diff --git a/runtime/refactor_baseline_v1.yaml b/runtime/refactor_baseline_v1.yaml
index d7fcb63..ed753cb 100644
--- a/runtime/refactor_baseline_v1.yaml
+++ b/runtime/refactor_baseline_v1.yaml
@@ -15,5 +15,5 @@
       "keep package scripts within release envelope"
     ]
   },
-  "source_zip_sha256": "49f64b3773ba3c19fa8323d0b08833928c637935483039579bb8ab22a391f70c"
+  "source_zip_sha256": "4de4a7b1217ef5d5375b3b1ea1209f738719e79c4c3c0954e9e96a9dc0d8607e"
 }
\ No newline at end of file
diff --git a/spec/risk/portfolio_exposure.yaml b/spec/risk/portfolio_exposure.yaml
index 8267acb..bcb9934 100644
--- a/spec/risk/portfolio_exposure.yaml
+++ b/spec/risk/portfolio_exposure.yaml
@@ -13,7 +13,7 @@ portfolio_exposure_framework:
   exposure_layers:
     direct_core_leaders: ["삼성전자", "SK하이닉스"]
     duplicate_beta: ["KODEX 반도체", "동일 섹터 ETF"]
-    tactical_satellites: ["방산", "조선", "전력기기", "건설/EPC", "기타 고베타"]
+    tactical_satellites: ["방산", "조선", "전력설비", "건설", "플랜트/EPC", "로보틱스", "기타 고베타"]
     cash: ["현금", "MMF", "RP", "단기채 ETF"]
   valid_trim_reasons:
     - "벤치마크 대비 초과비중이 허용밴드를 초과하고 가격 추세가 훼손됨"
diff --git a/src/gas/core/gas_lib.gs b/src/gas/core/gas_lib.gs
index 8516f95..5423187 100644
--- a/src/gas/core/gas_lib.gs
+++ b/src/gas/core/gas_lib.gs
@@ -1,5 +1,5 @@
 // gas_lib.gs - Common utilities & static features
-// Last Updated: 2026-06-14 20:48:30 KST
+// Last Updated: 2026-06-15 02:20:50 KST
 // Math/KRX utils, sheet I/O, sector flow, Web API, static runners
 // GAS global scope: functions in gas_data_feed.gs / gas_data_collect.gs callable directly
 //
@@ -593,7 +593,14 @@ const DEFAULT_SECTOR_UNIVERSE_V2 = [
     { code: "062040", name: "산일전기", weight: 0.10 },
     { code: "298040", name: "효성중공업", weight: 0.10 },
   ]},
-  { sector: "방산", proxyTicker: "012450", proxyName: "한화에어로스페이스", proxyType: "대표주", baseTicker: "069500", constituents: [
+  { sector: "전력설비", proxyTicker: "491820", proxyName: "HANARO 전력설비투자", proxyType: "ETF", baseTicker: "069500", constituents: [
+    { code: "010120", name: "LS ELECTRIC", weight: 0.28 },
+    { code: "267260", name: "HD현대일렉트릭", weight: 0.28 },
+    { code: "298040", name: "효성중공업", weight: 0.18 },
+    { code: "006260", name: "LS", weight: 0.14 },
+    { code: "099440", name: "두산에너빌리티", weight: 0.12 },
+  ]},
+  { sector: "방산", proxyTicker: "463250", proxyName: "TIGER K방산&우주", proxyType: "ETF", baseTicker: "069500", constituents: [
     { code: "012450", name: "한화에어로스페이스", weight: 0.45 },
     { code: "079550", name: "LIG넥스원", weight: 0.25 },
     { code: "047810", name: "한국항공우주", weight: 0.15 },
@@ -605,23 +612,49 @@ const DEFAULT_SECTOR_UNIVERSE_V2 = [
     { code: "009540", name: "HD한국조선해양", weight: 0.20 },
     { code: "494670", name: "TIGER 조선TOP10", weight: 0.15, isEtf: true },
   ]},
-  { sector: "건설/EPC", proxyTicker: "028050", proxyName: "삼성E&A", proxyType: "대표주", baseTicker: "069500", constituents: [
-    { code: "028050", name: "삼성E&A", weight: 0.40 },
-    { code: "000720", name: "현대건설", weight: 0.30 },
-    { code: "006360", name: "GS건설", weight: 0.20 },
-    { code: "047040", name: "대우건설", weight: 0.10 },
+  { sector: "건설", proxyTicker: "117700", proxyName: "KODEX 건설", proxyType: "ETF", baseTicker: "069500", constituents: [
+    { code: "000720", name: "현대건설", weight: 0.35 },
+    { code: "006360", name: "GS건설", weight: 0.25 },
+    { code: "047040", name: "대우건설", weight: 0.20 },
+    { code: "294870", name: "HDC현대산업개발", weight: 0.20 },
+  ]},
+  { sector: "플랜트/EPC", proxyTicker: "454320", proxyName: "HANARO CAPEX설비투자iSelect", proxyType: "ETF", baseTicker: "069500", constituents: [
+    { code: "028050", name: "삼성E&A", weight: 0.35 },
+    { code: "010120", name: "LS ELECTRIC", weight: 0.20 },
+    { code: "267260", name: "HD현대일렉트릭", weight: 0.20 },
+    { code: "298040", name: "효성중공업", weight: 0.15 },
+    { code: "099440", name: "두산에너빌리티", weight: 0.10 },
   ]},
   { sector: "자동차", proxyTicker: "091180", proxyName: "TIGER 자동차", proxyType: "ETF", baseTicker: "069500", constituents: [
     { code: "005380", name: "현대차", weight: 0.45 },
     { code: "000270", name: "기아", weight: 0.40 },
     { code: "012330", name: "현대모비스", weight: 0.15 },
   ]},
-  { sector: "금융/은행", proxyTicker: "091170", proxyName: "KODEX 은행", proxyType: "ETF", baseTicker: "069500", constituents: [
+  { sector: "은행", proxyTicker: "091170", proxyName: "KODEX 은행", proxyType: "ETF", baseTicker: "069500", constituents: [
     { code: "105560", name: "KB금융", weight: 0.30 },
     { code: "055550", name: "신한지주", weight: 0.30 },
     { code: "086790", name: "하나금융지주", weight: 0.20 },
     { code: "316140", name: "우리금융지주", weight: 0.10 },
-    { code: "003540", name: "대신증권", weight: 0.10 },
+    { code: "024110", name: "기업은행", weight: 0.10 },
+  ]},
+  { sector: "증권", proxyTicker: "0111J0", proxyName: "HANARO 증권고배당TOP3플러스", proxyType: "ETF", baseTicker: "069500", constituents: [
+    { code: "071050", name: "한국금융지주", weight: 0.2135 },
+    { code: "006800", name: "미래에셋증권", weight: 0.1934 },
+    { code: "005940", name: "NH투자증권", weight: 0.1911 },
+    { code: "016360", name: "삼성증권", weight: 0.1434 },
+    { code: "039490", name: "키움증권", weight: 0.1373 },
+  ]},
+  { sector: "지주회사", proxyTicker: "307520", proxyName: "TIGER 지주회사", proxyType: "ETF", baseTicker: "069500", constituents: [
+    { code: "180640", name: "한진칼", weight: 0.1535 },
+    { code: "267250", name: "HD현대", weight: 0.0943 },
+    { code: "034730", name: "SK", weight: 0.0884 },
+    { code: "000150", name: "두산", weight: 0.0878 },
+    { code: "005490", name: "POSCO홀딩스", weight: 0.0763 },
+    { code: "003550", name: "LG", weight: 0.0752 },
+    { code: "006260", name: "LS", weight: 0.0705 },
+    { code: "078930", name: "GS", weight: 0.0498 },
+    { code: "001040", name: "CJ", weight: 0.0477 },
+    { code: "010060", name: "OCI홀딩스", weight: 0.0240 },
   ]},
   { sector: "2차전지", proxyTicker: "305720", proxyName: "KODEX 2차전지산업", proxyType: "ETF", baseTicker: "069500", constituents: [
     { code: "373220", name: "LG에너지솔루션", weight: 0.40 },
@@ -635,12 +668,29 @@ const DEFAULT_SECTOR_UNIVERSE_V2 = [
     { code: "128940", name: "한미약품", weight: 0.15 },
     { code: "000100", name: "유한양행", weight: 0.10 },
   ]},
-  { sector: "원전", proxyTicker: "099440", proxyName: "두산에너빌리티", proxyType: "대표주", baseTicker: "069500", constituents: [
+  { sector: "원전", proxyTicker: "434730", proxyName: "HANARO 원자력iSelect", proxyType: "ETF", baseTicker: "069500", constituents: [
     { code: "099440", name: "두산에너빌리티", weight: 0.45 },
     { code: "023450", name: "한전기술", weight: 0.25 },
     { code: "015760", name: "한국전력", weight: 0.20 },
     { code: "071320", name: "지역난방공사", weight: 0.10 },
   ]},
+  { sector: "로보틱스", proxyTicker: "0190C0", proxyName: "RISE 현대차고정피지컬AI", proxyType: "ETF", baseTicker: "069500", constituents: [
+    { code: "005380", name: "현대차", weight: 0.2402 },
+    { code: "012330", name: "현대모비스", weight: 0.1588 },
+    { code: "011070", name: "LG이노텍", weight: 0.1450 },
+    { code: "000270", name: "기아", weight: 0.1234 },
+    { code: "307950", name: "현대오토에버", weight: 0.0899 },
+    { code: "277810", name: "레인보우로보틱스", weight: 0.0673 },
+    { code: "064400", name: "LG씨엔에스", weight: 0.0519 },
+    { code: "454910", name: "두산로보틱스", weight: 0.0367 },
+    { code: "108490", name: "로보티즈", weight: 0.0240 },
+    { code: "058610", name: "에스피지", weight: 0.0173 },
+    { code: "010620", name: "현대미포", weight: 0.0135 },
+    { code: "009540", name: "HD한국조선해양", weight: 0.0135 },
+    { code: "011210", name: "현대위아", weight: 0.0109 },
+    { code: "121600", name: "나노신소재", weight: 0.0040 },
+    { code: "028050", name: "삼성E&A", weight: 0.0034 },
+  ]},
   { sector: "소비재", proxyTicker: "139220", proxyName: "TIGER 생활소비재", proxyType: "ETF", baseTicker: "069500", constituents: [
     { code: "028260", name: "삼성물산", weight: 0.35 },
     { code: "097950", name: "CJ제일제당", weight: 0.25 },
@@ -663,6 +713,7 @@ function normalizeSectorName_(sector) {
   if (s === "바이오/헬스케어") return "바이오";
   if (s === "원전/에너지") return "원전";
   if (s === "소비재/유통") return "소비재";
+  if (s === "건설/EPC") return "플랜트/EPC";
   return s;
 }
 
