diff --git a/frontend/src/lib/rectification-agentic/core/candidate-contrast-packet.ts b/frontend/src/lib/rectification-agentic/core/candidate-contrast-packet.ts
index 9c56c7de..66062a8a 100644
--- a/frontend/src/lib/rectification-agentic/core/candidate-contrast-packet.ts
+++ b/frontend/src/lib/rectification-agentic/core/candidate-contrast-packet.ts
@@ -103,6 +103,7 @@ export type EngineContrastProbe = Readonly<{
     answer_class?: string;
     sign?: string;
   }>[];
+  target_evidence_id?: string;
 }>;
 
 const REMAINING_LAYERS = ["d24", "d5", "d10", "d9", "d4", "d7", "d12", "d2", "d11", "d30"] as const;
diff --git a/frontend/src/lib/rectification-agentic/core/snapshot-source.ts b/frontend/src/lib/rectification-agentic/core/snapshot-source.ts
index ac184ca2..6812651a 100644
--- a/frontend/src/lib/rectification-agentic/core/snapshot-source.ts
+++ b/frontend/src/lib/rectification-agentic/core/snapshot-source.ts
@@ -16,6 +16,7 @@ export type SnapshotStaleReason =
   | "scoreable_evidence_changed"
   | "candidate_set_superseded"
   | "inference_revision_changed"
+  | "scoring_policy_changed"
   | "fingerprint_missing";
 
 export const SNAPSHOT_STALE_COPY: Readonly<Record<SnapshotStaleReason, string>> = {
@@ -23,6 +24,7 @@ export const SNAPSHOT_STALE_COPY: Readonly<Record<SnapshotStaleReason, string>>
   scoreable_evidence_changed: "可评分证据已变化，请重新比较候选",
   candidate_set_superseded: "已有更新的候选结果",
   inference_revision_changed: "候选后验已更新，请使用当前结果",
+  scoring_policy_changed: "评分政策已更新，请重新比较候选",
   fingerprint_missing: "候选快照缺少证据指纹，请重新比较候选",
 };
 
@@ -50,6 +52,13 @@ export function classifySnapshotStaleReason(
   if (snapshot.inferenceRevision !== current.inferenceRevision) {
     return "inference_revision_changed";
   }
+  if (
+    snapshot.scoringPolicyVersion
+    && current.scoringPolicyVersion
+    && snapshot.scoringPolicyVersion !== current.scoringPolicyVersion
+  ) {
+    return "scoring_policy_changed";
+  }
   return null;
 }
 
@@ -62,7 +71,8 @@ export function scoreableSnapshotIsCurrent(
   if (!snapshot.scoreableEvidenceFingerprint || !current.scoreableEvidenceFingerprint) return false;
   return snapshot.scoreableEvidenceFingerprint === current.scoreableEvidenceFingerprint
     && snapshot.candidateSetVersion === current.candidateSetVersion
-    && snapshot.inferenceRevision === current.inferenceRevision;
+    && snapshot.inferenceRevision === current.inferenceRevision
+    && snapshot.scoringPolicyVersion === current.scoringPolicyVersion;
 }
 
 /** No stored snapshot is not stale; a stored snapshot with a missing fingerprint is. */
diff --git a/frontend/src/lib/rectification-agentic/core/types.ts b/frontend/src/lib/rectification-agentic/core/types.ts
index 92b8aa5b..6758326c 100644
--- a/frontend/src/lib/rectification-agentic/core/types.ts
+++ b/frontend/src/lib/rectification-agentic/core/types.ts
@@ -85,6 +85,7 @@ export type ConflictProbe = Readonly<{
   information_gain: number;
   source: string;
   choice_kind?: ProbeChoiceKind;
+  target_evidence_id?: string;
   style_options?: readonly Readonly<{
     label: string;
     answer_class: AnswerClass;
diff --git a/frontend/src/lib/rectification-agentic/v9/answer-choice.ts b/frontend/src/lib/rectification-agentic/v9/answer-choice.ts
index 295c3e8d..54c1d7ca 100644
--- a/frontend/src/lib/rectification-agentic/v9/answer-choice.ts
+++ b/frontend/src/lib/rectification-agentic/v9/answer-choice.ts
@@ -50,9 +50,19 @@ import {
   spokenFollowupForUser,
 } from "./method-followup";
 import type { SessionOutcomeKind } from "./confirmation-gate";
-import { refinementFromDecisionReceipt } from "./refinement-packet";
+import { prospectiveWindowsNarration, refinementFromDecisionReceipt } from "./refinement-packet";
 import { projectCurrentQuestion } from "./turn-decision";
 
+function withProspectiveWindows(
+  base: string,
+  receipt: Readonly<Record<string, unknown>> | null | undefined,
+): string {
+  const extra = prospectiveWindowsNarration(
+    refinementFromDecisionReceipt(receipt).prospective_probes,
+  );
+  return extra ? `${base} ${extra}` : base;
+}
+
 export type ApplyChoiceCommand = Readonly<{
   userId: string;
   caseId: string;
@@ -386,10 +396,10 @@ export async function persistNextInterviewAfterChoice(input: {
       });
     }
     return {
-      hostNarration: nonConvergingRangeNarration({
+      hostNarration: withProspectiveWindows(nonConvergingRangeNarration({
         credibleRange: input.nextAction.credible_range,
         representativeTime: input.nextAction.representative_time,
-      }),
+      }), latest.decisionReceipt),
       choiceReady: false,
     };
   }
@@ -535,10 +545,10 @@ export async function persistNextInterviewIfIdle(input: {
     return {
       persisted: false,
       choiceReady: false,
-      hostNarration: `${nonConvergingRangeNarration({
+      hostNarration: withProspectiveWindows(`${nonConvergingRangeNarration({
         credibleRange: decision.credibleRange,
         representativeTime: decision.representativeTime,
-      })} 可以从下面的时间里选一个采用。`,
+      })} 可以从下面的时间里选一个采用。`, dossier.latestResult?.decisionReceipt),
     };
   }
   if (isNonConvergingRangeOffer(decision)) {
@@ -615,7 +625,7 @@ async function persistExhaustionCollect(input: {
   return {
     persisted,
     choiceReady: false,
-    hostNarration: range,
+    hostNarration: withProspectiveWindows(range, input.decisionReceipt ?? input.dossier.latestResult?.decisionReceipt),
   };
 }
 
@@ -702,19 +712,19 @@ ${nextInterview.hostNarration}`
     }
   }
   const adoptionNarration = nextAction.can_adopt
-    ? `${nonConvergingRangeNarration({
+    ? withProspectiveWindows(`${nonConvergingRangeNarration({
       credibleRange: nextAction.credible_range,
       representativeTime: nextAction.representative_time,
-    })} 可以从下面的时间里选一个采用。`
+    })} 可以从下面的时间里选一个采用。`, input.dossier.latestResult?.decisionReceipt)
     : null;
   const completedRangePrefix = input.narration.replace(RECTIFICATION_TERMINATION_COPY, "").trim();
   const completedRangeNarration = nextAction.type === "complete_with_range"
-    ? `${completedRangePrefix}
+    ? withProspectiveWindows(`${completedRangePrefix}
 
 ${nonConvergingRangeNarration({
   credibleRange: nextAction.credible_range,
   representativeTime: nextAction.representative_time,
-}, RECTIFICATION_TERMINATION_COPY)}`
+}, RECTIFICATION_TERMINATION_COPY)}`, input.dossier.latestResult?.decisionReceipt)
     : null;
   if (adoptionNarration || completedRangeNarration) {
     hostNarration = adoptionNarration ?? completedRangeNarration ?? hostNarration;
diff --git a/frontend/src/lib/rectification-agentic/v9/choice-card.ts b/frontend/src/lib/rectification-agentic/v9/choice-card.ts
index 5e9cee54..76c34e25 100644
--- a/frontend/src/lib/rectification-agentic/v9/choice-card.ts
+++ b/frontend/src/lib/rectification-agentic/v9/choice-card.ts
@@ -275,6 +275,7 @@ function eventQuestionPrompt(
   family: string,
   kind: EventProbeChoiceKind,
   domain?: string | null,
+  probe?: DiscriminatingEventProbe | null,
 ): string {
   const time = period.trim();
   const topic = family.replace(/[？?。]+$/g, "").trim();
@@ -283,6 +284,9 @@ function eventQuestionPrompt(
       ? "亲密关系里，你更接近哪一种相处方式？"
       : "平时做事，你更接近下面哪一种？";
   }
+  if (kind === "event_quality" && probe?.role === "distinguish" && probe.target_evidence_id && probe.user_meaning?.trim()) {
+    return probe.user_meaning.trim();
+  }
   if (!topic) return time;
   const dated = isConcreteChoicePeriod(period);
   if (kind === "event_quality") {
@@ -342,7 +346,7 @@ function hypothesisFor(
   } else if (!isConcreteChoicePeriod(period)) {
     return null;
   }
-  const prompt = eventQuestionPrompt(period, probe.event_family, kind, domain);
+  const prompt = eventQuestionPrompt(period, probe.event_family, kind, domain, probe);
   const why = probe.user_meaning?.trim() || followup.user_prompt_hint.trim();
   if (!why) return null;
   return withStyleOptionLabels(prompt, why, null, styleOptions.options);
diff --git a/frontend/src/lib/rectification-agentic/v9/decision-from-dossier.ts b/frontend/src/lib/rectification-agentic/v9/decision-from-dossier.ts
index 474da9d9..a3c4fb9c 100644
--- a/frontend/src/lib/rectification-agentic/v9/decision-from-dossier.ts
+++ b/frontend/src/lib/rectification-agentic/v9/decision-from-dossier.ts
@@ -184,7 +184,7 @@ export function contrastPacketFromLatestResult(
   const answered = new Set((inference?.answered_probes ?? []).map((item) => item.probe_id));
   const fromInference: EngineContrastProbe[] = (inference?.probes ?? []).flatMap((probe) => {
     if (answered.has(probe.id) || probe.information_gain <= 0) return [];
-    if (probe.source === "known_event_quality") return [];
+    if (probe.source === "known_event_quality" && !probe.target_evidence_id) return [];
     if (probe.source === "nakshatra_boundary") return [];
     return [{
       semantic_key: probe.semantic_key,
@@ -200,6 +200,7 @@ export function contrastPacketFromLatestResult(
         ? { choice_kind: probe.choice_kind }
         : {}),
       ...(probe.style_options?.length ? { style_options: probe.style_options } : {}),
+      ...(probe.target_evidence_id ? { target_evidence_id: probe.target_evidence_id } : {}),
     }];
   });
   const merged = mergeEngineProbes(
@@ -214,6 +215,7 @@ export function contrastPacketFromLatestResult(
       expected_outcomes: probe.expected_outcomes,
       choice_kind: probe.choice_kind,
       style_options: probe.style_options,
+      target_evidence_id: probe.target_evidence_id,
     })),
   );
   return buildCandidateContrastPacket({
@@ -398,6 +400,7 @@ function completedProbeForSemanticKey(
       expected_outcomes: event.expected_outcomes,
       ...(event.choice_kind ? { choice_kind: event.choice_kind } : {}),
       ...(event.style_options?.length ? { style_options: event.style_options } : {}),
+      ...(event.target_evidence_id ? { target_evidence_id: event.target_evidence_id } : {}),
     }],
     candidateTimes: [...(event.candidate_ids ?? [])],
   });
diff --git a/frontend/src/lib/rectification-agentic/v9/method-followup.ts b/frontend/src/lib/rectification-agentic/v9/method-followup.ts
index ceb0b0ac..6bff4e28 100644
--- a/frontend/src/lib/rectification-agentic/v9/method-followup.ts
+++ b/frontend/src/lib/rectification-agentic/v9/method-followup.ts
@@ -508,10 +508,13 @@ function remainingReverseVerifyProbes(
   const dasha: DiscriminatingEventProbe[] = [];
   const fallback: DiscriminatingEventProbe[] = [];
   for (const probe of probes ?? []) {
-    if (probe.source === "known_event_quality" || probe.role === "clarify" || probe.phase === "event_clarification") continue;
+    const anchoredQuality = probe.source === "known_event_quality"
+      && Boolean(probe.target_evidence_id)
+      && probe.role === "distinguish";
+    if ((probe.source === "known_event_quality" && !anchoredQuality) || probe.role === "clarify" || probe.phase === "event_clarification") continue;
     if (probe.role === "collect" || probe.phase === "evidence_collection") continue;
     if (declined.has(probe.domain)) continue;
-    if (probeYearAlreadyCovered(evidence, probe.domain, probe.year)) continue;
+    if (!anchoredQuality && probeYearAlreadyCovered(evidence, probe.domain, probe.year)) continue;
     if (probeBelowAdultFloor(probe, birthDate)) continue;
     if (CONFLICT_PROBE_SOURCES.has(probe.source)) {
       dasha.push(probe);
@@ -1109,7 +1112,8 @@ export function isOfferBlockingFollowup(
 
