({
         dataNeedGraph,
-        fallbackPrimitives: template.fallback_primitives,
+        fallbackPrimitives,
         requiredAxes: axes,
         metadataSurface: input.metadataSurface,
         actionFamily: action,

llm_normalizer/backend/tests/assistantMcpDiscoveryPlanner.test.ts

@@ -520,7 +520,7 @@ describe("assistant MCP discovery planner", () => {
     expect(result.reason_codes).not.toContain("planner_selected_movement_from_confirmed_metadata_surface_ref");
   });
 
-  it("can select value-flow chain from data need graph even when turn meaning family is still under-specified", () => {
+  it("can select bidirectional value-flow comparison from data need graph even when turn meaning family is still under-specified", () => {
     const result = planAssistantMcpDiscovery({
       dataNeedGraph: {
         schema_version: "assistant_data_need_graph_v1",
@@ -552,23 +552,18 @@ describe("assistant MCP discovery planner", () => {
     });
 
     expect(result.planner_status).toBe("ready_for_execution");
-    expect(result.selected_chain_id).toBe("value_flow");
-    expect(result.proposed_primitives).toEqual([
-      "resolve_entity_reference",
-      "query_movements",
-      "aggregate_by_axis",
-      "probe_coverage"
-    ]);
+    expect(result.selected_chain_id).toBe("value_flow_comparison");
+    expect(result.proposed_primitives).toEqual(["resolve_entity_reference", "query_movements", "probe_coverage"]);
     expect(result.required_axes).toEqual([
       "counterparty",
       "period",
-      "aggregate_axis",
       "amount",
       "coverage_target",
       "calendar_month"
     ]);
-    expect(result.reason_codes).toContain("planner_selected_monthly_value_flow_from_data_need_graph");
-    expect(result.reason_codes).toContain("planner_instantiated_catalog_chain_template_value_flow");
+    expect(result.reason_codes).toContain("planner_selected_bidirectional_value_flow_comparison_from_data_need_graph");
+    expect(result.reason_codes).toContain("planner_instantiated_catalog_chain_template_value_flow_comparison");
+    expect(result.catalog_chain_template_matches[0]).toBe("value_flow_comparison");
   });
 
   it("does not collapse a ranking-shaped value graph into entity-resolution just because no subject is preselected", () => {