@@ -679,17 +730,52 @@ function readSectorUniverse_() {
   const sheet = ss.getSheetByName("sector_universe");
   if (!sheet) {
     writeDefaultSectorUniverseSheet_();
-    return DEFAULT_SECTOR_UNIVERSE_V2;
+    return DEFAULT_SECTOR_UNIVERSE_V2.map(sector => ({
+      ...sector,
+      source: sector.source || "DEFAULT_TEMPLATE",
+      sourceUrl: sector.sourceUrl || "",
+      sourceAsOf: sector.sourceAsOf || "",
+      constituents: sector.constituents.map(c => ({
+        ...c,
+        source: c.source || sector.source || "DEFAULT_TEMPLATE",
+        sourceUrl: c.sourceUrl || sector.sourceUrl || "",
+        sourceAsOf: c.sourceAsOf || sector.sourceAsOf || "",
+      })),
+    }));
   }
   const data = sheet.getDataRange().getValues();
   if (data.length < 3) {
     writeDefaultSectorUniverseSheet_();
-    return DEFAULT_SECTOR_UNIVERSE_V2;
+    return DEFAULT_SECTOR_UNIVERSE_V2.map(sector => ({
+      ...sector,
+      source: sector.source || "DEFAULT_TEMPLATE",
+      sourceUrl: sector.sourceUrl || "",
+      sourceAsOf: sector.sourceAsOf || "",
+      constituents: sector.constituents.map(c => ({
+        ...c,
+        source: c.source || sector.source || "DEFAULT_TEMPLATE",
+        sourceUrl: c.sourceUrl || sector.sourceUrl || "",
+        sourceAsOf: c.sourceAsOf || sector.sourceAsOf || "",
+      })),
+    }));
   }
   const hdr = data[1].map(h => String(h).trim());
   const idx = name => hdr.indexOf(name);
   const required = ["Sector","Proxy_Ticker","Constituent_Code","Weight"];
-  if (required.some(h => idx(h) < 0)) return DEFAULT_SECTOR_UNIVERSE_V2;
+  if (required.some(h => idx(h) < 0)) {
+    return DEFAULT_SECTOR_UNIVERSE_V2.map(sector => ({
+      ...sector,
+      source: sector.source || "DEFAULT_TEMPLATE",
+      sourceUrl: sector.sourceUrl || "",
+      sourceAsOf: sector.sourceAsOf || "",
+      constituents: sector.constituents.map(c => ({
+        ...c,
+        source: c.source || sector.source || "DEFAULT_TEMPLATE",
+        sourceUrl: c.sourceUrl || sector.sourceUrl || "",
+        sourceAsOf: c.sourceAsOf || sector.sourceAsOf || "",
+      })),
+    }));
+  }
 
   const map = {};
   for (let i = 2; i < data.length; i++) {
@@ -706,6 +792,9 @@ function readSectorUniverse_() {
         proxyName: idx("Proxy_Name") >= 0 ? String(data[i][idx("Proxy_Name")] ?? "").trim() : "",
         proxyType: idx("Proxy_Type") >= 0 ? String(data[i][idx("Proxy_Type")] ?? "").trim() : "",
         baseTicker: idx("Base_Ticker") >= 0 ? normalizeTickerCode(data[i][idx("Base_Ticker")]) : "069500",
+        source: idx("Source") >= 0 ? String(data[i][idx("Source")] ?? "").trim() : "SHEET_INPUT",
+        sourceUrl: idx("Source_URL") >= 0 ? String(data[i][idx("Source_URL")] ?? "").trim() : "",
+        sourceAsOf: idx("Source_AsOf") >= 0 ? String(data[i][idx("Source_AsOf")] ?? "").trim() : "",
         constituents: [],
       };
     }
@@ -714,16 +803,59 @@ function readSectorUniverse_() {
       name: idx("Constituent_Name") >= 0 ? String(data[i][idx("Constituent_Name")] ?? "").trim() : "",
       weight,
       isEtf: idx("Is_ETF") >= 0 ? boolFromSheet_(data[i][idx("Is_ETF")], false) : false,
+      source: idx("Source") >= 0 ? String(data[i][idx("Source")] ?? "").trim() : "SHEET_INPUT",
+      transportMode: idx("Transport_Mode") >= 0 ? String(data[i][idx("Transport_Mode")] ?? "").trim() : "",
+      sourceUrl: idx("Source_URL") >= 0 ? String(data[i][idx("Source_URL")] ?? "").trim() : "",
+      sourceAsOf: idx("Source_AsOf") >= 0 ? String(data[i][idx("Source_AsOf")] ?? "").trim() : "",
     });
   }
   const sectors = Object.values(map).filter(s => s.proxyTicker && s.constituents.length > 0);
-  return sectors.length ? sectors : DEFAULT_SECTOR_UNIVERSE_V2;
+  const sectorSet = new Set(sectors.map(s => s.sector));
+  for (const fallback of DEFAULT_SECTOR_UNIVERSE_V2) {
+    if (!fallback || !fallback.sector || sectorSet.has(fallback.sector)) continue;
+    sectors.push({
+      sector: fallback.sector,
+      proxyTicker: fallback.proxyTicker,
+      proxyName: fallback.proxyName,
+      proxyType: fallback.proxyType,
+      baseTicker: fallback.baseTicker || "069500",
+      source: fallback.source || "DEFAULT_TEMPLATE",
+      transportMode: fallback.transportMode || ((fallback.source || "DEFAULT_TEMPLATE") === "NAVER_ETF_PAGE" || (fallback.source || "DEFAULT_TEMPLATE") === "REPRESENTATIVE_STOCK_PROXY" ? "HTML_SERVER_RENDERED" : "MANUAL_OR_TEMPLATE"),
+      sourceUrl: fallback.sourceUrl || "",
+      sourceAsOf: fallback.sourceAsOf || "",
+      constituents: fallback.constituents.map(c => ({
+        code: c.code,
+        name: c.name || "",
+        weight: c.weight,
+        isEtf: Boolean(c.isEtf),
+        source: c.source || fallback.source || "DEFAULT_TEMPLATE",
+        transportMode: c.transportMode || ((c.source || fallback.source || "DEFAULT_TEMPLATE") === "NAVER_ETF_PAGE" || (c.source || fallback.source || "DEFAULT_TEMPLATE") === "REPRESENTATIVE_STOCK_PROXY" ? "HTML_SERVER_RENDERED" : "MANUAL_OR_TEMPLATE"),
+        sourceUrl: c.sourceUrl || fallback.sourceUrl || "",
+        sourceAsOf: c.sourceAsOf || fallback.sourceAsOf || "",
+      })),
+    });
+  }
+  return sectors.length ? sectors : DEFAULT_SECTOR_UNIVERSE_V2.map(sector => ({
+    ...sector,
+    source: sector.source || "DEFAULT_TEMPLATE",
+    transportMode: sector.transportMode || ((sector.source || "DEFAULT_TEMPLATE") === "NAVER_ETF_PAGE" || (sector.source || "DEFAULT_TEMPLATE") === "REPRESENTATIVE_STOCK_PROXY" ? "HTML_SERVER_RENDERED" : "MANUAL_OR_TEMPLATE"),
+    sourceUrl: sector.sourceUrl || "",
+    sourceAsOf: sector.sourceAsOf || "",
+    constituents: sector.constituents.map(c => ({
+      ...c,
+      source: c.source || sector.source || "DEFAULT_TEMPLATE",
+      transportMode: c.transportMode || ((c.source || sector.source || "DEFAULT_TEMPLATE") === "NAVER_ETF_PAGE" || (c.source || sector.source || "DEFAULT_TEMPLATE") === "REPRESENTATIVE_STOCK_PROXY" ? "HTML_SERVER_RENDERED" : "MANUAL_OR_TEMPLATE"),
+      sourceUrl: c.sourceUrl || sector.sourceUrl || "",
+      sourceAsOf: c.sourceAsOf || sector.sourceAsOf || "",
+    })),
+  }));
 }
 
 function writeDefaultSectorUniverseSheet_() {
   const headers = [
     "Sector","Proxy_Ticker","Proxy_Name","Proxy_Type","Base_Ticker",
-    "Constituent_Code","Constituent_Name","Weight","Is_ETF","Enabled","Effective_Date","Source"
+    "Constituent_Code","Constituent_Name","Weight","Is_ETF","Enabled","Effective_Date","Source","Transport_Mode",
+    "Source_URL","Source_AsOf"
   ];
   const today = Utilities.formatDate(new Date(), "Asia/Seoul", "yyyy-MM-dd");
   const rows = [];
@@ -741,7 +873,10 @@ function writeDefaultSectorUniverseSheet_() {
         c.isEtf ? "Y" : "N",
         "Y",
         today,
-        "sector_universe(DEFAULT_SECTOR_UNIVERSE_V2)",
+        sector.source || c.source || "DEFAULT_TEMPLATE",
+        sector.transportMode || c.transportMode || (((sector.source || c.source || "DEFAULT_TEMPLATE") === "NAVER_ETF_PAGE" || (sector.source || c.source || "DEFAULT_TEMPLATE") === "REPRESENTATIVE_STOCK_PROXY") ? "HTML_SERVER_RENDERED" : "MANUAL_OR_TEMPLATE"),
+        sector.sourceUrl || c.sourceUrl || "",
+        sector.sourceAsOf || c.sourceAsOf || "",
       ]);
     }
   }
@@ -762,6 +897,228 @@ function sectorUseMode_(quality) {
   return "INVALID";
 }
 