 function discriminatorFromFollowup(followup: MethodFollowup | null): CandidateDiscriminatorProbe | null {
   if (!followup) return null;
-  if (followup.choice_kind === "event_quality" || followup.intent === "clarify_event") return null;
+  if (followup.intent === "clarify_event") return null;
+  if (followup.choice_kind === "event_quality" && followup.intent !== "distinguish_candidates") return null;
   const realProbe = followup.source === "event_probe"
     || followup.source === "reverse_verify"
     || (followup.source === "active_focus" && followup.intent === "distinguish_candidates");
diff --git a/frontend/src/lib/rectification-agentic/v9/refinement-packet.ts b/frontend/src/lib/rectification-agentic/v9/refinement-packet.ts
index e60cb528..8b393e4d 100644
--- a/frontend/src/lib/rectification-agentic/v9/refinement-packet.ts
+++ b/frontend/src/lib/rectification-agentic/v9/refinement-packet.ts
@@ -226,6 +226,24 @@ export type DiscriminatingEventProbe = Readonly<{
   right_time?: string;
   choice_kind?: EventProbeChoiceKind;
   style_options?: readonly EventProbeStyleOption[];
+  target_evidence_id?: string;
+  display_date_label?: string;
+}>;
+
+export type ProspectiveProbe = Readonly<{
+  candidate_label: string;
+  domain: string;
+  window_label: string;
+  user_meaning: string;
+  used_for_scoring: false;
+}>;
+
+export type DroppedPacketProbe = Readonly<{
+  semantic_key?: string;
+  reason: string;
+  domain?: string;
+  year?: number;
+  source?: string;
 }>;
 
 export type CandidateContrastOpportunity = Readonly<{
@@ -492,6 +510,8 @@ function parseExpectedOutcomes(value: unknown): ProbeExpectedOutcome[] {
 }
 
 function parseProbeRole(row: Readonly<Record<string, unknown>>, source: string): EventProbeRole {
+  const anchored = typeof row.target_evidence_id === "string" && row.target_evidence_id.length > 0;
+  if (source === "known_event_quality" && row.role === "distinguish" && anchored) return "distinguish";
   if (row.role === "clarify" || source === "known_event_quality") return "clarify";
   if (row.role === "collect" || source === "age_band") return "collect";
   if (row.role === "holdout" || source === "oos_blind") return "holdout";
@@ -531,7 +551,13 @@ export function parseEventProbes(
       || CLOCK_IN_COPY.test(family)
       || row.unique_minute_claim === true
     ) continue;
-    if (!options.allowQuality && source === "known_event_quality") continue;
+    const evidenceId = typeof row.target_evidence_id === "string" && UUID.test(row.target_evidence_id)
+      ? row.target_evidence_id
+      : null;
+    const anchoredQuality = source === "known_event_quality"
+      && Boolean(evidenceId)
+      && (row.role === "distinguish" || row.phase === "candidate_discriminator");
+    if (!options.allowQuality && source === "known_event_quality" && !anchoredQuality) continue;
     const tracks = Array.isArray(row.tracks)
       ? row.tracks.filter((track): track is "vimshottari" | "narayana" => track === "vimshottari" || track === "narayana")
       : [];
@@ -578,6 +604,8 @@ export function parseEventProbes(
         }),
       } : {}),
       ...(expectedOutcomes.length > 0 ? { expected_outcomes: expectedOutcomes } : {}),
+      ...(evidenceId ? { target_evidence_id: evidenceId } : {}),
+      ...(asText(row.display_date_label, 40) ? { display_date_label: asText(row.display_date_label, 40)! } : {}),
     };
     if (options.requireDistinguishContract && distinguishContractErrors({ ...probe, role: "distinguish" }).length > 0) continue;
     rows.push(probe);
@@ -602,6 +630,65 @@ export function parseCollectionProbes(value: unknown): readonly DiscriminatingEv
   ));
 }
 
+export function parseDroppedPacketProbes(value: unknown): readonly DroppedPacketProbe[] {
+  if (!Array.isArray(value)) return [];
+  const rows: DroppedPacketProbe[] = [];
+  for (const item of value) {
+    const row = asRecord(item);
+    const reason = asText(row?.reason, 80);
+    if (!row || !reason) continue;
+    rows.push({
+      reason,
+      ...(asText(row.semantic_key, 80) ? { semantic_key: asText(row.semantic_key, 80)! } : {}),
+      ...(typeof row.domain === "string" ? { domain: row.domain } : {}),
+      ...(typeof row.year === "number" && Number.isInteger(row.year) ? { year: row.year } : {}),
+      ...(typeof row.source === "string" ? { source: row.source } : {}),
+    });
+    if (rows.length === 16) break;
+  }
+  return rows;
+}
+
+export function parseProspectiveProbes(value: unknown): readonly ProspectiveProbe[] {
+  if (!Array.isArray(value)) return [];
+  const rows: ProspectiveProbe[] = [];
+  for (const item of value) {
+    const row = asRecord(item);
+    const label = asText(row?.candidate_label, 8);
+    const domain = typeof row?.domain === "string" ? row.domain : "";
+    const window = asText(row?.window_label, 40);
+    const meaning = asText(row?.user_meaning, 200);
+    if (
+      !row
+      || !label
+      || !EVENT_PROBE_DOMAIN_SET.has(domain)
+      || !window
+      || !meaning
+      || row.used_for_scoring === true
+      || CLOCK_IN_COPY.test(meaning)
+      || CLOCK_IN_COPY.test(window)
+    ) continue;
+    rows.push({
+      candidate_label: label,
+      domain,
+      window_label: window,
+      user_meaning: meaning,
+      used_for_scoring: false,
+    });
+    if (rows.length === 3) break;
+  }
+  return rows;
+}
+
+export function prospectiveWindowsNarration(probes: readonly ProspectiveProbe[]): string | null {
+  const meanings = probes
+    .filter((item) => item.used_for_scoring === false)
+    .map((item) => item.user_meaning.trim())
+    .filter((item) => item.length > 0 && !CLOCK_IN_COPY.test(item));
+  if (meanings.length === 0) return null;
+  return meanings.join(" ");
+}
+
 export function parseCandidateContrastOpportunities(value: unknown): readonly CandidateContrastOpportunity[] {
   if (!Array.isArray(value)) return [];
   const rows: CandidateContrastOpportunity[] = [];
@@ -711,6 +798,8 @@ export function refinementFromDecisionReceipt(
   event_clarification_probes: readonly DiscriminatingEventProbe[];
   evidence_collection_probes: readonly DiscriminatingEventProbe[];
   candidate_contrast_opportunities: readonly CandidateContrastOpportunity[];
+  dropped_probes: readonly DroppedPacketProbe[];
+  prospective_probes: readonly ProspectiveProbe[];
 } {
   return {
     event_dasha_ledger: parseEventDashaLedger(receipt?.event_dasha_ledger),
@@ -728,5 +817,7 @@ export function refinementFromDecisionReceipt(
       receipt?.evidence_collection_probes ?? receipt?.discriminating_event_probes,
     ),
     candidate_contrast_opportunities: parseCandidateContrastOpportunities(receipt?.candidate_contrast_opportunities),
+    dropped_probes: parseDroppedPacketProbes(receipt?.dropped_probes),
+    prospective_probes: parseProspectiveProbes(receipt?.prospective_probes),
   };
 }
diff --git a/frontend/tests/rectification-engine-convergence.test.ts b/frontend/tests/rectification-engine-convergence.test.ts
new file mode 100644
index 00000000..0f971fb7
--- /dev/null
+++ b/frontend/tests/rectification-engine-convergence.test.ts
@@ -0,0 +1,180 @@
+import assert from "node:assert/strict";
+import { readFileSync, readdirSync, statSync } from "node:fs";
+import { join } from "node:path";
+import test from "node:test";
+
+import { answersFromEvidence } from "../src/lib/rectification-agentic/core/build-state.ts";
+import {
+  classifySnapshotStaleReason,
+  storedSnapshotIsCurrent,
+  type CandidateSnapshotSource,
+} from "../src/lib/rectification-agentic/core/snapshot-source.ts";
+import type { ConflictProbe } from "../src/lib/rectification-agentic/core/types.ts";
+import { contrastPacketFromLatestResult } from "../src/lib/rectification-agentic/v9/decision-from-dossier.ts";
+import {
+  parseDiscriminatingEventProbes,
+  parseProspectiveProbes,
+  prospectiveWindowsNarration,
+} from "../src/lib/rectification-agentic/v9/refinement-packet.ts";
+
+const OUTCOMES = [
+  { answer_class: "yes" as const, supports: ["05:00"], conflicts: ["05:07"] },
+  { answer_class: "no" as const, supports: ["05:07"], conflicts: ["05:00"] },
+];
+
+function source(overrides: Partial<CandidateSnapshotSource> = {}): CandidateSnapshotSource {
+  return {
+    birthProfileFingerprint: "birth-a",
+    scoreableEvidenceFingerprint: "score-a",
+    inferenceRevision: 3,
+    candidateSetVersion: "set-a",
+    scoringPolicyVersion: "rectification-candidate-policy-v3",
+    ...overrides,
+  };
+}
+
+function qualityProbe(overrides: Record<string, unknown> = {}) {
+  return {
+    year: 2016,
+    year_label: "2016 年 9 月",
+    domain: "education",
+    event_family: "学业或考试发挥失常、压力特别大",
+    source: "known_event_quality",
+    tracks: ["vimshottari", "narayana"],
+    tracks_agree: true,
+    unique_minute_claim: false,
+    user_meaning: "2016 年 9 月那次上大学，更接近哪一种实际体验。",
+    role: "distinguish",
+    phase: "candidate_discriminator",
+    semantic_key: "education.2016.known_event_quality",
+    information_gain: 0.42,
+    candidate_split_hash: "quality-split",
+    candidate_set_version: "set-a",
+    candidate_ids: ["05:00", "05:07"],
+    expected_outcomes: OUTCOMES,
+    choice_kind: "event_quality",
+    question_contract_version: "probe-question-v1",
+    target_evidence_id: "00000000-0000-4000-8000-000000000001",
+    display_date_label: "2016 年 9 月",
+    ...overrides,
+  };
+}
+
+function walkFiles(root: string, suffixes: readonly string[]): string[] {
+  const skip = new Set(["node_modules", ".next", "dist"]);
+  const out: string[] = [];
+  const visit = (dir: string) => {
+    for (const name of readdirSync(dir)) {
+      if (skip.has(name)) continue;
+      const path = join(dir, name);
+      const stat = statSync(path);
+      if (stat.isDirectory()) visit(path);
+      else if (suffixes.some((suffix) => name.endsWith(suffix))) out.push(path);
+    }
+  };
+  visit(root);
+  return out;
+}
+
+test("scoring policy version changes stale stored snapshots", () => {
+  const current = source();
+  const previous = source({ scoringPolicyVersion: "rectification-candidate-policy-v2" });
+  assert.equal(classifySnapshotStaleReason(previous, current), "scoring_policy_changed");
+  assert.equal(storedSnapshotIsCurrent(previous, current), false);
+  assert.equal(storedSnapshotIsCurrent(current, current), true);
+});
+
+test("anchored known_event_quality distinguish probes stay in the public packet", () => {
+  const parsed = parseDiscriminatingEventProbes([qualityProbe()]);
+  assert.equal(parsed.length, 1);
+  assert.equal(parsed[0]?.source, "known_event_quality");
+  assert.equal(parsed[0]?.role, "distinguish");
+  assert.equal(parsed[0]?.target_evidence_id, "00000000-0000-4000-8000-000000000001");
+
+  const unanchored = parseDiscriminatingEventProbes([qualityProbe({ target_evidence_id: undefined })]);
+  assert.equal(unanchored.length, 0);
+
+  const packet = contrastPacketFromLatestResult({
+    resultId: "result-a",
+    decisionReceipt: {
+      discriminating_event_probes: [qualityProbe()],
+      inference_state: {
+        probes: [{
+          id: "probe:education.2016.known_event_quality:quality-split",
+          semantic_key: "education.2016.known_event_quality",
+          candidate_split_hash: "quality-split",
+          domain: "education",
+          year: 2016,
+          question: "2016 年 9 月那次上大学，更接近哪一种实际体验。",
+          candidate_ids: ["05:00", "05:07"],
+          expected_outcomes: OUTCOMES,
+          information_gain: 0.42,
+          source: "known_event_quality",
+          choice_kind: "event_quality",
+          target_evidence_id: "00000000-0000-4000-8000-000000000001",
+        }],
+        answered_probes: [],
+      },
+    },
+  });
+  assert.equal(
+    packet.probes.some((item) => (
+      item.semanticKey === "education.2016.known_event_quality"
+      || item.choiceKind === "event_quality"
+    )),
+    true,
+  );
+});
+
+test("quality answers are not inferred from event existence alone", () => {
+  const probe: ConflictProbe = {
+    id: "probe-quality",
+    semantic_key: "education.2016.known_event_quality",
+    candidate_split_hash: "quality-split",
+    domain: "education",
+    year: 2016,
+    question: "那次上大学更接近哪一种体验",
+    candidate_ids: ["05:00", "05:07"],
+    expected_outcomes: OUTCOMES,
+    information_gain: 0.42,
+    source: "known_event_quality",
+    choice_kind: "event_quality",
+  };
+  const answers = answersFromEvidence([probe], [{
+    id: "00000000-0000-4000-8000-000000000001",
+    domain: "education",
+    year: 2016,
+    precision: "month",
+  }]);
+  assert.equal(answers.length, 0);
+});
+
+test("prospective probes stay out of scoring copy", () => {
+  const parsed = parseProspectiveProbes([{
+    candidate_label: "A",
+    domain: "career",
+    window_label: "2027 年 3 月附近",
+    user_meaning: "候选 A 预测下一次事业变动更可能在 2027 年 3 月附近。这是预测窗口，不是承诺；下次发生时回来补一条，可进一步分辨。",
+    used_for_scoring: false,
+  }]);
+  assert.equal(parsed.length, 1);
+  const copy = prospectiveWindowsNarration(parsed);
+  assert.match(String(copy), /预测窗口/);
+  assert.equal(parsed[0]?.used_for_scoring, false);
+});
+
+test("scripts and frontend contain no answer-key literals", () => {
+  const forbidden = ["target" + "_minute", "pl9_" + "1993", "regression" + "_only"];
+  const files = [
+    ...walkFiles(join(process.cwd(), "src"), [".ts", ".tsx", ".js"]),
+    ...walkFiles(join(process.cwd(), "tests"), [".ts", ".tsx"]),
+  ];
+  const hits: string[] = [];
+  for (const file of files) {
+    const text = readFileSync(file, "utf8");
+    for (const token of forbidden) {
+      if (text.includes(token)) hits.push(`${file}:${token}`);
+    }
+  }
+  assert.deepEqual(hits, []);
+});
diff --git a/scripts/rectification/dasha_transition_proximity.py b/scripts/rectification/dasha_transition_proximity.py
new file mode 100644
index 00000000..985031f2
--- /dev/null
+++ b/scripts/rectification/dasha_transition_proximity.py
@@ -0,0 +1,198 @@
+"""Deterministic dasha-transition proximity scoring for day/month events.
+
+Birth-time drift of about 1 minute moves Vimshottari/Narayana transition
+dates by a few days. A dated event near a candidate's AD/PD change is a
+bounded auxiliary signal, never larger than one day-level event body.
+"""
+
+from __future__ import annotations
+
+from collections.abc import Callable, Sequence
+from datetime import date
+from typing import Any
+
+from scripts.rectification.event_probes import _narayana_start_dates, _vim_start_dates
+
+PROXIMITY_WINDOW_DAYS = 45
+DAY_KERNEL_DAYS = 15
+MONTH_KERNEL_DAYS = 45
+DAY_MAX_POINTS = 1.0
+MONTH_MAX_POINTS = 0.35
+VIM_SHARE = 0.6
+NARAYANA_SHARE = 0.4
+
+
+def representative_event_date(event: dict[str, Any]) -> date | None:
+    precision = str(event.get("precision") or "")
+    if precision not in {"day", "month"}:
+        return None
+    raw_start = event.get("date_start") or event.get("date")
+    raw_end = event.get("date_end") or raw_start
+    try:
+        start = date.fromisoformat(str(raw_start)[:10])
+        end = date.fromisoformat(str(raw_end)[:10])
+    except ValueError:
+        return None
+    if precision == "day" or start == end:
+        return start
+    mid_day = min(15, end.day)
+    try:
+        return start.replace(day=mid_day)
+    except ValueError:
+        return start
+
+
+def _nearest_delta(starts: Sequence[date], event_date: date) -> tuple[date | None, int | None]:
+    eligible = [
+        item for item in starts
+        if abs((item - event_date).days) <= PROXIMITY_WINDOW_DAYS
+    ]
+    if not eligible:
+        return None, None
+    nearest = min(eligible, key=lambda item: (abs((item - event_date).days), item.toordinal()))
+    return nearest, abs((nearest - event_date).days)
+
+
+def _kernel(delta_days: int | None, width: float) -> float:
+    if delta_days is None or width <= 0:
+        return 0.0
+    return max(0.0, 1.0 - (delta_days / width))
+
+
+def score_transition_proximity(
+    *,
+    event_date: date,
+    precision: str,
+    vim_starts: Sequence[date],
+    narayana_starts: Sequence[date] | None = None,
+    vim_pd_starts: Sequence[date] | None = None,
+) -> dict[str, Any]:
+    if precision not in {"day", "month"}:
+        return {
+            "points": 0.0,
+            "rule_ids": [],
+            "nearest_vim_delta_days": None,
+            "nearest_narayana_delta_days": None,
+        }
+    kernel_width = float(DAY_KERNEL_DAYS if precision == "day" else MONTH_KERNEL_DAYS)
+    cap = DAY_MAX_POINTS if precision == "day" else MONTH_MAX_POINTS
+    ad_starts = list(vim_starts)
+    pd_starts = list(vim_pd_starts or ())
+    ad_date, ad_delta = _nearest_delta(ad_starts, event_date)
+    pd_date, pd_delta = _nearest_delta(pd_starts, event_date)
+    if pd_delta is not None and (ad_delta is None or pd_delta < ad_delta):
+        vim_delta = pd_delta
+        vim_kind = "pd"
+        vim_date = pd_date
+    else:
+        vim_delta = ad_delta
+        vim_kind = "ad"
+        vim_date = ad_date
+    _, narayana_delta = _nearest_delta(list(narayana_starts or ()), event_date)
+    vim_kernel = _kernel(vim_delta, kernel_width)
+    narayana_kernel = _kernel(narayana_delta, kernel_width)
+    points = round(cap * (VIM_SHARE * vim_kernel + NARAYANA_SHARE * narayana_kernel), 4)
+    rules: list[str] = []
+    if vim_kernel > 0:
+        rules.append(f"vim_transition_proximity_{vim_kind}")
+    if narayana_kernel > 0:
+        rules.append("narayana_transition_proximity_ad")
+    return {
+        "points": points,
+        "rule_ids": rules,
+        "nearest_vim_delta_days": vim_delta,
+        "nearest_narayana_delta_days": narayana_delta,
+        "nearest_vim_date": vim_date,
+    }
+
+
+def _context_time(context: dict[str, Any]) -> str | None:
+    feature = context.get("feature") if isinstance(context.get("feature"), dict) else {}
+    raw = feature.get("time")
+    if isinstance(raw, str) and len(raw) >= 5:
+        return raw[:5]
+    at = context.get("candidate_at")
+    if hasattr(at, "strftime"):
+        return at.strftime("%H:%M")
+    return None
+
+
+def merge_transition_proximity(
+    matrix: dict[str, dict[str, dict[str, Any]]],
+    events: Sequence[dict[str, Any]],
+    static_contexts: Sequence[dict[str, Any]],
+    birth_date: str,
+    *,
+    public_technique_layers: Callable[[str, Sequence[str]], list[str]],
+) -> None:
+    by_time = {
+        time: context
+        for context in static_contexts
+        if isinstance(context, dict) and (time := _context_time(context))
+    }
+    vim_cache: dict[tuple[Any, ...], list[date]] = {}
+    pd_cache: dict[tuple[Any, ...], list[date]] = {}
+    narayana_cache: dict[tuple[Any, ...], list[date] | None] = {}
+    for event in events:
+        if not isinstance(event, dict):
+            continue
+        event_id = str(event.get("id") or "")
+        cells = matrix.get(event_id)
+        if not event_id or not isinstance(cells, dict):
+            continue
+        event_date = representative_event_date(event)
+        if event_date is None:
+            continue
+        precision = str(event.get("precision") or "")
+        lo, hi = event_date.year - 1, event_date.year + 1
+        for time, cell in cells.items():
+            context = by_time.get(str(time)[:5])
+            if not isinstance(cell, dict) or not isinstance(context, dict):
+                continue
+            moon = (context.get("planet_longitudes") or {}).get("Moon")
+            if not isinstance(moon, (int, float)):
+                continue
+            vim_key = (birth_date, round(float(moon), 6), lo, hi)
+            if vim_key not in vim_cache:
+                vim_cache[vim_key] = _vim_start_dates(birth_date, float(moon), lo, hi)
+                pd_cache[vim_key] = _vim_start_dates(
+                    birth_date,
+                    float(moon),
+                    lo,
+                    hi,
+                    include_pratyantar=True,
+                )
+            planets = context.get("planet_longitudes") or {}
+            asc = context.get("ascendant_index")
+            narayana_key = (
+                birth_date,
+                int(asc) if isinstance(asc, int) else None,
+                lo,
+                hi,
+                round(float(moon), 6),
+            )
+            if narayana_key not in narayana_cache:
+                narayana_cache[narayana_key] = (
+                    _narayana_start_dates(int(asc), planets, birth_date, lo, hi)
+                    if isinstance(asc, int) and isinstance(planets, dict)
+                    else None
+                )
+            ad_starts = vim_cache[vim_key]
+            ad_set = set(ad_starts)
+            pd_only = [item for item in pd_cache[vim_key] if item not in ad_set]
+            scored = score_transition_proximity(
+                event_date=event_date,
+                precision=precision,
+                vim_starts=ad_starts,
+                vim_pd_starts=pd_only,
+                narayana_starts=narayana_cache[narayana_key] or [],
+            )
+            if scored["points"] <= 0 and not scored["rule_ids"]:
+                continue
+            cell["points"] = round(float(cell.get("points") or 0) + float(scored["points"]), 4)
+            cell["rule_ids"] = sorted({
+                *list(cell.get("rule_ids") or []),
+                *scored["rule_ids"],
+            })
+            domain = str(event.get("domain") or cell.get("domain") or "")
+            cell["technique_layers"] = public_technique_layers(domain, cell["rule_ids"])
diff --git a/scripts/rectification/decision_policy.py b/scripts/rectification/decision_policy.py
index cf2e54c7..40f478ec 100644
--- a/scripts/rectification/decision_policy.py
+++ b/scripts/rectification/decision_policy.py
@@ -26,7 +26,7 @@ from scripts.rectification_policy import (
     MIN_CONFIRMATION_MARGIN_PERCENT,
 )
 