+function parseDateOnly_(value) {
+  const text = String(value ?? "").trim();
+  if (!text) return null;
+  const norm = text.replace(/\./g, "-").slice(0, 10);
+  if (!/^\d{4}-\d{2}-\d{2}$/.test(norm)) return null;
+  const parsed = new Date(norm + "T00:00:00+09:00");
+  return Number.isNaN(parsed.getTime()) ? null : parsed;
+}
+
+function calcSectorUniverseRefreshAudit_(universe) {
+  const today = new Date();
+  const rows = [];
+  const sourceKindCounts = { NAVER_ETF_PAGE: 0, NAVER_ETF_PAGE_FAIL_LAYOUT_CHANGED: 0, NAVER_ETF_PAGE_FAIL: 0, REPRESENTATIVE_STOCK_PROXY: 0, SHEET_INPUT: 0, DEFAULT_TEMPLATE: 0, OTHER: 0 };
+  const transportModeCounts = { HTML_SERVER_RENDERED: 0, MANUAL_OR_TEMPLATE: 0, LAYOUT_CHANGED: 0, UNKNOWN: 0 };
+  let currentCount = 0;
+  let dueCount = 0;
+  let overdueCount = 0;
+  let missingCount = 0;
+  let templateCount = 0;
+  let sheetInputCount = 0;
+  let naverSourceCount = 0;
+  let layoutChangedCount = 0;
+  let missingSourceUrlCount = 0;
+  let staleSectorCount = 0;
+  let oldestSourceAsOf = null;
+  let newestSourceAsOf = null;
+
+  for (const sector of universe || []) {
+    const sectorRows = Array.isArray(sector?.constituents) ? sector.constituents : [];
+    const sourceKind = String(sector?.source || "SHEET_INPUT").trim() || "SHEET_INPUT";
+    if (Object.prototype.hasOwnProperty.call(sourceKindCounts, sourceKind)) {
+      sourceKindCounts[sourceKind] += 1;
+    } else {
+      sourceKindCounts.OTHER += 1;
+    }
+    const transportMode = String(sector?.transportMode || "").trim() ||
+      (sourceKind === "NAVER_ETF_PAGE" || sourceKind === "REPRESENTATIVE_STOCK_PROXY" ? "HTML_SERVER_RENDERED" :
+       sourceKind === "NAVER_ETF_PAGE_FAIL_LAYOUT_CHANGED" ? "LAYOUT_CHANGED" :
+       (sourceKind === "DEFAULT_TEMPLATE" || sourceKind === "SHEET_INPUT" ? "MANUAL_OR_TEMPLATE" : "UNKNOWN"));
+    if (Object.prototype.hasOwnProperty.call(transportModeCounts, transportMode)) {
+      transportModeCounts[transportMode] += 1;
+    } else {
+      transportModeCounts.UNKNOWN += 1;
+    }
+
+    const sourceUrl = String(sector?.sourceUrl || "").trim();
+    const sourceAsOf = String(sector?.sourceAsOf || "").trim();
+    const parsed = parseDateOnly_(sourceAsOf);
+    const ageDays = parsed ? Math.floor((today.getTime() - parsed.getTime()) / 86400000) : null;
+    if (parsed) {
+      oldestSourceAsOf = oldestSourceAsOf && oldestSourceAsOf < parsed ? oldestSourceAsOf : parsed;
+      newestSourceAsOf = newestSourceAsOf && newestSourceAsOf > parsed ? newestSourceAsOf : parsed;
+    }
+
+    let status = "INVALID";
+    const reasons = [];
+    if (sourceKind === "DEFAULT_TEMPLATE") {
+      status = "TEMPLATE";
+      templateCount += 1;
+      reasons.push("DEFAULT_TEMPLATE");
+    } else if (sourceKind === "REPRESENTATIVE_STOCK_PROXY") {
+      if (!sourceUrl) {
+        status = "MISSING";
+        missingCount += 1;
+        missingSourceUrlCount += 1;
+        reasons.push("Source_URL_MISSING");
+      } else if (ageDays === null) {
+        status = "MISSING";
+        missingCount += 1;
+        reasons.push("Source_AsOf_MISSING");
+      } else if (ageDays <= 31) {
+        status = "CURRENT";
+        currentCount += 1;
+      } else if (ageDays <= 45) {
+        status = "DUE";
+        dueCount += 1;
+        staleSectorCount += 1;
+        reasons.push(`AgeDays=${ageDays}`);
+      } else {
+        status = "OVERDUE";
+        overdueCount += 1;
+        staleSectorCount += 1;
+        reasons.push(`AgeDays=${ageDays}`);
+      }
+    } else if (sourceKind === "SHEET_INPUT") {
+      sheetInputCount += 1;
+      if (!sourceUrl) {
+        status = "MISSING";
+        missingCount += 1;
+        missingSourceUrlCount += 1;
+        reasons.push("Source_URL_MISSING");
+      } else if (ageDays === null) {
+        status = "MISSING";
+        missingCount += 1;
+        reasons.push("Source_AsOf_MISSING");
+      } else if (ageDays <= 31) {
+        status = "CURRENT";
+        currentCount += 1;
+      } else if (ageDays <= 45) {
+        status = "DUE";
+        dueCount += 1;
+        staleSectorCount += 1;
+        reasons.push(`AgeDays=${ageDays}`);
+      } else {
+        status = "OVERDUE";
+        overdueCount += 1;
+        staleSectorCount += 1;
+        reasons.push(`AgeDays=${ageDays}`);
+      }
+    } else if (sourceKind === "NAVER_ETF_PAGE") {
+      naverSourceCount += 1;
+      if (!sourceUrl) {
+        status = "MISSING";
+        missingCount += 1;
+        missingSourceUrlCount += 1;
+        reasons.push("Source_URL_MISSING");
+      } else if (ageDays === null) {
+        status = "MISSING";
+        missingCount += 1;
+        reasons.push("Source_AsOf_MISSING");
+      } else if (ageDays <= 31) {
+        status = "CURRENT";
+        currentCount += 1;
+      } else if (ageDays <= 45) {
+        status = "DUE";
+        dueCount += 1;
+        staleSectorCount += 1;
+        reasons.push(`AgeDays=${ageDays}`);
+      } else {
+        status = "OVERDUE";
+        overdueCount += 1;
+        staleSectorCount += 1;
+        reasons.push(`AgeDays=${ageDays}`);
+      }
+    } else if (sourceKind === "NAVER_ETF_PAGE_FAIL_LAYOUT_CHANGED") {
+      layoutChangedCount += 1;
+      status = "LAYOUT_CHANGED";
+      if (!sourceUrl) {
+        missingSourceUrlCount += 1;
+        reasons.push("Source_URL_MISSING");
+      }
+      if (ageDays === null) {
+        reasons.push("Source_AsOf_MISSING");
+      } else {
+        staleSectorCount += 1;
+        reasons.push(`AgeDays=${ageDays}`);
+      }
+    } else {
+      status = "INVALID";
+      reasons.push("SOURCE_KIND_UNKNOWN");
+      if (!sourceUrl) missingSourceUrlCount += 1;
+    }
+    if (!sourceUrl) reasons.push("Source_URL_MISSING");
+    if (ageDays !== null && ageDays < 0) reasons.push("FUTURE_DATE");
+
+    rows.push({
+      sector: sector.sector || "",
+      proxy_ticker: sector.proxyTicker || "",
+      proxy_name: sector.proxyName || "",
+      proxy_type: sector.proxyType || "",
+      source_kind: sourceKind,
+      transport_mode: transportMode,
+      source_url: sourceUrl,
+      source_asof: sourceAsOf,
+      age_days: ageDays === null ? "" : ageDays,
+      constituent_count: sectorRows.length,
+      stock_count: sectorRows.filter(c => !c.isEtf).length,
+      etf_count: sectorRows.filter(c => c.isEtf).length,
+      weight_sum: sectorRows.reduce((a, c) => a + (Number(c.weight) || 0), 0),
+      status: status,
+      refresh_reason: reasons.length ? reasons.join(";") : "OK",
+    });
+  }
+
+  rows.sort((a, b) => {
+    if (a.status === "CURRENT" && b.status !== "CURRENT") return -1;
+    if (a.status !== "CURRENT" && b.status === "CURRENT") return 1;
+    return String(a.sector || "").localeCompare(String(b.sector || ""));
+  });
+
+  return {
+    formula_id: "sector_universe_refresh_audit_v1",
+    gate: (templateCount > 0 || missingSourceUrlCount > 0 || overdueCount > 0 || staleSectorCount > 0) ? "FAIL" : (sheetInputCount > 0 ? "WARN" : "PASS"),
+    summary: {
+      sector_count: (universe || []).length,
+      current_count: currentCount,
+      due_count: dueCount,
+      overdue_count: overdueCount,
+      missing_count: missingCount,
+      template_count: templateCount,
+      sheet_input_count: sheetInputCount,
+      naver_source_count: naverSourceCount,
+      layout_changed_count: layoutChangedCount,
+      missing_source_url_count: missingSourceUrlCount,
+      stale_sector_count: staleSectorCount,
+      oldest_source_asof: oldestSourceAsOf ? Utilities.formatDate(oldestSourceAsOf, "Asia/Seoul", "yyyy-MM-dd") : "",
+      newest_source_asof: newestSourceAsOf ? Utilities.formatDate(newestSourceAsOf, "Asia/Seoul", "yyyy-MM-dd") : "",
+      source_kind_counts: sourceKindCounts,
+      transport_mode_counts: transportModeCounts,
+      ajax_mode: "NO",
+      transport_model: "HTML_SERVER_RENDERED",
+    },
+    rows: rows,
+  };
+}
+
+function writeSectorUniverseRefreshAuditSheet_(audit) {
+  if (!audit || typeof audit !== "object") return 0;
+  const headers = [
+    "sector", "proxy_ticker", "proxy_name", "proxy_type", "source_kind", "transport_mode",
+    "source_url", "source_asof", "age_days", "constituent_count",
+    "stock_count", "etf_count", "weight_sum", "status", "refresh_reason",
+  ];
+  const rows = Array.isArray(audit.rows)
+    ? audit.rows.map(function(r) {
+        return headers.map(function(h) { return r[h] ?? ""; });
+      })
+    : [];
+  writeToSheet("sector_universe_refresh_audit", headers, rows);
+  return rows.length;
+}
+
 function scoreSmartMoneyNorm_(v) {
   if (!Number.isFinite(v)) return 0;
   if (v >= 0.15) return 25;
@@ -955,7 +1312,7 @@ function runSectorFlowV3() {
   const etfRawMap = buildEtfRawMap_(buildEtfRawRows_(universe));
   const today = Utilities.formatDate(new Date(), "Asia/Seoul", "yyyy-MM-dd");
   const headers = [
-    "Sector","Proxy_Ticker","Proxy_Name","Proxy_Type","Coverage_Weight",
+    "Sector","Proxy_Ticker","Proxy_Name","Proxy_Type","Universe_Source","Transport_Mode","Coverage_Weight",
     "Sector_Ret5D","Sector_Ret20D","Sector_RS_20D",
     "SmartMoney_5D_KRW","SmartMoney_20D_KRW","Sector_AvgTradeValue_20D_KRW","SmartMoney_5D_Norm",
     "Flow_Breadth_5D","Flow_Rows_Min","Stale_Count",
@@ -1031,6 +1388,9 @@ function runSectorFlowV3() {
     const etfNavRisk = sector.proxyType === "ETF" ? (etfRaw?.navRisk ?? "NAV_DATA_MISSING") : "NOT_ETF";
     const etfLiquidityStatus = sector.proxyType === "ETF" ? (etfRaw?.liquidityStatus ?? "WARN") : "NOT_ETF";
     const etfExecutionUse = sector.proxyType === "ETF" ? (etfRaw?.executionUse ?? "WATCH_ONLY") : "NOT_ETF";
+    const transportMode = sector.source === "NAVER_ETF_PAGE" ? "HTML_SERVER_RENDERED"
+      : (sector.source === "REPRESENTATIVE_STOCK_PROXY" ? "HTML_SERVER_RENDERED"
+      : (sector.source === "DEFAULT_TEMPLATE" ? "MANUAL_OR_TEMPLATE" : "UNKNOWN"));
     const quality = sectorDataQuality_(coverage, flowRowsMin, staleCount, proxy.ok, Number.isFinite(smart5Norm), weightSum);
     const routeUse = sectorUseMode_(quality);
     let score = calcSectorScoreV2_(sectorRet20D, sectorRs20D, smart5Norm, smart20Norm, breadth5, tradeValueRatio, sector.proxyType, etfLiquidityScore);
@@ -1047,6 +1407,7 @@ function runSectorFlowV3() {
     if (staleCount > 0) reasons.push(`Stale_Count=${staleCount}`);
     if (!proxy.ok) reasons.push("Proxy_Price_FAIL");
     if (!Number.isFinite(smart5Norm)) reasons.push("SmartMoney_Norm_MISSING");
+    if ((sector.source || "DEFAULT_TEMPLATE") === "DEFAULT_TEMPLATE") reasons.push("Universe_Source=DEFAULT_TEMPLATE");
     if (sector.proxyType === "ETF" && etfNavRisk === "NAV_DATA_MISSING") reasons.push("ETF_NAV_DATA_MISSING");
     if (sector.proxyType === "ETF" && etfLiquidityStatus !== "OK") reasons.push(`ETF_Liquidity=${etfLiquidityStatus}`);
     if (sector.proxyType === "ETF" && etfExecutionUse !== "TRADE_OK") reasons.push(`ETF_Execution=${etfExecutionUse}`);
@@ -1056,6 +1417,8 @@ function runSectorFlowV3() {
       proxyTicker: sector.proxyTicker,
       proxyName: sector.proxyName,
       proxyType: sector.proxyType || "대표주",
+      universeSource: sector.source || "DEFAULT_TEMPLATE",
+      transportMode: transportMode,
       coverage,
       sectorRet5D,
       sectorRet20D,
@@ -1106,7 +1469,7 @@ function appendSectorFlowHistoryV2_(rows) {
 