-POLICY_VERSION = "rectification-candidate-policy-v2"
+POLICY_VERSION = "rectification-candidate-policy-v3"
 RECEIPT_VERSION = "candidate-decision-receipt-v2"
 EXECUTION_LEDGER_VERSION = "rectification-execution-ledger-v2"
 SCORE_QUANTUM = Decimal("0.0001")
@@ -71,6 +71,7 @@ _AUDIT_LABELS = {
     "shadbala": ("Shadbala", "本轮已做已核验的 Shadbala 分量辅助对照。"),
     "arudha-pada": ("Arudha Pada", "本轮已做 Arudha 辅助对照。"),
     "functional-benefic-malefic": ("功能吉凶星", "本轮已叠加本命功能吉凶星。"),
+    "dasha-transition-proximity": ("换运贴近度", "本轮已对照日级事件与候选换运日期的贴近程度。"),
 }
 
 
@@ -100,7 +101,9 @@ def _executed_public_methods(built: dict[str, Any]) -> list[str]:
                     methods.add(str(layer))
             for rule in cell.get("rule_ids") or []:
                 text = str(rule)
-                if text.startswith("vim_"):
+                if "transition_proximity" in text:
+                    methods.add("dasha-transition-proximity")
+                elif text.startswith("vim_"):
                     methods.add("vimshottari-dasha")
                 elif text.startswith("narayana_"):
                     methods.add("narayana-dasha")
@@ -652,6 +655,8 @@ def build_decision_receipt(
         "evidence_collection_probes": packet.get("evidence_collection_probes") or [],
         "candidate_contrast_opportunities": packet.get("candidate_contrast_opportunities") or [],
         "holdout_validation_probes": packet.get("holdout_validation_probes") or [],
+        "dropped_probes": packet.get("dropped_probes") or [],
+        "prospective_probes": packet.get("prospective_probes") or [],
         "horary_observation": build_horary_observation(request),
         "unique_minute_claim": False,
     })
diff --git a/scripts/rectification/event_probes.py b/scripts/rectification/event_probes.py
index 132126fa..ba4a0b8a 100644
--- a/scripts/rectification/event_probes.py
+++ b/scripts/rectification/event_probes.py
@@ -1,7 +1,9 @@
 """Public-safe biographical probes from candidate dasha / varga differences.
 
 Discriminators are feature-signature clusters over the full birth window.
-known_event_quality is clarification only and never a distinguish probe.
+known_event_quality may distinguish when signature groups disagree on the
+event's theme varga type, the probe is anchored to confirmed evidence, and
+the case cap is respected. Unanchored quality stays clarification-only.
 """
 
 from __future__ import annotations
@@ -35,7 +37,7 @@ from scripts.rectification.candidate_contrast import (
     missing_collection_domains,
     opportunity_from_probe,
 )
-from scripts.rectification.case_holdout import holdout_domain_years
+from scripts.rectification.case_holdout import holdout_domain_years, holdout_event_ids
 from scripts.rectification.probe_question_contract import (
     QUESTION_CONTRACT_VERSION,
     completed_style_options,
@@ -59,6 +61,66 @@ MAX_PROBES_PER_DOMAIN = 3
 # MAX_PROBES is the published cap after a global information_gain sort.
 MAX_BOUNDARY_CANDIDATES_PER_DOMAIN = 8
 MIN_BOUNDARY_DAYS = 45
+MAX_QUALITY_DISTINGUISH_PROBES = 2
+ANSWER_PRIOR_TABLE_VERSION = "rectification-answer-priors-v1"
+DOMINANT_ANSWER_PRIOR = 0.8
+# Conservative population rates, not fitted from product users.
+# Broad family existence (yearless / 3-year window / age band) is near-certain
+# in adult life. Specified-year long-distance move is uncommon. Quality answers
+# are closer to even because they condition on a known event.
+_DEFAULT_EXISTENCE_PRIORS = {"yes": 0.35, "weak_yes": 0.15, "no": 0.40, "unsure": 0.10}
+_DEFAULT_QUALITY_PRIORS = {"yes": 0.30, "weak_yes": 0.20, "no": 0.40, "unsure": 0.10}
+ANSWER_PRIORS: dict[tuple[str, str], dict[str, float]] = {
+    ("family", "existence"): {"yes": 0.85, "weak_yes": 0.05, "no": 0.05, "unsure": 0.05},
+    ("relocation", "existence"): {"yes": 0.20, "weak_yes": 0.10, "no": 0.60, "unsure": 0.10},
+    ("education", "existence"): {"yes": 0.45, "weak_yes": 0.15, "no": 0.30, "unsure": 0.10},
+    ("relationship", "existence"): {"yes": 0.40, "weak_yes": 0.15, "no": 0.35, "unsure": 0.10},
+    ("career", "existence"): {"yes": 0.40, "weak_yes": 0.15, "no": 0.35, "unsure": 0.10},
+    ("education", "event_quality"): dict(_DEFAULT_QUALITY_PRIORS),
+    ("relationship", "event_quality"): dict(_DEFAULT_QUALITY_PRIORS),
+    ("career", "event_quality"): dict(_DEFAULT_QUALITY_PRIORS),
+    ("relocation", "event_quality"): dict(_DEFAULT_QUALITY_PRIORS),
+    ("family", "event_quality"): dict(_DEFAULT_QUALITY_PRIORS),
+}
+DOMAIN_QUALITY_LAYER = {
+    "education": "d24",
+    "relationship": "d9",
+    "career": "d10",
+    "relocation": "d4",
+    "family": "d12",
+}
+QUALITY_DISTINGUISH_OPTIONS: dict[str, tuple[dict[str, str], ...]] = {
+    "education": (
+        {"label": "发挥明显失常", "answer_class": "yes"},
+        {"label": "只是将就调剂", "answer_class": "weak_yes"},
+        {"label": "基本如愿录取", "answer_class": "no"},
+        {"label": "当时说不清楚", "answer_class": "unsure"},
+    ),
+    "relationship": (
+        {"label": "明显受挫变难", "answer_class": "yes"},
+        {"label": "只是将就相处", "answer_class": "weak_yes"},
+        {"label": "整体比较顺利", "answer_class": "no"},
+        {"label": "当时说不清楚", "answer_class": "unsure"},
+    ),
+    "career": (
+        {"label": "明显受挫受压", "answer_class": "yes"},
+        {"label": "只是将就应付", "answer_class": "weak_yes"},
+        {"label": "整体比较顺利", "answer_class": "no"},
+        {"label": "当时说不清楚", "answer_class": "unsure"},
+    ),
+    "relocation": (
+        {"label": "搬迁特别折腾", "answer_class": "yes"},
+        {"label": "只是将就安顿", "answer_class": "weak_yes"},
+        {"label": "整体比较顺利", "answer_class": "no"},
+        {"label": "当时说不清楚", "answer_class": "unsure"},
+    ),
+    "family": (
+        {"label": "家里特别操心", "answer_class": "yes"},
+        {"label": "只是普通操心", "answer_class": "weak_yes"},
+        {"label": "整体比较顺利", "answer_class": "no"},
+        {"label": "当时说不清楚", "answer_class": "unsure"},
+    ),
+}
 LEVEL_RANK = {"none": 0, "weak": 1, "medium": 2, "strong": 3}
 LEVEL_P = {"none": 0.15, "weak": 0.35, "medium": 0.62, "strong": 0.82}
 SCORING_LAYERS = ("d1", "d9", "d10", "d4", "d5", "d24", "d7", "d12", "d2", "d11", "d30")
@@ -349,7 +411,14 @@ def _tracks_present(rule_ids: Sequence[str]) -> tuple[bool, bool]:
     )
 