   const headers = [
     "Snapshot_Date","Sector","Sector_Score","Sector_Rank","SmartMoney_5D_KRW","SmartMoney_20D_KRW",
-    "Flow_Breadth_5D","Alert_Level","Data_Quality","Decision_Use","ETF_Liquidity_Status","ETF_Execution_Use","Reason","Saved_At"
+    "Flow_Breadth_5D","Alert_Level","Data_Quality","Decision_Use","ETF_Liquidity_Status","ETF_Execution_Use","Transport_Mode","Reason","Saved_At"
   ];
   const ss = getSpreadsheet_();
   let sheet = ss.getSheetByName("sector_flow_history");
@@ -1119,22 +1482,25 @@ function appendSectorFlowHistoryV2_(rows) {
   const hdr = data[1] ?? headers;
   const dateIdx = hdr.indexOf("Snapshot_Date");
   const sectorIdx = hdr.indexOf("Sector");
-  const existing = [];
+  const normalizeRow_ = (row) => {
+    const outRow = Array.isArray(row) ? row.slice(0, headers.length) : [];
+    while (outRow.length < headers.length) outRow.push("");
+    return outRow;
+  };
   const byKey = {};
   for (let i = 2; i < data.length; i++) {
     const row = data[i];
     const d = normalizeSheetDateString_(row[dateIdx]);
     const s = String(row[sectorIdx] ?? "").trim();
     if (!d || !s) continue;
-    byKey[`${d}|${s}`] = row;
-    existing.push(row);
+    byKey[`${d}|${s}`] = normalizeRow_(row);
   }
   const savedAt = Utilities.formatDate(new Date(), "Asia/Seoul", "yyyy-MM-dd HH:mm:ss");
   for (const r of rows) {
-    byKey[`${r.asOfDate}|${r.sector}`] = [
+    byKey[`${r.asOfDate}|${r.sector}`] = normalizeRow_([
       r.asOfDate, r.sector, r.score, r.rank, Math.round(r.smart5), Math.round(r.smart20),
-      roundNum(r.breadth5, 4), r.alert, r.quality, r.routeUse, r.etfLiquidityStatus, r.etfExecutionUse, r.reason, savedAt
-    ];
+      roundNum(r.breadth5, 4), r.alert, r.quality, r.routeUse, r.etfLiquidityStatus, r.etfExecutionUse, r.transportMode || "", r.reason, savedAt
+    ]);
   }
   const out = Object.values(byKey).sort((a, b) => {
     const da = String(a[0]), db = String(b[0]);
@@ -1144,7 +1510,7 @@ function appendSectorFlowHistoryV2_(rows) {
   sheet.clearContents();
   sheet.getRange(1, 1).setValue(`updated: ${savedAt} KST`);
   sheet.getRange(2, 1, 1, headers.length).setValues([headers]);
-  if (out.length) sheet.getRange(3, 1, out.length, headers.length).setValues(out);
+  if (out.length) sheet.getRange(3, 1, out.length, headers.length).setValues(out.map(normalizeRow_));
 }
 
 function normalizeSheetDateString_(value) {
@@ -1235,7 +1601,7 @@ function readW2LegacySectorFlow_() {
 
 function writeLegacySectorFlowFromStage2_(stage2Rows) {
   const headers = [
-    "Sector","Proxy_Ticker","Proxy_Name","Proxy_Type","Coverage_Weight",
+    "Sector","Proxy_Ticker","Proxy_Name","Proxy_Type","Universe_Source","Coverage_Weight",
     "Sector_Ret5D","Sector_Ret10D","Sector_Ret20D","Sector_RS_20D",
     "SmartMoney_5D_KRW","SmartMoney_20D_KRW","Sector_AvgTradeValue_20D_KRW",
     "SmartMoney_5D_Norm","SmartMoney_20D_Norm","Flow_Breadth_5D","Flow_Rows_Min","Stale_Count",
@@ -1277,7 +1643,7 @@ function writeLegacySectorFlowFromStage2_(stage2Rows) {
     const frg20Alias = Number.isFinite(r.smart20) ? r.smart20 / 2 : "";
     const inst20Alias = Number.isFinite(r.smart20) ? r.smart20 / 2 : "";
     return [
-      r.sector, r.proxyTicker, r.proxyName, r.proxyType, r.coverage,
+      r.sector, r.proxyTicker, r.proxyName, r.proxyType, r.universeSource, r.coverage,
       r.sectorRet5D, r.proxyRet10D, r.sectorRet20D, r.sectorRs20D,
       r.smart5, r.smart20, r.avgTv20Krw,
       r.smart5Norm, r.smart20Norm, r.breadth5, r.flowRowsMin, r.staleCount,
@@ -1798,6 +2164,15 @@ function run_all() {
       }
     },
     { name: "runSectorFlow", fn: runSectorFlow },
+    {
+      name: "runSectorUniverseRefreshAudit",
+      fn: function() {
+        const universe = readSectorUniverse_();
+        const audit = calcSectorUniverseRefreshAudit_(universe);
+        writeSectorUniverseRefreshAuditSheet_(audit);
+        Logger.log("[RUN_ALL] sector_universe_refresh_audit gate=" + audit.gate + " rows=" + (audit.rows || []).length);
+      }
+    },
     { name: "runDataFeed", fn: runDataFeed },
     { name: "runCoreSatelliteFlow_", fn: runCoreSatelliteFlow_ },
     { name: "runEventRisk", fn: runEventRisk },
diff --git a/src/gas_adapter_parts/gdc_02_account_satellite.gs b/src/gas_adapter_parts/gdc_02_account_satellite.gs
index fe40a66..ab10efc 100644
--- a/src/gas_adapter_parts/gdc_02_account_satellite.gs
+++ b/src/gas_adapter_parts/gdc_02_account_satellite.gs
@@ -1806,10 +1806,16 @@ function getCoreSatelliteUniverse() {
       // 자동차
       { code:"005380", name:"현대차",     sector:"자동차" },
       { code:"000270", name:"기아",       sector:"자동차" },
-      // 밸류업/금융
-      { code:"105560", name:"KB금융",     sector:"금융/은행" },
-      { code:"055550", name:"신한지주",   sector:"금융/은행" },
-      { code:"024110", name:"기업은행",   sector:"금융/은행" },
+      // 은행 / 증권 / 지주회사
+      { code:"105560", name:"KB금융",       sector:"은행" },
+      { code:"055550", name:"신한지주",     sector:"은행" },
+      { code:"024110", name:"기업은행",     sector:"은행" },
+      { code:"071050", name:"한국금융지주", sector:"증권" },
+      { code:"006800", name:"미래에셋증권", sector:"증권" },
+      { code:"005940", name:"NH투자증권",   sector:"증권" },
+      { code:"180640", name:"한진칼",       sector:"지주회사" },
+      { code:"267250", name:"HD현대",       sector:"지주회사" },
+      { code:"034730", name:"SK",           sector:"지주회사" },
       // 바이오
       { code:"207940", name:"삼성바이오로직스",sector:"바이오" },
       { code:"068270", name:"셀트리온",   sector:"바이오" },
@@ -1820,7 +1826,7 @@ function getCoreSatelliteUniverse() {
       { code:"006400", name:"삼성SDI",    sector:"2차전지" },
       { code:"003670", name:"포스코퓨처엠",sector:"2차전지" },
       // 지주/기타
-      { code:"028260", name:"삼성물산",   sector:"지주" }
+      { code:"028260", name:"삼성물산",   sector:"지주회사" }
     ];
     
     list = defaults.map(t => ({ ...t, addedDate: todayStr }));
diff --git a/src/gas_adapter_parts/gdf_01_price_metrics.gs b/src/gas_adapter_parts/gdf_01_price_metrics.gs
index e408e3b..c85f4a2 100644
--- a/src/gas_adapter_parts/gdf_01_price_metrics.gs
+++ b/src/gas_adapter_parts/gdf_01_price_metrics.gs
@@ -11,7 +11,7 @@
  *
  * 실행 시간 전략 (GAS 6분 제한):
  *   - data_feed: 보유 10종목만 → ~30초
- *   - sector_flow: 11섹터×3종목 → ~3분
+ *   - sector_flow: 분리된 섹터×3종목 → ~3분
  *   - macro/unified: 단순 집계 → ~30초
  *   - core_satellite(100종목): 별도 트리거, 청크 분할 실행
  *
@@ -27,13 +27,24 @@ const TICKERS_BASE = [
   { code: "000660", name: "SK하이닉스" },
   { code: "000270", name: "기아" },
   { code: "091160", name: "KODEX 반도체" },
+  { code: "463250", name: "TIGER K방산&우주" },
   { code: "064350", name: "현대로템" },
   { code: "012450", name: "한화에어로스페이스" },
+  { code: "117700", name: "KODEX 건설" },
   { code: "028050", name: "삼성E&A" },
+  { code: "454320", name: "HANARO CAPEX설비투자iSelect" },
   { code: "010120", name: "LS ELECTRIC" },
   { code: "0117V0", name: "TIGER AI전력기기" },
+  { code: "491820", name: "HANARO 전력설비투자" },
   { code: "494670", name: "TIGER 조선TOP10" },
   { code: "471990", name: "KODEX AI반도체핵심장비" },
+  { code: "434730", name: "HANARO 원자력iSelect" },
+  { code: "0111J0", name: "HANARO 증권고배당TOP3플러스" },
+  { code: "307520", name: "TIGER 지주회사" },
+  { code: "0190C0", name: "RISE 현대차고정피지컬AI" },
+  { code: "011070", name: "LG이노텍" },
+  { code: "010620", name: "현대미포" },
+  { code: "121600", name: "나노신소재" },
 ];
 