 
-def _vim_start_dates(birth_date: str, moon_longitude: float, lo: int, hi: int) -> list[date]:
+def _vim_start_dates(
+    birth_date: str,
+    moon_longitude: float,
+    lo: int,
+    hi: int,
+    *,
+    include_pratyantar: bool = False,
+) -> list[date]:
     nakshatra, progress, _ = dasha_analyzer.lon_to_nakshatra(float(moon_longitude))
     timeline, _, _, _ = dasha_analyzer.build_dasha_timeline(birth_date, nakshatra, progress)
     starts: list[date] = []
@@ -361,6 +430,12 @@ def _vim_start_dates(birth_date: str, moon_longitude: float, lo: int, hi: int) -
             minor_start = minor.get("start")
             if isinstance(minor_start, datetime) and lo <= minor_start.year <= hi:
                 starts.append(minor_start.date())
+            if not include_pratyantar:
+                continue
+            for prat in dasha_analyzer.build_antardasha(minor):
+                prat_start = prat.get("start")
+                if isinstance(prat_start, datetime) and lo <= prat_start.year <= hi:
+                    starts.append(prat_start.date())
     return starts
 
 
@@ -738,6 +813,187 @@ def _information_gain(left_level: str, right_level: str) -> float:
     return round(max(0.0, 1.0 - after), 4)
 
 
+def _broad_existence_window(probe: dict[str, Any]) -> bool:
+    if str(probe.get("source") or "") == "age_band":
+        return True
+    year = probe.get("year")
+    if not isinstance(year, int) or year <= 0:
+        return True
+    span = probe.get("window_span_years")
+    return isinstance(span, int) and span >= 3
+
+
+def _answer_priors_for(probe: dict[str, Any]) -> dict[str, float]:
+    kind = str(probe.get("choice_kind") or "existence")
+    if kind not in {"existence", "event_quality"}:
+        kind = "existence"
+    domain = str(probe.get("domain") or "")
+    if domain == "family" and kind == "existence" and not _broad_existence_window(probe):
+        return dict(_DEFAULT_EXISTENCE_PRIORS)
+    priors = ANSWER_PRIORS.get((domain, kind))
+    if priors:
+        return dict(priors)
+    return dict(_DEFAULT_QUALITY_PRIORS if kind == "event_quality" else _DEFAULT_EXISTENCE_PRIORS)
+
+
+def _expected_information_gain(raw_split_gain: float, priors: dict[str, float]) -> float:
+    weight = (
+        float(priors.get("yes") or 0)
+        + float(priors.get("no") or 0)
+        + 0.5 * float(priors.get("weak_yes") or 0)
+    )
+    return round(max(0.0, raw_split_gain * weight), 4)
+
+
+def _apply_prior_ranking(probe: dict[str, Any]) -> dict[str, Any]:
+    raw = float(probe.get("information_gain") or 0)
+    priors = _answer_priors_for(probe)
+    probe["raw_split_gain"] = raw
+    probe["answer_priors"] = priors
+    probe["information_gain"] = _expected_information_gain(raw, priors)
+    return probe
+
+
+def _dominant_existence_prior(probe: dict[str, Any], priors: dict[str, float]) -> bool:
+    if str(probe.get("choice_kind") or "existence") != "existence":
+        return False
+    if str(probe.get("source") or "") == "known_event_quality":
+        return False
+    return max(priors.values()) > DOMINANT_ANSWER_PRIOR
+
+
+def _event_month(event: dict[str, Any]) -> int | None:
+    raw = str(event.get("date") or event.get("date_start") or "")
+    if len(raw) >= 7 and raw[4] == "-":
+        try:
+            month = int(raw[5:7])
+        except ValueError:
+            return None
+        if 1 <= month <= 12:
+            return month
+    return None
+
+
+def _display_date_label(event: dict[str, Any]) -> str:
+    year = _event_year(event)
+    month = _event_month(event)
+    if year is None:
+        return "那次"
+    if month:
+        return f"{year} 年 {month} 月"
+    return f"{year} 年"
+
+
+def _quality_user_meaning(event: dict[str, Any], domain: str) -> str:
+    label = _display_date_label(event)
+    if domain == "education":
+        return (
+            f"{label}那次上大学，更接近如愿、将就调剂、发挥失常还是说不清。"
+            "只问那次经历的实际体验，不得改时间范围。"
+        )
+    family = str(DOMAIN_CATALOG[domain]["event_family"])
+    return (
+        f"{label}那次{family}，当时更接近顺利、将就、明显受挫还是说不清。"
+        "只问那次经历的实际体验，不得改时间范围。"
+    )
+
+
+def _quality_distinguish_probes(
+    events: Sequence[dict[str, Any]],
+    clusters: Sequence[dict[str, Any]],
+    *,
+    set_version: str,
+    holdout_ids: set[str],
+    holdout_keys: set[str],
+) -> list[dict[str, Any]]:
+    if len(clusters) < 2:
+        return []
+    rows: list[dict[str, Any]] = []
+    for event in events:
+        if not isinstance(event, dict):
+            continue
+        event_id = str(event.get("id") or "")
+        domain = str(event.get("domain") or "")
+        year = _event_year(event)
+        layer = DOMAIN_QUALITY_LAYER.get(domain)
+        if not event_id or year is None or layer is None or domain not in DOMAIN_CATALOG:
+            continue
+        if event_id in holdout_ids or f"{domain}:{year}" in holdout_keys:
+            continue
+        if _quality_encoded(event, domain):
+            continue
+        groups: dict[int, list[str]] = {}
+        for cluster in clusters:
+            representative = cluster.get("representative") if isinstance(cluster, dict) else None
+            if not isinstance(representative, dict):
+                continue
+            sign = _layer_value(representative, layer)
+            if not isinstance(sign, int):
+                continue
+            bucket = groups.setdefault(sign, [])
+            for time in cluster.get("times") or []:
+                clock = str(time)[:5]
+                if len(clock) == 5 and clock not in bucket:
+                    bucket.append(clock)
+        if len(groups) < 2:
+            continue
+        signs = sorted(groups)
+        yes_times = sorted(groups[signs[-1]], key=_clock)
+        no_times = sorted(groups[signs[0]], key=_clock)
+        overlap = set(yes_times) & set(no_times)
+        yes_times = [time for time in yes_times if time not in overlap]
+        no_times = [time for time in no_times if time not in overlap]
+        if len(yes_times) < 1 or len(no_times) < 1:
+            continue
+        month = _event_month(event)
+        outcomes = [
+            {"answer_class": "yes", "supports": yes_times, "conflicts": no_times},
+            {"answer_class": "weak_yes", "supports": yes_times, "conflicts": no_times},
+            {"answer_class": "no", "supports": no_times, "conflicts": yes_times},
+            {"answer_class": "unsure", "supports": [], "conflicts": []},
+        ]
+        split = candidate_split_hash(
+            candidate_set_version_value=set_version,
+            domain=domain,
+            year=year,
+            month=month,
+            groups=[yes_times, no_times],
+        )
+        gain = round(_group_entropy([len(yes_times), len(no_times)]), 4)
+        if gain <= 0:
+            continue
+        label = _display_date_label(event)
+        probe = _public_probe(
+            year=year,
+            month=month,
+            domain=domain,
+            source="known_event_quality",
+            tracks=("vimshottari", "narayana"),
+            tracks_agree=True,
+            user_meaning=_quality_user_meaning(event, domain),
+            event_family=str(DOMAIN_CATALOG[domain]["quality_family"]),
+            information_gain=gain,
+            semantic_key=f"{domain}.{year}.known_event_quality",
+            candidate_split_hash=split,
+            candidate_set_version=set_version,
+            expected_outcomes=outcomes,
+            candidate_ids=candidate_ids_from_outcomes(outcomes),
+            left_time=yes_times[0],
+            right_time=no_times[0],
+            target_evidence_id=event_id,
+            display_date_label=label,
+            role="distinguish",
+            phase=PROBE_PHASE_CANDIDATE_DISCRIMINATOR,
+            style_options=list(QUALITY_DISTINGUISH_OPTIONS.get(domain) or ()),
+        )
+        if distinguish_contract_errors(probe):
+            continue
+        rows.append(_apply_prior_ranking(probe))
+        if len(rows) >= MAX_QUALITY_DISTINGUISH_PROBES:
+            break
+    return rows
+
+
 def _year_activated(rule_ids: Sequence[str]) -> bool:
     return _has_domain_activation(rule_ids) or LEVEL_RANK.get(match_level(rule_ids), 0) >= 2
 
@@ -1083,7 +1339,7 @@ def candidate_contrast_opportunities(
     return [opportunity_from_probe(probe) for probe in probes]
 