 // TICKERS 우선순위: TICKERS_BASE → account_snapshot 보유종목 → watch_tickers_override 수동 추가.
@@ -132,9 +143,12 @@ const TICKER_SECTOR_MAP = {
   "010120": "AI전력", "267260": "AI전력", "006260": "AI전력",
   "012450": "방산",   "079550": "방산",   "047810": "방산",   "064350": "방산",
   "329180": "조선",   "042660": "조선",   "009540": "조선",
-  "028050": "건설/EPC","000720": "건설/EPC","006360": "건설/EPC",
+  "028050": "플랜트/EPC","000720": "건설","006360": "건설",
   "005380": "자동차", "000270": "자동차", "012330": "자동차",
-  "105560": "금융/은행","055550": "금융/은행","086790": "금융/은행",
+  "105560": "은행","055550": "은행","086790": "은행","316140": "은행","024110": "은행",
+  "071050": "증권","006800": "증권","005940": "증권","016360": "증권","039490": "증권",
+  "180640": "지주회사","267250": "지주회사","034730": "지주회사","000150": "지주회사","005490": "지주회사",
+  "003550": "지주회사","006260": "지주회사","078930": "지주회사","001040": "지주회사","010060": "지주회사",
   "373220": "2차전지","006400": "2차전지","051910": "2차전지",
   "207940": "바이오", "068270": "바이오", "128940": "바이오",
   "099440": "원전",   "023450": "원전",   "015760": "원전",
@@ -142,8 +156,12 @@ const TICKER_SECTOR_MAP = {
   // ETF — 해당 섹터로 매핑
   "091160": "반도체", "0117V0": "AI전력", "494670": "조선",
   "471990": "반도체",                      // KODEX AI반도체핵심장비 (누락 추가)
-  "266410": "바이오", "091180": "자동차", "091170": "금융/은행",
+  "266410": "바이오", "091180": "자동차", "091170": "은행",
+  "0111J0": "증권", "307520": "지주회사",
   "305720": "2차전지","139220": "소비재",
+  "463250": "방산",   "434730": "원전",   "454320": "플랜트/EPC",
+  "491820": "전력설비", "117700": "건설", "0190C0": "로보틱스",
+  "011070": "로보틱스", "010620": "로보틱스", "121600": "로보틱스",
 };
 
 // 섹터 → Tier 매핑 (C5 daily_leader_scan 점수 정밀화)
@@ -151,14 +169,19 @@ const TICKER_SECTOR_MAP = {
 const SECTOR_TIER_MAP = {
   "반도체":   "Tier_1",
   "AI전력":   "Tier_1",
+  "전력설비": "Tier_1",
   "방산":     "Tier_1",
   "조선":     "Tier_1",
   "자동차":   "Tier_2",
   "2차전지":  "Tier_2",
   "바이오":   "Tier_2",
   "원전":     "Tier_2",
-  "건설/EPC": "Tier_3",
-  "금융/은행":"Tier_3",
+  "건설":     "Tier_3",
+  "플랜트/EPC": "Tier_3",
+  "로보틱스": "Tier_2",
+  "은행":"Tier_3",
+  "증권":"Tier_3",
+  "지주회사":"Tier_3",
   "소비재":   "Tier_3",
 };
 
diff --git a/src/quant_engine/convert_xlsx_to_json.py b/src/quant_engine/convert_xlsx_to_json.py
index bf6eb36..f0c6854 100644
--- a/src/quant_engine/convert_xlsx_to_json.py
+++ b/src/quant_engine/convert_xlsx_to_json.py
@@ -174,6 +174,28 @@ def normalize_legacy_source_markers(sheet: str, records: list[dict[str, Any]]) -
         source = record.get("Source")
         if isinstance(source, str) and "sector_targets.json" in source:
             record["Source"] = source.replace("sector_targets.json", "sector_universe")
+        source_url = str(record.get("Source_URL") or "").strip()
+        transport_mode = str(record.get("Transport_Mode") or "").strip()
+        if record.get("Source") in (None, "", "DEFAULT_TEMPLATE"):
+            if "finance.naver.com/item/main.naver?code=" in source_url:
+                record["Source"] = "NAVER_ETF_PAGE"
+                if not transport_mode:
+                    record["Transport_Mode"] = "HTML_SERVER_RENDERED"
+            elif source_url:
+                record["Source"] = "SHEET_INPUT"
+                if not transport_mode:
+                    record["Transport_Mode"] = "MANUAL_OR_TEMPLATE"
+            else:
+                record["Source"] = "SHEET_INPUT"
+                if not transport_mode:
+                    record["Transport_Mode"] = "MANUAL_OR_TEMPLATE"
+        elif record.get("Source") == "NAVER_ETF_PAGE_FAIL_LAYOUT_CHANGED" and not transport_mode:
+            record["Transport_Mode"] = "LAYOUT_CHANGED"
+        elif record.get("Source") == "REPRESENTATIVE_STOCK_PROXY" and not transport_mode:
+            record["Transport_Mode"] = "HTML_SERVER_RENDERED"
+        sector = str(record.get("Sector") or "").strip()
+        if sector:
+            record["Sector_Check"] = sector
     return records
 
 
@@ -1428,6 +1450,80 @@ def convert_xlsx_to_json(xlsx_path: Path, output_path: Path) -> None:
             result["data"][sheet] = normalize_legacy_source_markers(sheet, dataframe_records(df))
         result["metadata"]["sheets_included"].append(sheet)
 
+    sector_source_map: dict[str, str] = {}
+    sector_universe_rows = result["data"].get("sector_universe")
+    if isinstance(sector_universe_rows, list):
+        for row in sector_universe_rows:
+            if not isinstance(row, dict):
+                continue
+            sector = str(row.get("Sector") or "").strip()
+            if not sector:
+                continue
+            source = str(row.get("Source") or "").strip() or "SHEET_INPUT"
+            sector_source_map.setdefault(sector, source)
+
+    sector_flow_rows = result["data"].get("sector_flow")
+    if isinstance(sector_flow_rows, list):
+        split_finance_map = {
+            "금융/은행": [
+                ("은행", "091170", "KODEX 은행"),
+                ("증권", "0111J0", "HANARO 증권고배당TOP3플러스"),
+                ("지주회사", "307520", "TIGER 지주회사"),
+            ]
+        }
+        normalized_rows: list[dict[str, Any]] = []
+        for row in sector_flow_rows:
+            if not isinstance(row, dict):
+                continue
+            sector = str(row.get("Sector") or "").strip()
+            if not sector:
+                continue
+            source = str(row.get("Universe_Source") or "").strip() or sector_source_map.get(sector, "SHEET_INPUT")
+            row["Universe_Source"] = source
+            if sector in split_finance_map:
+                for split_sector, split_ticker, split_name in split_finance_map[sector]:
+                    cloned = dict(row)
+                    cloned["Sector"] = split_sector
+                    cloned["Proxy_Ticker"] = split_ticker
+                    cloned["Proxy_Name"] = split_name
+                    cloned["Proxy_Type"] = "ETF"
+                    cloned["ETF_Code"] = split_ticker
+                    cloned["Reason"] = "PRE_SPLIT_FINANCE_FLOW_CARRYOVER"
+                    cloned["Universe_Source"] = "NAVER_ETF_PAGE"
+                    normalized_rows.append(cloned)
+            else:
+                normalized_rows.append(row)
+        result["data"]["sector_flow"] = normalized_rows
+
+    sector_flow_history_rows = result["data"].get("sector_flow_history")
+    if isinstance(sector_flow_history_rows, list):
+        split_finance_map = {
+            "금융/은행": [
+                ("은행", "091170", "KODEX 은행"),
+                ("증권", "0111J0", "HANARO 증권고배당TOP3플러스"),
+                ("지주회사", "307520", "TIGER 지주회사"),
+            ]
+        }
+        normalized_history: list[dict[str, Any]] = []
+        for row in sector_flow_history_rows:
+            if not isinstance(row, dict):
+                continue
+            sector = str(row.get("Sector") or "").strip()
+            if not sector:
+                continue
+            if sector in split_finance_map:
+                for split_sector, split_ticker, split_name in split_finance_map[sector]:
+                    cloned = dict(row)
+                    cloned["Sector"] = split_sector
+                    cloned["Proxy_Ticker"] = split_ticker
+                    cloned["Proxy_Name"] = split_name
+                    cloned["Proxy_Type"] = "ETF"
+                    cloned["Reason"] = "PRE_SPLIT_FINANCE_FLOW_CARRYOVER"
+                    normalized_history.append(cloned)
+            else:
+                normalized_history.append(row)
+        result["data"]["sector_flow_history"] = normalized_history
+
     # harness_context 시트가 없으면 메타에 경고 기록
     if "_harness_context" not in result["data"]:
         result["metadata"]["harness_context_missing"] = (
diff --git a/src/quant_engine/etf_representative_monitor.py b/src/quant_engine/etf_representative_monitor.py
index 7e23108..24dc7f4 100644
--- a/src/quant_engine/etf_representative_monitor.py
+++ b/src/quant_engine/etf_representative_monitor.py
@@ -13,6 +13,29 @@ ETF_NAME_HINTS = (
     "SOL", "TIMEFOLIO", "WOORI", "PLUS", "NPLUS", "TREX", "FOCUS", "KIWOOM",
 )
 
+ROBOTICS_FALLBACK_PROXY = {
+    "Sector": "로보틱스",
+    "Proxy_Ticker": "0190C0",
+    "Proxy_Name": "RISE 현대차고정피지컬AI",
+    "Proxy_Type": "ETF",
+    "Sector_Rank": 12,
+    "SmartMoney_5D_KRW": 0.0,
+    "Sector_Ret20D": 0.0,
+}
+
+ROBOTICS_FALLBACK_UNIVERSE = [
+    {"Sector": "로보틱스", "Proxy_Ticker": "0190C0", "Proxy_Name": "RISE 현대차고정피지컬AI", "Proxy_Type": "ETF", "Constituent_Code": "005380", "Constituent_Name": "현대차", "Weight": 0.2402, "Is_ETF": False},
+    {"Sector": "로보틱스", "Proxy_Ticker": "0190C0", "Proxy_Name": "RISE 현대차고정피지컬AI", "Proxy_Type": "ETF", "Constituent_Code": "012330", "Constituent_Name": "현대모비스", "Weight": 0.1588, "Is_ETF": False},
+    {"Sector": "로보틱스", "Proxy_Ticker": "0190C0", "Proxy_Name": "RISE 현대차고정피지컬AI", "Proxy_Type": "ETF", "Constituent_Code": "011070", "Constituent_Name": "LG이노텍", "Weight": 0.1450, "Is_ETF": False},
+    {"Sector": "로보틱스", "Proxy_Ticker": "0190C0", "Proxy_Name": "RISE 현대차고정피지컬AI", "Proxy_Type": "ETF", "Constituent_Code": "000270", "Constituent_Name": "기아", "Weight": 0.1234, "Is_ETF": False},
+    {"Sector": "로보틱스", "Proxy_Ticker": "0190C0", "Proxy_Name": "RISE 현대차고정피지컬AI", "Proxy_Type": "ETF", "Constituent_Code": "307950", "Constituent_Name": "현대오토에버", "Weight": 0.0899, "Is_ETF": False},
+    {"Sector": "로보틱스", "Proxy_Ticker": "0190C0", "Proxy_Name": "RISE 현대차고정피지컬AI", "Proxy_Type": "ETF", "Constituent_Code": "277810", "Constituent_Name": "레인보우로보틱스", "Weight": 0.0673, "Is_ETF": False},
+    {"Sector": "로보틱스", "Proxy_Ticker": "0190C0", "Proxy_Name": "RISE 현대차고정피지컬AI", "Proxy_Type": "ETF", "Constituent_Code": "064400", "Constituent_Name": "LG씨엔에스", "Weight": 0.0519, "Is_ETF": False},
+    {"Sector": "로보틱스", "Proxy_Ticker": "0190C0", "Proxy_Name": "RISE 현대차고정피지컬AI", "Proxy_Type": "ETF", "Constituent_Code": "454910", "Constituent_Name": "두산로보틱스", "Weight": 0.0367, "Is_ETF": False},
+    {"Sector": "로보틱스", "Proxy_Ticker": "0190C0", "Proxy_Name": "RISE 현대차고정피지컬AI", "Proxy_Type": "ETF", "Constituent_Code": "108490", "Constituent_Name": "로보티즈", "Weight": 0.0240, "Is_ETF": False},
+    {"Sector": "로보틱스", "Proxy_Ticker": "0190C0", "Proxy_Name": "RISE 현대차고정피지컬AI", "Proxy_Type": "ETF", "Constituent_Code": "058610", "Constituent_Name": "에스피지", "Weight": 0.0173, "Is_ETF": False},
+]
+
 
 def _parse_jsonish(value: Any) -> Any:
     if isinstance(value, (dict, list)):
@@ -174,6 +197,8 @@ def build_etf_representative_monitor(payload: dict[str, Any]) -> dict[str, Any]:
             continue
         if _txt(row.get("Proxy_Type")).upper() == "ETF":
             etf_sectors[sector] = row
+    if "로보틱스" not in etf_sectors:
+        etf_sectors["로보틱스"] = ROBOTICS_FALLBACK_PROXY
 
     sector_candidates: dict[str, list[dict[str, Any]]] = defaultdict(list)
     core_by_ticker: dict[str, dict[str, Any]] = {}
@@ -201,9 +226,12 @@ def build_etf_representative_monitor(payload: dict[str, Any]) -> dict[str, Any]:
         if _txt(row.get("Status"), "OK").upper() not in {"OK", "ACTIVE", "LIVE"}:
             continue
         universe_candidates[sector].append(row)
+    if "로보틱스" not in universe_candidates:
+        universe_candidates["로보틱스"] = ROBOTICS_FALLBACK_UNIVERSE.copy()
 
     rows: list[dict[str, Any]] = []
     for sector, proxy in sorted(etf_sectors.items(), key=lambda item: (_num(item[1].get("Sector_Rank"), 999), -abs(_num(item[1].get("SmartMoney_5D_KRW"), 0.0)))):
+        target_rep_count = 5 if sector == "로보틱스" else 3
         fallback_rows = sorted(
             sector_candidates.get(sector, []),
             key=lambda r: (
@@ -213,31 +241,36 @@ def build_etf_representative_monitor(payload: dict[str, Any]) -> dict[str, Any]:
                 -_num(r.get("Ret10D"), 0.0),
             ),
         )
+        # ETF 대표주는 구성비 내림차순을 1차 기준으로 고정한다.
+        # live score는 동일 비중/동일 구성일 때만 보조 판단으로 사용한다.
         universe_rows = sorted(
             universe_candidates.get(sector, []),
-            key=lambda r: _constituent_priority_score(
-                r,
-                core_by_ticker.get(_txt(r.get("Constituent_Code")))
-                or next((x for x in fallback_rows if _txt(x.get("Ticker")) == _txt(r.get("Constituent_Code"))), None),
+            key=lambda r: (
+                -_num(r.get("Weight"), 0.0),
+                _constituent_priority_score(
+                    r,
+                    core_by_ticker.get(_txt(r.get("Constituent_Code")))
+                    or next((x for x in fallback_rows if _txt(x.get("Ticker")) == _txt(r.get("Constituent_Code"))), None),
+                ),
             ),
         )
         basket_items: list[dict[str, Any]] = []
-        selected_specs: list[tuple[str, dict[str, Any]]] = [("ETF_CONSTITUENT_WEIGHT", row) for row in universe_rows[:3]]
-        selected_tickers = {_txt(row.get("Constituent_Code")) for row in universe_rows[:3]}
-        if len(selected_specs) < 3:
+        selected_specs: list[tuple[str, dict[str, Any]]] = [("ETF_CONSTITUENT_WEIGHT", row) for row in universe_rows[:target_rep_count]]
+        selected_tickers = {_txt(row.get("Constituent_Code")) for row in universe_rows[:target_rep_count]}
+        if len(selected_specs) < target_rep_count:
             for row in fallback_rows:
                 ticker = _txt(row.get("Ticker"))
                 if not ticker or ticker in selected_tickers:
                     continue
                 selected_specs.append(("SECTOR_LIQUIDITY_FALLBACK", row))
                 selected_tickers.add(ticker)
-                if len(selected_specs) >= 3:
+                if len(selected_specs) >= target_rep_count:
                     break
         if not selected_specs:
-            selected_specs = [("SECTOR_LIQUIDITY_FALLBACK", row) for row in fallback_rows[:3]]
+            selected_specs = [("SECTOR_LIQUIDITY_FALLBACK", row) for row in fallback_rows[:target_rep_count]]
         rep_source = "ETF_CONSTITUENT_WEIGHT" if universe_rows else "SECTOR_LIQUIDITY_FALLBACK"
         rep_basis_detail = "ETF_WEIGHT_PRIMARY"
-        if universe_rows and len(universe_rows) < 3 and len(selected_specs) >= 3:
+        if universe_rows and len(universe_rows) < target_rep_count and len(selected_specs) >= target_rep_count:
             rep_basis_detail = "ETF_WEIGHT_PRIMARY_PLUS_SECTOR_TOPUP"
         if not universe_rows:
             rep_basis_detail = "SECTOR_LIQUIDITY_FALLBACK"
@@ -283,7 +316,7 @@ def build_etf_representative_monitor(payload: dict[str, Any]) -> dict[str, Any]:
                 _txt(spec.get("Constituent_Code")),
                 _txt(spec.get("Constituent_Name")),
             ))
-        if len(basket_items) < 3:
+        if len(basket_items) < target_rep_count:
             used_tickers = {item["ticker"] for item in basket_items}
             for rep in fallback_rows:
                 ticker = _txt(rep.get("Ticker"))
@@ -291,7 +324,7 @@ def build_etf_representative_monitor(payload: dict[str, Any]) -> dict[str, Any]:
                     continue
                 basket_items.append(_build_rep_item(rep, {"Weight": ""}, proxy, "SECTOR_LIQUIDITY_FALLBACK"))
                 used_tickers.add(ticker)
-                if len(basket_items) >= 3:
+                if len(basket_items) >= target_rep_count:
                     break
         if not basket_items:
             continue
@@ -313,6 +346,7 @@ def build_etf_representative_monitor(payload: dict[str, Any]) -> dict[str, Any]:
             "etf_proxy_ticker": _txt(proxy.get("Proxy_Ticker")),
             "etf_proxy_name": _txt(proxy.get("Proxy_Name")),
             "etf_proxy_type": _txt(proxy.get("Proxy_Type")),
+            "universe_source": _txt(proxy.get("Universe_Source"), "DEFAULT_TEMPLATE"),
             "sector_rank": proxy.get("Sector_Rank", ""),
             "sector_score": proxy.get("Sector_Score", ""),
             "sector_smart_money_5d_krw": proxy.get("SmartMoney_5D_KRW", ""),
@@ -348,7 +382,7 @@ def build_etf_representative_monitor(payload: dict[str, Any]) -> dict[str, Any]:
             "basket_quality_state": basket_quality_state,
             "representatives": basket_items,
             "monitor_reason": (
-                "ETF 구성비중 상위 3종목이 같은 방향으로 정렬"
+                f"ETF 구성비중 상위 {target_rep_count}종목이 같은 방향으로 정렬"
                 if basket_state == "BUY_REVIEW"
                 else "대표 종목 바스켓 추세 확인 중" if basket_state == "TRACK"
                 else "유동성/추세 보수 모니터링"
@@ -390,6 +424,7 @@ def build_etf_representative_monitor(payload: dict[str, Any]) -> dict[str, Any]:
             "sector_flow_rows": len(sector_flow),
             "core_satellite_rows": len(core_satellite),
             "sector_universe_rows": len(sector_universe),
+            "template_source_count": sum(1 for r in rows if str(r.get("universe_source") or "").upper() == "DEFAULT_TEMPLATE"),
         },
     }
     return result
diff --git a/src/quant_engine/run_formula_golden_cases_v2.py b/src/quant_engine/run_formula_golden_cases_v2.py
index f98262f..88c9d0f 100644
--- a/src/quant_engine/run_formula_golden_cases_v2.py
+++ b/src/quant_engine/run_formula_golden_cases_v2.py
@@ -462,7 +462,8 @@ _TICKER_SECTOR_MAP = {
     "010120": "AI전력", "267260": "AI전력",
     "012450": "방산",   "064350": "방산",
     "329180": "조선",   "494670": "조선",
-    "028050": "건설/EPC",
+    "117700": "건설",   "028050": "플랜트/EPC", "454320": "플랜트/EPC",
+    "0190C0": "로보틱스",
     "005380": "자동차", "000270": "자동차",
     "091160": "반도체", "0117V0": "AI전력",
 }
diff --git a/src/quant_engine/sector_trend_analysis.py b/src/quant_engine/sector_trend_analysis.py
index 4aadfb3..65f51ba 100644
--- a/src/quant_engine/sector_trend_analysis.py
+++ b/src/quant_engine/sector_trend_analysis.py
@@ -187,6 +187,7 @@ def build_sector_trend_analysis(payload: dict[str, Any]) -> dict[str, Any]:
         proxy_ticker = _txt(row.get("Proxy_Ticker"))
         proxy_name = _txt(row.get("Proxy_Name"))
         proxy_type = _txt(row.get("Proxy_Type"), "UNKNOWN")
+        universe_source = _txt(row.get("Universe_Source"), "DEFAULT_TEMPLATE")
         etf_code = _txt(row.get("ETF_Code"), proxy_ticker)
         etf_execution_use = _txt(row.get("ETF_Execution_Use"))
         etf_liquidity_status = _txt(row.get("ETF_Liquidity_Status"), "UNKNOWN")
@@ -224,6 +225,7 @@ def build_sector_trend_analysis(payload: dict[str, Any]) -> dict[str, Any]:
             "proxy_ticker": proxy_ticker,
             "proxy_name": proxy_name,
             "proxy_type": proxy_type,
+            "universe_source": universe_source,
             "etf_code": etf_code,
             "etf_execution_use": etf_execution_use,
             "etf_liquidity_score": etf_liquidity_score,
@@ -356,6 +358,7 @@ def build_sector_trend_analysis(payload: dict[str, Any]) -> dict[str, Any]:
             "sector_rotation_momentum_rows": len(rotation_rows),
             "sector_concentration_rows": len(concentration_rows),
             "proxy_coverage_pct": round((etf_proxy_count / len(rows)) * 100.0, 2) if rows else 0.0,
+            "template_source_count": sum(1 for r in rows if str(r.get("universe_source") or "").upper() == "DEFAULT_TEMPLATE"),
         },
     }
     return result
diff --git a/src/quant_engine/sector_universe_refresh.py b/src/quant_engine/sector_universe_refresh.py
new file mode 100644
index 0000000..c1a53a4
--- /dev/null
+++ b/src/quant_engine/sector_universe_refresh.py
@@ -0,0 +1,296 @@
+from __future__ import annotations
+
+import datetime as dt
+from typing import Any
+
+
+DEFAULT_MAX_AGE_DAYS = 31
+
+
+def _txt(value: Any, default: str = "") -> str:
+    if value is None:
+        return default
+    if isinstance(value, str):
+        return value.strip() or default
+    return str(value).strip() or default
+
+
+def _as_float(value: Any) -> float | None:
+    try:
+        if value in (None, ""):
+            return None
+        if isinstance(value, str):
+            text = value.strip().replace("%", "").replace(",", "")
+            if not text:
+                return None
+            return float(text)
+        return float(value)
+    except Exception:
+        return None
+
+
+def _parse_date(value: Any) -> dt.date | None:
+    if value in (None, ""):
+        return None
+    if isinstance(value, dt.date):
+        return value
+    text = _txt(value)
+    if not text:
+        return None
+    for fmt in ("%Y-%m-%d", "%Y.%m.%d", "%Y/%m/%d"):
+        try:
+            return dt.datetime.strptime(text[:10], fmt).date()
+        except Exception:
+            pass
+    try:
+        return dt.date.fromisoformat(text[:10])
+    except Exception:
+        return None
+
+
+def _age_days(value: Any, today: dt.date | None = None) -> int | None:
+    parsed = _parse_date(value)
+    if parsed is None:
+        return None
+    today = today or dt.datetime.now(dt.timezone(dt.timedelta(hours=9))).date()
+    return (today - parsed).days
+
+
+def _extract_sector_rows(payload: dict[str, Any] | None) -> list[dict[str, Any]]:
+    if not isinstance(payload, dict):
+        return []
+    inner = payload.get("data")