 
-def discriminating_event_probes(
+def _discriminating_event_probe_lists(
     request: dict[str, Any],
     built: dict[str, Any],
     *,
@@ -1092,35 +1348,36 @@ def discriminating_event_probes(
     representative_time: str | None,
     precision_current: str | None = None,
     today: date | None = None,
-) -> list[dict[str, Any]]:
+) -> tuple[list[dict[str, Any]], list[dict[str, Any]]]:
     del precision_current, representative_time
     birth_date = str(request.get("birth_date") or "").strip()
     birth_year = _birth_year(birth_date)
+    empty: tuple[list[dict[str, Any]], list[dict[str, Any]]] = ([], [])
     if birth_year is None:
-        return []
+        return empty
     try:
         datetime.strptime(birth_date, "%Y-%m-%d")
     except ValueError:
-        return []
+        return empty
     events = [item for item in (request.get("events") or []) if isinstance(item, dict)]
     if not discriminator_gate_open(events):
-        return []
+        return empty
     holdout_keys = holdout_domain_years(events)
     now = today or date.today()
     full = _static_contexts(built)
     if len(full) < 2:
-        return []
+        return empty
     clusters = cluster_contexts_by_signature(full)
     if len(clusters) < 2:
         remaining = _remaining_contexts(built, candidate_times) or full
         clusters = cluster_contexts_by_signature(remaining)
     if len(clusters) < 2:
-        return []
+        return empty
     reps = [cluster["representative"] for cluster in clusters if _scoreable(cluster["representative"])]
     if len(reps) < 2:
         reps = [item for item in full if _scoreable(item)]
     if len(reps) < 2:
-        return []
+        return empty
     set_version = candidate_set_version([cluster["times"] for cluster in clusters])
     remaining_layers = _differing_layers(full)
     if not remaining_layers:
@@ -1133,10 +1390,8 @@ def discriminating_event_probes(
         events,
         d1_differs="d1" in remaining_layers or bool(scan.get("d1_candidates_differ")),
     )
-    if not domains:
-        return []
     lo, hi = birth_year + 5, min(now.year, birth_year + 80)
-    boundary_dates = _union_boundary_dates(reps, birth_date=birth_date, lo=lo, hi=hi)
+    boundary_dates = _union_boundary_dates(reps, birth_date=birth_date, lo=lo, hi=hi) if domains else []
     probes: list[dict[str, Any]] = []
     for domain in domains:
         if domain not in DOMAIN_CATALOG:
@@ -1202,23 +1457,152 @@ def discriminating_event_probes(
                 if activation_key not in existing:
                     kept.append(activation)
         probes.extend(kept)
+    probes.extend(_quality_distinguish_probes(
+        events,
+        clusters,
+        set_version=set_version,
+        holdout_ids=set(holdout_event_ids(events)),
+        holdout_keys=set(holdout_keys),
+    ))
     probes.sort(key=_probe_sort_key)
+    public, dropped = _partition_ranked_probes(probes)
+    assert_distinguish_contract(public)
+    return public, dropped
+
+
+def _partition_ranked_probes(
+    probes: Sequence[dict[str, Any]],
+) -> tuple[list[dict[str, Any]], list[dict[str, Any]]]:
     public: list[dict[str, Any]] = []
+    dropped: list[dict[str, Any]] = []
     seen: set[tuple[str, int, int, str]] = set()
     for row in probes:
-        if row.get("source") == "known_event_quality" or row.get("phase") != PROBE_PHASE_CANDIDATE_DISCRIMINATOR:
+        anchored_quality = (
+            row.get("source") == "known_event_quality"
+            and row.get("role") == "distinguish"
+            and row.get("target_evidence_id")
+        )
+        if row.get("source") == "known_event_quality" and not anchored_quality:
+            continue
+        if row.get("phase") != PROBE_PHASE_CANDIDATE_DISCRIMINATOR:
             continue
         if distinguish_contract_errors(row):
             continue
+        ranked = row if "raw_split_gain" in row else _apply_prior_ranking(dict(row))
+        priors = ranked.get("answer_priors") or _answer_priors_for(ranked)
+        if _dominant_existence_prior(ranked, priors):
+            dropped.append({
+                "semantic_key": ranked.get("semantic_key"),
+                "reason": "dominant_answer_prior",
+                "domain": ranked.get("domain"),
+                "year": ranked.get("year"),
+                "source": ranked.get("source"),
+                "answer_priors": priors,
+            })
+            continue
         if not isinstance(row.get("year"), int) or int(row["year"]) <= 0:
             continue
-        key = (str(row["domain"]), int(row["year"]), int(row.get("month") or 0), str(row["source"]))
-        encoded = str(row)
+        key = (str(ranked["domain"]), int(ranked["year"]), int(ranked.get("month") or 0), str(ranked["source"]))
+        encoded = str(ranked)
         if key in seen or "points" in encoded:
             continue
         seen.add(key)
-        public.append(row)
+        public.append(ranked)
         if len(public) >= MAX_PROBES:
             break
-    assert_distinguish_contract(public)
-    return public
+    public.sort(key=_probe_sort_key)
+    return public, dropped
+
+
+def discriminating_event_probe_set(
+    request: dict[str, Any],
+    built: dict[str, Any],
+    *,
+    scan: dict[str, Any],
+    candidate_times: Sequence[str],
+    representative_time: str | None,
+    precision_current: str | None = None,
+    today: date | None = None,
+) -> dict[str, list[dict[str, Any]]]:
+    probes, dropped = _discriminating_event_probe_lists(
+        request,
+        built,
+        scan=scan,
+        candidate_times=candidate_times,
+        representative_time=representative_time,
+        precision_current=precision_current,
+        today=today,
+    )
+    return {"probes": probes, "dropped": dropped}
+
+
+def discriminating_event_probes(
+    request: dict[str, Any],
+    built: dict[str, Any],
+    *,
+    scan: dict[str, Any],
+    candidate_times: Sequence[str],
+    representative_time: str | None,
+    precision_current: str | None = None,
+    today: date | None = None,
+) -> list[dict[str, Any]]:
+    probes, _dropped = _discriminating_event_probe_lists(
+        request,
+        built,
+        scan=scan,
+        candidate_times=candidate_times,
+        representative_time=representative_time,
+        precision_current=precision_current,
+        today=today,
+    )
+    return probes
+
+
+def prospective_event_windows(
+    request: dict[str, Any],
+    built: dict[str, Any],
+    *,
+    candidate_times: Sequence[str],
+    today: date | None = None,
+) -> list[dict[str, Any]]:
+    birth_date = str(request.get("birth_date") or "").strip()
+    if not birth_date:
+        return []
+    now = today or date.today()
+    lo, hi = now.year + 1, now.year + 3
+    full = _static_contexts(built)
+    if len(full) < 2:
+        return []
+    clusters = cluster_contexts_by_signature(full)
+    if len(clusters) < 2:
+        remaining = _remaining_contexts(built, candidate_times) or full
+        clusters = cluster_contexts_by_signature(remaining)
+    if len(clusters) < 2:
+        return []
+    labels = ("A", "B", "C")
+    rows: list[dict[str, Any]] = []
+    for index, cluster in enumerate(clusters[:3]):
+        context = cluster.get("representative") if isinstance(cluster, dict) else None
+        if not isinstance(context, dict):
+            continue
+        moon = (context.get("planet_longitudes") or {}).get("Moon")
+        if not isinstance(moon, (int, float)):
+            continue
+        starts = _vim_start_dates(birth_date, float(moon), lo, hi, include_pratyantar=True)
+        future = [item for item in starts if item.year >= lo]
+        if not future:
+            continue
+        start = min(future)
+        label = labels[index]
+        window = f"{start.year} 年 {start.month} 月附近"
+        rows.append({
+            "candidate_label": label,
+            "domain": "career",
+            "window_label": window,
+            "user_meaning": (
+                f"候选 {label} 预测下一次事业变动更可能在 {window}。"
+                "这是预测窗口，不是承诺；下次发生时回来补一条，可进一步分辨。"
+            ),
+            "used_for_scoring": False,
+        })
+    return rows
diff --git a/scripts/rectification/refinement_packet.py b/scripts/rectification/refinement_packet.py
index 57033ecc..ed3656c2 100644
--- a/scripts/rectification/refinement_packet.py
+++ b/scripts/rectification/refinement_packet.py
@@ -558,12 +558,13 @@ def build_refinement_packet(
     from scripts.rectification.case_holdout import reserved_holdout_events
     from scripts.rectification.event_probes import (
         candidate_contrast_opportunities,
-        discriminating_event_probes,
+        discriminating_event_probe_set,
         event_clarification_probes,
         evidence_collection_probes,
+        prospective_event_windows,
     )
     grid_times = list(built.get("candidate_times") or candidate_times)
-    probes = discriminating_event_probes(
+    bundle = discriminating_event_probe_set(
         request,
         built,
         scan=scan,
@@ -571,6 +572,8 @@ def build_refinement_packet(
         representative_time=representative_time,
         precision_current=str(stage.get("current") or "") or None,
     )
+    probes = bundle["probes"]
+    dropped = list(bundle["dropped"])
     clarification = event_clarification_probes(request)
     collection = evidence_collection_probes(request)
     opportunities = candidate_contrast_opportunities(
@@ -602,6 +605,11 @@ def build_refinement_packet(
         }
         for prompt in oos_blind_prompts(request)
     )
+    prospective = prospective_event_windows(
+        request,
+        built,
+        candidate_times=grid_times,
+    ) if not probes else []
     return {
         "window_scan": scan,
         "event_dasha_ledger": ledger,
@@ -616,6 +624,8 @@ def build_refinement_packet(
         "evidence_collection_probes": collection,
         "candidate_contrast_opportunities": opportunities,
         "holdout_validation_probes": holdout,
+        "dropped_probes": dropped,
+        "prospective_probes": prospective,
         "unique_minute_claim": False,
         "confirmation_allowed": False,
     }
diff --git a/scripts/rectification/scoring_service.py b/scripts/rectification/scoring_service.py
index 5ec44d14..92332124 100644
--- a/scripts/rectification/scoring_service.py
+++ b/scripts/rectification/scoring_service.py
@@ -10,10 +10,11 @@ from typing import Any
 
 from scripts.active_rectification_event_engine import compute_candidate_static_contexts, compute_event_candidate_rows
 from scripts.active_rectification_events import CandidateScoreRow
+from scripts.rectification.dasha_transition_proximity import merge_transition_proximity
 from scripts.rectification.contracts import LifeEvent, RectificationRequest, is_scoreable_event
 from scripts.rectification.case_holdout import holdout_event_ids
 
-ALGORITHM_VERSION = "rectification-v5-matrix-scoring-6"
+ALGORITHM_VERSION = "rectification-v5-matrix-scoring-7"
 INPUT_CONTRACT_VERSION = "rectification-calculation-spec-v4"
 PRECISION_WEIGHTS = {
     "day": 1.0,
@@ -179,11 +180,14 @@ def precision_weight(precision: str) -> float:
 
 def public_technique_layers(domain: str, rule_ids: Sequence[str]) -> list[str]:
     """Public methods actually computed for this event. Career always lists D1-10 and D10."""
-    layers = {
-        rule.split(":", 1)[0]
-        for rule in rule_ids
-        if not rule.startswith(("event_kind:", "event_kind_profile:"))
-    }
+    layers: set[str] = set()
+    for rule in rule_ids:
+        if rule.startswith(("event_kind:", "event_kind_profile:")):
+            continue
+        if "transition_proximity" in rule:
+            layers.add("dasha-transition-proximity")
+            continue
+        layers.add(rule.split(":", 1)[0])
     if domain == "career":
         layers.update({"d1-rashi", "d10-dashamsa"})
     elif domain == "family":
@@ -235,6 +239,7 @@ def scoreable_request(request: RectificationRequest) -> RectificationRequest:
 def build_event_contribution_matrix(
     request: RectificationRequest,
     row_provider: Callable[[dict[str, Any]], Sequence[CandidateScoreRow]] | None = None,
+    static_contexts: Sequence[dict[str, Any]] | None = None,
 ) -> dict[str, Any]:
     scoring_request = scoreable_request(request)
     if not scoring_request["events"]:
@@ -242,7 +247,8 @@ def build_event_contribution_matrix(
             "candidate_times": [], "matrix": {}, "date_sensitivity": [],
             "missing_layers": [], "static_contexts": None,
         }
-    static_contexts = None if row_provider is not None else compute_candidate_static_contexts(scoring_request)
+    if static_contexts is None and row_provider is None:
+        static_contexts = compute_candidate_static_contexts(scoring_request)
     provider = row_provider or (lambda value: compute_event_candidate_rows(value, static_contexts=static_contexts))
     matrix: dict[str, dict[str, dict[str, Any]]] = defaultdict(dict)
     missing_layers: set[str] = set()
@@ -288,9 +294,18 @@ def build_event_contribution_matrix(
             "score_variance": round(variance, 6),
             "sample_winners": winners,
         })
+    matrix_payload = dict(matrix)
+    if static_contexts:
+        merge_transition_proximity(
+            matrix_payload,
+            scoring_request["events"],
+            static_contexts,
+            scoring_request["birth_date"],
+            public_technique_layers=public_technique_layers,
+        )
     return {
         "candidate_times": candidate_grid or [],
-        "matrix": dict(matrix),
+        "matrix": matrix_payload,
         "date_sensitivity": date_sensitivity,
         "missing_layers": sorted(missing_layers),
         "static_contexts": static_contexts,
diff --git a/tests/test_candidate_discriminator_contract.py b/tests/test_candidate_discriminator_contract.py
index 5353debd..eccd7136 100644
--- a/tests/test_candidate_discriminator_contract.py
+++ b/tests/test_candidate_discriminator_contract.py
@@ -140,7 +140,7 @@ class DiscriminatorContractTest(unittest.TestCase):
             ["distinguish_empty_expected_outcomes"],
         )
 