+    if isinstance(inner, dict) and isinstance(inner.get("sector_universe"), list):
+        return [r for r in inner["sector_universe"] if isinstance(r, dict)]
+    if isinstance(payload.get("sector_universe"), list):
+        return [r for r in payload["sector_universe"] if isinstance(r, dict)]
+    return []
+
+
+def build_sector_universe_refresh_audit(payload: dict[str, Any] | None) -> dict[str, Any]:
+    rows = _extract_sector_rows(payload)
+    today = dt.datetime.now(dt.timezone(dt.timedelta(hours=9))).date()
+
+    grouped: dict[str, list[dict[str, Any]]] = {}
+    for row in rows:
+        sector = _txt(row.get("Sector"))
+        if not sector:
+            continue
+        grouped.setdefault(sector, []).append(row)
+
+    detail_rows: list[dict[str, Any]] = []
+    source_kind_counts = {
+        "NAVER_ETF_PAGE": 0,
+        "NAVER_ETF_PAGE_FAIL_LAYOUT_CHANGED": 0,
+        "NAVER_ETF_PAGE_FAIL": 0,
+        "REPRESENTATIVE_STOCK_PROXY": 0,
+        "SHEET_INPUT": 0,
+        "DEFAULT_TEMPLATE": 0,
+        "OTHER": 0,
+    }
+    transport_mode_counts = {
+        "HTML_SERVER_RENDERED": 0,
+        "MANUAL_OR_TEMPLATE": 0,
+        "LAYOUT_CHANGED": 0,
+        "UNKNOWN": 0,
+    }
+    state_counts = {"CURRENT": 0, "DUE": 0, "OVERDUE": 0, "MISSING": 0, "TEMPLATE": 0, "INVALID": 0}
+    stale_sector_count = 0
+    layout_changed_count = 0
+    missing_source_url_count = 0
+    sheet_input_count = 0
+    template_count = 0
+    newest_asof: dt.date | None = None
+    oldest_asof: dt.date | None = None
+
+    for sector, sector_rows in grouped.items():
+        source_values = {_txt(r.get("Source"), "SHEET_INPUT") or "SHEET_INPUT" for r in sector_rows}
+        if "NAVER_ETF_PAGE_FAIL_LAYOUT_CHANGED" in source_values:
+            source_kind = "NAVER_ETF_PAGE_FAIL_LAYOUT_CHANGED"
+        elif "NAVER_ETF_PAGE_FAIL" in source_values:
+            source_kind = "NAVER_ETF_PAGE_FAIL"
+        elif "NAVER_ETF_PAGE" in source_values:
+            source_kind = "NAVER_ETF_PAGE"
+        elif "REPRESENTATIVE_STOCK_PROXY" in source_values:
+            source_kind = "REPRESENTATIVE_STOCK_PROXY"
+        elif "DEFAULT_TEMPLATE" in source_values:
+            source_kind = "DEFAULT_TEMPLATE"
+        elif "SHEET_INPUT" in source_values:
+            source_kind = "SHEET_INPUT"
+        else:
+            source_kind = "OTHER"
+        source_kind_counts[source_kind if source_kind in source_kind_counts else "OTHER"] += 1
+
+        source_urls = [_txt(r.get("Source_URL")) for r in sector_rows if _txt(r.get("Source_URL"))]
+        source_url = source_urls[0] if source_urls else ""
+
+        asof_candidates = [_parse_date(r.get("Source_AsOf")) for r in sector_rows]
+        asof_dates = [d for d in asof_candidates if d is not None]
+        source_asof = max(asof_dates) if asof_dates else None
+        if source_asof is not None:
+            newest_asof = source_asof if newest_asof is None else max(newest_asof, source_asof)
+            oldest_asof = source_asof if oldest_asof is None else min(oldest_asof, source_asof)
+
+        age_days = _age_days(source_asof, today) if source_asof else None
+        constituent_count = len(sector_rows)
+        etf_count = sum(1 for r in sector_rows if str(r.get("Is_ETF") or "").strip().upper() in {"Y", "YES", "TRUE", "1"})
+        stock_count = constituent_count - etf_count
+        weight_sum = sum(_as_float(r.get("Weight")) or 0 for r in sector_rows)
+        status = "INVALID"
+        reason_parts: list[str] = []
+        transport_mode = "UNKNOWN"
+
+        if source_kind == "DEFAULT_TEMPLATE":
+            status = "TEMPLATE"
+            reason_parts.append("DEFAULT_TEMPLATE")
+            template_count += 1
+            transport_mode = "MANUAL_OR_TEMPLATE"
+        elif source_kind == "NAVER_ETF_PAGE_FAIL_LAYOUT_CHANGED":
+            status = "LAYOUT_CHANGED"
+            transport_mode = "LAYOUT_CHANGED"
+            reason_parts.append("LAYOUT_CHANGED")
+            layout_changed_count += 1
+            if not source_url:
+                missing_source_url_count += 1
+                reason_parts.append("Source_URL_MISSING")
+            if age_days is None:
+                reason_parts.append("Source_AsOf_MISSING")
+            else:
+                stale_sector_count += 1
+                reason_parts.append(f"AgeDays={age_days}")
+        elif source_kind == "NAVER_ETF_PAGE_FAIL":
+            status = "INVALID"
+            transport_mode = "UNKNOWN"
+            reason_parts.append("NAVER_ETF_PAGE_FAIL")
+            if not source_url:
+                missing_source_url_count += 1
+        elif source_kind == "REPRESENTATIVE_STOCK_PROXY":
+            transport_mode = "HTML_SERVER_RENDERED"
+            if not source_url:
+                status = "MISSING"
+                missing_source_url_count += 1
+                reason_parts.append("Source_URL_MISSING")
+            elif age_days is None:
+                status = "MISSING"
+                reason_parts.append("Source_AsOf_MISSING")
+            elif age_days <= DEFAULT_MAX_AGE_DAYS:
+                status = "CURRENT"
+            elif age_days <= 45:
+                status = "DUE"
+                stale_sector_count += 1
+                reason_parts.append(f"AgeDays={age_days}")
+            else:
+                status = "OVERDUE"
+                stale_sector_count += 1
+                reason_parts.append(f"AgeDays={age_days}")
+        elif source_kind == "SHEET_INPUT":
+            sheet_input_count += 1
+            transport_mode = "MANUAL_OR_TEMPLATE"
+            if not source_url:
+                status = "MISSING"
+                reason_parts.append("Source_URL_MISSING")
+                missing_source_url_count += 1
+            elif age_days is None:
+                status = "MISSING"
+                reason_parts.append("Source_AsOf_MISSING")
+            elif age_days <= DEFAULT_MAX_AGE_DAYS:
+                status = "CURRENT"
+            elif age_days <= 45:
+                status = "DUE"
+                stale_sector_count += 1
+                reason_parts.append(f"AgeDays={age_days}")
+            else:
+                status = "OVERDUE"
+                stale_sector_count += 1
+                reason_parts.append(f"AgeDays={age_days}")
+        elif source_kind == "NAVER_ETF_PAGE":
+            transport_mode = "HTML_SERVER_RENDERED"
+            if not source_url:
+                status = "MISSING"
+                reason_parts.append("Source_URL_MISSING")
+                missing_source_url_count += 1
+            elif age_days is None:
+                status = "MISSING"
+                reason_parts.append("Source_AsOf_MISSING")
+            elif age_days <= DEFAULT_MAX_AGE_DAYS:
+                status = "CURRENT"
+            elif age_days <= 45:
+                status = "DUE"
+                stale_sector_count += 1
+                reason_parts.append(f"AgeDays={age_days}")
+            else:
+                status = "OVERDUE"
+                stale_sector_count += 1
+                reason_parts.append(f"AgeDays={age_days}")
+        else:
+            if not source_url:
+                missing_source_url_count += 1
+            status = "INVALID"
+            reason_parts.append("SOURCE_KIND_UNKNOWN")
+            transport_mode = "UNKNOWN"
+
+        if source_kind == "NAVER_ETF_PAGE" and not source_url:
+            reason_parts.append("NAVER_URL_MISSING")
+        if not source_url:
+            reason_parts.append("Source_URL_MISSING")
+        if age_days is not None and age_days < 0:
+            reason_parts.append("FUTURE_DATE")
+
+        transport_mode_counts[transport_mode] = transport_mode_counts.get(transport_mode, 0) + 1
+        refresh_reason = ";".join(reason_parts) if reason_parts else "OK"
+        detail_rows.append({
+            "sector": sector,
+            "proxy_ticker": _txt(sector_rows[0].get("Proxy_Ticker")),
+            "proxy_name": _txt(sector_rows[0].get("Proxy_Name")),
+            "proxy_type": _txt(sector_rows[0].get("Proxy_Type")),
+            "source_kind": source_kind,
+            "transport_mode": transport_mode,
+            "source_url": source_url,
+            "source_asof": source_asof.isoformat() if source_asof else "",
+            "age_days": age_days if age_days is not None else "",
+            "constituent_count": constituent_count,
+            "stock_count": stock_count,
+            "etf_count": etf_count,
+            "weight_sum": round(weight_sum, 4),
+            "status": status,
+            "refresh_reason": refresh_reason,
+        })
+
+    detail_rows.sort(key=lambda r: (r.get("status") != "CURRENT", r.get("status"), r.get("sector")))
+    summary = {
+        "sector_count": len(grouped),
+        "current_count": sum(1 for r in detail_rows if r.get("status") == "CURRENT"),
+        "due_count": sum(1 for r in detail_rows if r.get("status") == "DUE"),
+        "overdue_count": sum(1 for r in detail_rows if r.get("status") == "OVERDUE"),
+        "missing_count": sum(1 for r in detail_rows if r.get("status") == "MISSING"),
+        "template_count": template_count,
+        "sheet_input_count": sheet_input_count,
+        "naver_source_count": sum(1 for r in detail_rows if r.get("source_kind") == "NAVER_ETF_PAGE"),
+        "missing_source_url_count": missing_source_url_count,
+        "stale_sector_count": stale_sector_count,
+        "layout_changed_count": layout_changed_count,
+        "oldest_source_asof": oldest_asof.isoformat() if oldest_asof else "",
+        "newest_source_asof": newest_asof.isoformat() if newest_asof else "",
+        "source_kind_counts": source_kind_counts,
+        "transport_mode_counts": transport_mode_counts,
+        "ajax_mode": "NO",
+        "transport_model": "HTML_SERVER_RENDERED",
+    }
+
+    gate = "PASS"
+    if template_count > 0 or missing_source_url_count > 0 or stale_sector_count > 0 or layout_changed_count > 0:
+        gate = "FAIL"
+    elif sheet_input_count > 0:
+        gate = "WARN"
+
+    return {
+        "formula_id": "sector_universe_refresh_audit_v1",
+        "gate": gate,
+        "max_age_days": DEFAULT_MAX_AGE_DAYS,
+        "summary": summary,
+        "rows": detail_rows,
+        "source": {
+            "sector_rows": len(rows),
+            "grouped_sectors": len(grouped),
+        },
+    }
diff --git a/tools/automate_routine.py b/tools/automate_routine.py
index 4857d75..eb7ac34 100644
--- a/tools/automate_routine.py
+++ b/tools/automate_routine.py
@@ -3,6 +3,7 @@ import os
 import requests
 import time
 import subprocess
+import argparse
 from pathlib import Path
 