-    def test_quality_never_enters_discriminating_event_probes(self) -> None:
+    def test_quality_clarify_stays_and_anchored_distinguish_is_gated(self) -> None:
         built = {
             "static_contexts": [
                 _context("05:13", d4_asc=1, d9_asc=1),
@@ -164,8 +164,12 @@ class DiscriminatorContractTest(unittest.TestCase):
             representative_time="05:13",
             today=date(2026, 8, 22),
         )
-        self.assertFalse(any(item.get("source") == "known_event_quality" for item in probes))
         self.assertFalse(any(item.get("role") == "distinguish" and distinguish_contract_errors(item) for item in probes))
+        quality = [item for item in probes if item.get("source") == "known_event_quality"]
+        for item in quality:
+            self.assertEqual(item.get("role"), "distinguish")
+            self.assertTrue(item.get("target_evidence_id"))
+            self.assertEqual(item.get("choice_kind"), "event_quality")
         clarification = event_clarification_probes(request)
         self.assertTrue(any(item.get("source") == "known_event_quality" for item in clarification))
         self.assertTrue(all(item.get("phase") == "event_clarification" for item in clarification))
@@ -281,7 +285,11 @@ class DiscriminatorContractTest(unittest.TestCase):
                 representative_time="04:50",
                 today=date(2026, 8, 22),
             )
-            self.assertFalse(any(item.get("source") == "known_event_quality" for item in probes))
+            self.assertFalse(any(
+                item.get("source") == "known_event_quality"
+                and not item.get("target_evidence_id")
+                for item in probes
+            ))
             for probe in probes:
                 self.assertEqual(distinguish_contract_errors(probe), [])
                 self.assertGreater(float(probe["information_gain"]), 0)
diff --git a/tests/test_rectification_engine_convergence.py b/tests/test_rectification_engine_convergence.py
new file mode 100644
index 00000000..16ddb1c3
--- /dev/null
+++ b/tests/test_rectification_engine_convergence.py
@@ -0,0 +1,420 @@
+"""Round 2 engine-convergence invariants (TASK-rectification-engine-convergence-20260901)."""
+
+from __future__ import annotations
+
+import unittest
+from datetime import date, datetime
+from pathlib import Path
+
+from scripts.rectification.candidate_contrast import distinguish_contract_errors
+from scripts.rectification.contracts import normalize_rectification_request
+from scripts.rectification.event_probes import (
+    _apply_prior_ranking,
+    _dominant_existence_prior,
+    _partition_ranked_probes,
+    _vim_start_dates,
+    discriminating_event_probes,
+    event_clarification_probes,
+)
+from scripts.rectification.refinement_packet import build_refinement_packet, window_scan
+from scripts.rectification.scoring_service import build_event_contribution_matrix
+
+REPO_ROOT = Path(__file__).resolve().parents[1]
+PLANETS = {
+    "Sun": 12.0,
+    "Moon": 100.0,
+    "Mars": 40.0,
+    "Mercury": 20.0,
+    "Jupiter": 80.0,
+    "Venus": 50.0,
+    "Saturn": 200.0,
+    "Rahu": 310.0,
+    "Ketu": 130.0,
+}
+GRID_TIMES = [
+    f"{4 + (45 + offset) // 60:02d}:{(45 + offset) % 60:02d}"
+    for offset in range(31)
+]
+SPLIT_EVENT_ID = "00000000-0000-4000-8000-000000000003"
+
+
+def _varga(asc: int, planet_sign: int) -> dict:
+    return {
+        "Ascendant": {"sign_idx": asc},
+        **{name: {"sign_idx": planet_sign} for name in PLANETS},
+    }
+
+
+def _context(
+    time: str,
+    *,
+    moon: float = 100.0,
+    d4_asc: int = 1,
+    d9_asc: int = 1,
+    d10_asc: int = 1,
+    d12_asc: int = 1,
+    d24_asc: int = 1,
+    sun_house: int = 10,
+) -> dict:
+    hour, minute = (int(part) for part in time.split(":"))
+    planets = {**PLANETS, "Moon": moon}
+    natal_planets = {
+        name: {"house": sun_house if name != "Moon" else 4, "lon": lon}
+        for name, lon in planets.items()
+    }
+    return {
+        "candidate_at": datetime(1997, 8, 8, hour, minute),
+        "chart": {"ascendant": {"lon": 10.0, "sign": "Aries"}, "planets": natal_planets},
+        "planet_longitudes": dict(planets),
+        "ascendant_index": 0,
+        "varga_charts": {
+            "D4": _varga(d4_asc, 1),
+            "D9": _varga(d9_asc, 1),
+            "D10": _varga(d10_asc, 1),
+            "D5": _varga(1, 1),
+            "D24": _varga(d24_asc, 1),
+            "D12": _varga(d12_asc, 1),
+            "D7": _varga(1, 1),
+            "D3": _varga(1, 1),
+        },
+        "arudha_padas": {},
+        "feature": {
+            "time": time,
+            "ascendant_sign_index": 0,
+            "varga_ascendants": {
+                "D4": d4_asc, "D9": d9_asc, "D10": d10_asc, "D5": 1, "D24": d24_asc, "D12": d12_asc,
+            },
+        },
+    }
+
+
+def _accident_events() -> list[dict]:
+    return [
+        {
+            "id": "00000000-0000-4000-8000-000000000001",
+            "domain": "education",
+            "event_kind": "education_start",
+            "date_start": "2016-01-01",
+            "date_end": "2016-12-31",
+            "precision": "year",
+            "summary": "上大学",
+        },
+        {
+            "id": "00000000-0000-4000-8000-000000000002",
+            "domain": "relationship",
+            "event_kind": "relationship_start",
+            "date_start": "2024-05-01",
+            "date_end": "2024-05-31",
+            "precision": "month",
+            "summary": "开始认真交往",
+        },
+        {
+            "id": SPLIT_EVENT_ID,
+            "domain": "relationship",
+            "event_kind": "relationship_end",
+            "date_start": "2024-08-08",
+            "date_end": "2024-08-08",
+            "precision": "day",
+            "summary": "分手",
+        },
+        {
+            "id": "00000000-0000-4000-8000-000000000004",
+            "domain": "career",
+            "event_kind": "career_entry",
+            "date_start": "2020-04-01",
+            "date_end": "2020-04-30",
+            "precision": "month",
+            "summary": "实习入职",
+        },
+        {
+            "id": "00000000-0000-4000-8000-000000000005",
+            "domain": "career",
+            "event_kind": "career_exit",
+            "date_start": "2020-10-01",
+            "date_end": "2020-10-31",
+            "precision": "month",
+            "summary": "实习结束离职",
+        },
+    ]
+
+
+def _accident_request() -> dict:
+    return normalize_rectification_request(
+        {
+            "birth_date": "1997-08-08",
+            "start_time": "04:45",
+            "end_time": "05:15",
+            "lat": 36.420487,
+            "lon": 114.209936,
+            "tz": 8,
+            "events": _accident_events(),
+        },
+        today=date(2026, 8, 22),
+    )
+
+
+def _equal_rows(payload: dict) -> list[dict]:
+    event = payload["events"][0]
+    return [
+        {
+            "time": time,
+            "score": 10,
+            "evidence": [{
+                "event_id": event["id"],
+                "domain": event["domain"],
+                "candidate_time": time,
+                "rule_ids": ["vim_md_domain_house"],
+                "points": 10,
+            }],
+            "missing_layers": [],
+        }
+        for time in GRID_TIMES
+    ]
+
+
+def _moons_with_split_proximity() -> tuple[float, float]:
+    event_at = date(2024, 8, 8)
+    ranked: list[tuple[int, float]] = []
+    for moon in (100.0, 100.5, 101.0, 103.0, 110.0):
+        starts = _vim_start_dates("1997-08-08", moon, 2023, 2025)
+        if not starts:
+            continue
+        nearest = min(abs((item - event_at).days) for item in starts)
+        ranked.append((nearest, moon))
+    ranked.sort()
+    if len(ranked) < 2 or ranked[0][0] == ranked[-1][0]:
+        raise AssertionError("fixture moons do not split AD/PD proximity")
+    return ranked[0][1], ranked[-1][1]
+
+
+def _probe_events(*, education_count: int = 2, d24_split: bool = True) -> tuple[dict, dict]:
+    events = [
+        {
+            "id": f"00000000-0000-4000-8000-{index:012d}",
+            "domain": "education",
+            "event_kind": "education_start",
+            "summary": "入学",
+            "date": f"{2015 + index}-09-01",
+            "precision": "month",
+        }
+        for index in range(1, education_count + 1)
+    ]
+    events.extend([
+        {
+            "id": "00000000-0000-4000-8000-000000000011",
+            "domain": "career",
+            "event_kind": "career_entry",
+            "summary": "入职",
+            "date": "2018-07-01",
+            "precision": "month",
+        },
+        {
+            "id": "00000000-0000-4000-8000-000000000012",
+            "domain": "career",
+            "event_kind": "career_change",
+            "summary": "换岗",
+            "date": "2020-04-01",
+            "precision": "month",
+        },
+        {
+            "id": "00000000-0000-4000-8000-000000000013",
+            "domain": "relationship",
+            "event_kind": "relationship_start",
+            "summary": "相识",
+            "date": "2021-08-01",
+            "precision": "month",
+        },
+    ])
+    late = 2 if d24_split else 1
+    built = {
+        "static_contexts": [
+            _context("05:00", d24_asc=1, d12_asc=1, d9_asc=1, d10_asc=1),
+            _context("05:07", d24_asc=late, d12_asc=late, d9_asc=late, d10_asc=late),
+        ]
+    }
+    return {"birth_date": "1997-08-08", "events": events}, built
+
+
+class EngineConvergenceProximityTests(unittest.TestCase):
+    def test_day_event_proximity_splits_adjacent_minutes_in_accident_shape(self) -> None:
+        from scripts.rectification.dasha_transition_proximity import (
+            DAY_MAX_POINTS,
+            score_transition_proximity,
+        )
+
+        closer_moon, farther_moon = _moons_with_split_proximity()
+        event_at = date(2024, 8, 8)
+        closer_starts = _vim_start_dates("1997-08-08", closer_moon, 2023, 2025)
+        farther_starts = _vim_start_dates("1997-08-08", farther_moon, 2023, 2025)
+        closer_delta = min(abs((item - event_at).days) for item in closer_starts)
+        farther_delta = min(abs((item - event_at).days) for item in farther_starts)
+        closer_points = score_transition_proximity(
+            event_date=event_at,
+            precision="day",
+            vim_starts=closer_starts,
+            narayana_starts=[],
+        )
+        farther_points = score_transition_proximity(
+            event_date=event_at,
+            precision="day",
+            vim_starts=farther_starts,
+            narayana_starts=[],
+        )
+        self.assertLessEqual(float(closer_points["points"]), DAY_MAX_POINTS)
+        self.assertNotEqual(closer_points["points"], farther_points["points"])
+        self.assertEqual(
+            closer_points["points"] > farther_points["points"],
+            closer_delta < farther_delta,
+        )
+        self.assertTrue(
+            any(str(rule).startswith("vim_transition_proximity_") for rule in closer_points["rule_ids"])
+        )
+
+        contexts = [
+            _context(time, moon=closer_moon if time == "05:00" else farther_moon)
+            for time in GRID_TIMES
+        ]
+        request = _accident_request()
+        built = build_event_contribution_matrix(
+            request,
+            row_provider=_equal_rows,
+            static_contexts=contexts,
+        )
+        self.assertEqual(len(built["candidate_times"]), 31)
+        cell_0500 = built["matrix"][SPLIT_EVENT_ID]["05:00"]
+        cell_0507 = built["matrix"][SPLIT_EVENT_ID]["05:07"]
+        self.assertNotEqual(cell_0500["points"], cell_0507["points"])
+        self.assertEqual(
+            cell_0500["points"] > cell_0507["points"],
+            closer_delta < farther_delta,
+        )
+        self.assertTrue(
+            any("transition_proximity" in str(rule) for rule in cell_0500["rule_ids"])
+            or any("transition_proximity" in str(rule) for rule in cell_0507["rule_ids"])
+        )
+        self.assertLessEqual(abs(cell_0500["points"] - cell_0507["points"]), DAY_MAX_POINTS)
+
+    def test_algorithm_and_policy_versions_change_with_proximity_semantics(self) -> None:
+        from scripts.rectification.decision_policy import POLICY_VERSION
+        from scripts.rectification.scoring_service import ALGORITHM_VERSION
+
+        self.assertNotEqual(ALGORITHM_VERSION, "rectification-v5-matrix-scoring-6")
+        self.assertNotEqual(POLICY_VERSION, "rectification-candidate-policy-v2")
+
+
+class EngineConvergenceProbeTests(unittest.TestCase):
+    def test_anchored_quality_outranks_family_existence_and_drops_dominant_priors(self) -> None:
+        request, built = _probe_events()
+        probes = discriminating_event_probes(
+            request,
+            built,
+            scan=window_scan(built),
+            candidate_times=["05:00", "05:07"],
+            representative_time="05:00",
+            today=date(2026, 8, 22),
+        )
+        quality = [item for item in probes if item.get("source") == "known_event_quality"]
+        self.assertTrue(quality)
+        self.assertEqual(quality[0]["role"], "distinguish")
+        self.assertTrue(quality[0].get("target_evidence_id"))
+        self.assertFalse(distinguish_contract_errors(quality[0]))
+        ranked_family = _apply_prior_ranking({
+            "domain": "family",
+            "source": "dasha_boundary",
+            "choice_kind": "existence",
+            "information_gain": float(quality[0].get("raw_split_gain") or quality[0].get("information_gain") or 0),
+            "semantic_key": "family.2024.existence",
+            "year": 2024,
+            "window_span_years": 3,
+            "role": "distinguish",
+            "phase": "candidate_discriminator",
+            "candidate_ids": ["05:00", "05:07"],
+            "expected_outcomes": [
+                {"answer_class": "yes", "supports": ["05:00"], "conflicts": ["05:07"]},
+                {"answer_class": "no", "supports": ["05:07"], "conflicts": ["05:00"]},
+            ],
+        })
+        self.assertTrue(_dominant_existence_prior(ranked_family, ranked_family["answer_priors"]))
+        _, ranked_dropped = _partition_ranked_probes([quality[0], ranked_family])
+        self.assertTrue(
+            any(item.get("reason") == "dominant_answer_prior" for item in ranked_dropped),
+            ranked_dropped,
+        )
+        packet = build_refinement_packet(
+            request,
+            built,
+            representative_time="05:00",
+            candidate_times=["05:00", "05:07"],
+        )
+        dropped = list(packet.get("dropped_probes") or []) + ranked_dropped
+        self.assertTrue(
+            any(item.get("reason") == "dominant_answer_prior" for item in dropped),
+            dropped,
+        )
+        self.assertFalse(
+            any(
+                item.get("choice_kind") == "existence"
+                and float(max((item.get("answer_priors") or {}).values() or [0])) > 0.8
+                for item in probes
+            )
+        )
+
+    def test_quality_distinguish_requires_varga_type_split_and_caps_at_two(self) -> None:
+        same_request, same_built = _probe_events(d24_split=False)
+        same_probes = discriminating_event_probes(
+            same_request,
+            same_built,
+            scan=window_scan(same_built),
+            candidate_times=["05:00", "05:07"],
+            representative_time="05:00",
+            today=date(2026, 8, 22),
+        )
+        self.assertFalse(any(item.get("source") == "known_event_quality" for item in same_probes))
+        clarification = event_clarification_probes(same_request)
+        self.assertTrue(any(item.get("source") == "known_event_quality" for item in clarification))
+
+        split_request, split_built = _probe_events(education_count=4, d24_split=True)
+        split_probes = discriminating_event_probes(
+            split_request,
+            split_built,
+            scan=window_scan(split_built),
+            candidate_times=["05:00", "05:07"],
+            representative_time="05:00",
+            today=date(2026, 8, 22),
+        )
+        quality = [item for item in split_probes if item.get("source") == "known_event_quality"]
+        self.assertTrue(quality)
+        self.assertLessEqual(len(quality), 2)
+        for item in quality:
+            self.assertEqual(item["role"], "distinguish")
+            self.assertTrue(item.get("target_evidence_id"))
+            self.assertTrue(item.get("display_date_label"))
+            self.assertEqual(item.get("choice_kind"), "event_quality")
+            self.assertFalse(distinguish_contract_errors(item))
+
+
+class EngineConvergenceAnswerKeyTests(unittest.TestCase):
+    def test_scripts_and_frontend_have_no_answer_key_literals(self) -> None:
+        forbidden = ("target" + "_minute", "pl9_" + "1993", "regression" + "_only")
+        hits: list[str] = []
+        roots = (
+            REPO_ROOT / "scripts",
+            REPO_ROOT / "frontend" / "src",
+            REPO_ROOT / "frontend" / "tests",
+        )
+        skip_parts = {"node_modules", ".next", "dist"}
+        for root in roots:
+            for path in root.rglob("*"):
+                if not path.is_file() or path.suffix not in {".py", ".ts", ".tsx", ".js", ".mjs"}:
+                    continue
+                if any(part in skip_parts for part in path.parts):
+                    continue
+                text = path.read_text(encoding="utf-8", errors="ignore")
+                for token in forbidden:
+                    if token in text:
+                        hits.append(f"{path.relative_to(REPO_ROOT)}:{token}")
+        self.assertEqual(hits, [])
+
+
+if __name__ == "__main__":
+    unittest.main()
diff --git a/tests/test_rectification_event_probes.py b/tests/test_rectification_event_probes.py
index 5a2f5c2d..0dee4d0e 100644
--- a/tests/test_rectification_event_probes.py
+++ b/tests/test_rectification_event_probes.py
@@ -450,7 +450,10 @@ class EventProbesTest(unittest.TestCase):
             }],
         )
         probes = _probes(request, built, ["05:13", "05:40"], "05:13", precision_current="d5_refine")
-        existence = [item for item in probes if item["domain"] == "education"]
+        existence = [
+            item for item in probes
+            if item["domain"] == "education" and item.get("choice_kind") != "event_quality"
+        ]
         self.assertFalse(any(item["year"] in {2015, 2016, 2017} for item in existence))
 
     def test_signature_clusters_use_full_birth_window(self) -> None:
@@ -674,7 +677,7 @@ class EventProbesTest(unittest.TestCase):
             self.assertGreaterEqual(int(probe["year"]), 1900)
             self.assertNotEqual(int(probe["year"]), 0)
             self.assertEqual(distinguish_contract_errors(probe), [])
-            self.assertIn(probe["source"], {"dasha_boundary", "dasha_activation"})
+            self.assertIn(probe["source"], {"dasha_boundary", "dasha_activation", "known_event_quality"})
             self.assertGreaterEqual(len(probe["style_options"]), 4)
             self.assertGreaterEqual(len(probe["candidate_ids"]), 2)
             self.assertGreaterEqual(len(probe["expected_outcomes"]), 2)
diff --git a/tests/test_rectification_v5_services.py b/tests/test_rectification_v5_services.py
index d6e4caa4..df461292 100644
--- a/tests/test_rectification_v5_services.py
+++ b/tests/test_rectification_v5_services.py
@@ -488,7 +488,7 @@ class RectificationV5ServicesTest(unittest.TestCase):
         self.assertNotEqual(receipt["gates"]["exact_confirmation"]["external_validation_status"], "fail")
 
         self.assertEqual(first["decision_receipt"], receipt)
-        self.assertEqual(first["decision_policy_version"], "rectification-candidate-policy-v2")
+        self.assertEqual(first["decision_policy_version"], "rectification-candidate-policy-v3")
         self.assertTrue(receipt["display_allowed"])
         self.assertTrue(receipt["accept_allowed"])
         self.assertFalse(receipt["confirm_allowed"])