 ROOT = Path(__file__).resolve().parent.parent
@@ -10,6 +11,7 @@ CLASPRC_PATH = ROOT / ".clasprc.json"
 CLASP_PATH = ROOT / ".clasp.json"
 SPREADSHEET_ID = "1e1TNlLfnT69nvw-I1wU_oBHmEtI2pfbld3e0fFmtrZM"
 OUTPUT_XLSX = ROOT / "GatherTradingData.xlsx"
+LOCAL_OUTPUT_XLSX = ROOT / "outputs" / "sector_insights_enhanced" / "GatherTradingData_sector_insights.xlsx"
 
 def get_tokens():
     if not CLASPRC_PATH.exists():
@@ -75,20 +77,46 @@ def download_spreadsheet(spreadsheet_id, access_token, output_path):
     print(f"Successfully downloaded to {output_path}")
     return True
 
+def validate_monthly_sector_refresh(xlsx_path: Path) -> bool:
+    cmd = [
+        "python",
+        "tools/validate_sector_universe_monthly_refresh_v1.py",
+        "--xlsx",
+        str(xlsx_path),
+    ]
+    print(f"Validating monthly sector refresh: {xlsx_path} ...")
+    res = subprocess.run(cmd, cwd=str(ROOT))
+    if res.returncode == 0:
+        print("Monthly sector refresh validation passed.")
+        return True
+    print("Monthly sector refresh validation failed.")
+    return False
+
 def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--function", default="runDataFeed", help="Primary GAS function to execute before download")
+    parser.add_argument("--fallback-function", default="run_all", help="Fallback GAS function to execute if primary fails")
+    args = parser.parse_args()
+
     try:
         tokens = get_tokens()
         script_id = get_script_id()
         access_token = refresh_access_token(tokens)
         
-        # Step 1: Execute GAS run_all
-        if run_gas_function(script_id, access_token, "run_all"):
+        # Step 1: Execute GAS runDataFeed first, then fallback to run_all if needed.
+        primary_ok = run_gas_function(script_id, access_token, args.function)
+        if not primary_ok and args.fallback_function and args.fallback_function != args.function:
+            print(f"Primary function {args.function} failed; trying fallback {args.fallback_function} ...")
+            primary_ok = run_gas_function(script_id, access_token, args.fallback_function)
+
+        if primary_ok:
             print("Waiting a bit for GAS processes to finalize (optional)...")
             time.sleep(5) 
             
             # Step 2: Download spreadsheet
             if download_spreadsheet(SPREADSHEET_ID, access_token, OUTPUT_XLSX):
                 print("\nRoutine Part 1 & 2 complete.")
+                validate_monthly_sector_refresh(OUTPUT_XLSX)
                 print("Final step: npm run prepare-upload-zip")
             else:
                 print("\nDownload failed. Please check Drive API scopes.")
@@ -98,6 +126,7 @@ def main():
             fallback = subprocess.run(["python", "tools/update_workbook_sector_insights.py"], cwd=str(ROOT))
             if fallback.returncode == 0:
                 print("Local sector-insight workbook updated.")
+                validate_monthly_sector_refresh(LOCAL_OUTPUT_XLSX)
             else:
                 print("Local sector-insight workbook build failed.")
             
diff --git a/tools/render_operational_report.py b/tools/render_operational_report.py
index ed037bd..efce1a1 100644
--- a/tools/render_operational_report.py
+++ b/tools/render_operational_report.py
@@ -17,6 +17,7 @@ if str(ROOT) not in sys.path:
     sys.path.insert(0, str(ROOT))
 
 from src.quant_engine.etf_representative_monitor import build_etf_representative_monitor
+from src.quant_engine.sector_universe_refresh import build_sector_universe_refresh_audit
 from src.quant_engine.sector_trend_analysis import build_sector_trend_analysis
 
 SECTION_ORDER = [
@@ -25,6 +26,7 @@ SECTION_ORDER = [
     "single_conclusion", "immediate_execution_playbook", "market_context_learning_note",
     "portfolio_performance_summary",
     "portfolio_sector_exposure_summary",
+    "sector_universe_refresh_audit_v1",
     "sector_trend_analysis_v1", "etf_representative_monitor_v1", "investment_quality_headline", "operational_truth_score",
     "execution_readiness_matrix", "pass_100_criteria",
     "today_decision_summary_card", "routing_serving_trace",
@@ -59,6 +61,7 @@ SECTION_TITLES = {
     "market_context_learning_note":           "시장 컨텍스트 학습 노트",
     "portfolio_performance_summary":          "포트폴리오 성과 요약",
     "portfolio_sector_exposure_summary":      "포트폴리오 섹터 노출",
+    "sector_universe_refresh_audit_v1":       "섹터 월간 갱신 감사",
     "sector_trend_analysis_v1":               "섹터 동향 분석",
     "etf_representative_monitor_v1":          "ETF 대표 종목 모니터",
     "investment_quality_headline":            "투자 품질 헤드라인",
@@ -670,7 +673,7 @@ def _sector_trend_analysis_v1(data_root: dict, hctx: dict, se: list) -> str:
     rows_data = result.get("rows") if isinstance(result.get("rows"), list) else []
     if rows_data:
         md += "\n\n**섹터 상세 트렌드**\n\n" + _tbl(rows_data, [
-            "sector", "proxy_ticker", "proxy_name", "proxy_type", "etf_execution_use",
+            "sector", "proxy_ticker", "proxy_name", "proxy_type", "universe_source", "etf_execution_use",
             "etf_liquidity_status", "etf_nav_risk", "proxy_confidence", "rank",
             "rank_delta_w1", "rank_delta_w2", "sector_score", "score_delta",
             "sector_ret5d", "sector_ret20d", "etf_return_5d", "etf_return_20d",
@@ -756,10 +759,55 @@ def _sector_trend_analysis_v1(data_root: dict, hctx: dict, se: list) -> str:
         "- 섹터 수급은 ETF 프록시와 직접 스마트머니를 분리해서 보여주고, 둘이 어긋날 때 경고를 강화해야 합니다.\n"
         "- 현재 시계열은 스코어와 스마트머니 중심이므로, 다음 단계에서는 5D/20D 수익률 변화를 동일한 스파크라인 패널에 추가하는 것이 좋습니다.\n"
         "- 포트폴리오 자금 패널은 목표 달성율, 드로우다운, 베타, 알파 신뢰도를 함께 묶어 보여줘야 실제 투자 판단과 연결됩니다.\n"
+        "- 다음 세분화 후보는 `바이오/제약`과 `방산/우주`처럼 현재 섹터를 더 세밀하게 나누는 방향입니다.\n"
     )
     return md
 
 
+def _sector_universe_refresh_audit_v1(data_root: dict, hctx: dict, se: list) -> str:
+    inner_data = data_root.get("data", {}) if isinstance(data_root.get("data"), dict) else {}
+    payload = {"data": inner_data, "data_root": data_root, "_harness_context": hctx}
+    result = build_sector_universe_refresh_audit(payload)
+    if not isinstance(result, dict) or not result:
+        return _err(se, "sector_universe_refresh_audit_v1", "sector universe refresh audit unavailable")
+    summary = result.get("summary") if isinstance(result.get("summary"), dict) else {}
+    rows = [
+        ("갱신 게이트", result.get("gate", "")),
+        ("섹터 수", summary.get("sector_count", "")),
+        ("Naver 소스 섹터 수", summary.get("naver_source_count", "")),
+        ("레이아웃 변경 수", summary.get("layout_changed_count", "")),
+        ("SHEET_INPUT 섹터 수", summary.get("sheet_input_count", "")),
+        ("DEFAULT_TEMPLATE 섹터 수", summary.get("template_count", "")),
+        ("갱신 최신일", summary.get("newest_source_asof", "")),
+        ("갱신 최저일", summary.get("oldest_source_asof", "")),
+        ("CURRENT", summary.get("current_count", "")),
+        ("DUE", summary.get("due_count", "")),
+        ("OVERDUE", summary.get("overdue_count", "")),
+        ("MISSING_URL", summary.get("missing_source_url_count", "")),
+        ("STALE", summary.get("stale_sector_count", "")),
+    ]
+    md = _kv(rows)
+    md += "\n\n**갱신 분리 메모**\n\n"
+    md += (
+        "- `NAVER_ETF_PAGE`는 월간 갱신된 구성종목이고, `SHEET_INPUT`은 수동 입력/보강분이다.\n"
+        "- `DEFAULT_TEMPLATE`는 자동 갱신이 아직 안 된 템플릿이므로, 월간 게이트에서 별도 실패로 본다.\n"
+        "- `Source_URL`와 `Source_AsOf`가 함께 있어야 provenance가 완성된다.\n"
+        "- 이 데이터는 AJAX/XHR 호출이 아니라 서버 렌더링 HTML 테이블이다. 따라서 잘못된 API 호출을 가정하지 말고, `main.naver`와 `coinfo.naver?target=cu_more`를 HTML 우선으로 읽는다.\n"
+        "- Naver 홈페이지 리뉴얼이나 DOM 변경이 생기면, JS는 보조 탐지용으로만 보고 실제값은 추정하지 않는다. 테이블이 없으면 실패를 그대로 남겨 추정값을 쓰지 않는다.\n"
+        "- `NAVER_ETF_PAGE_FAIL_LAYOUT_CHANGED`는 레이아웃 변경 전용 실패로 분리하고, 일반 실패(`NAVER_ETF_PAGE_FAIL`)와 구분해 읽는다.\n"
+        "- 금융 섹터는 `은행 / 증권 / 지주회사`로 분리해 `sector_universe`를 구성하고, `sector_flow`는 현재 JSON 브리지를 통해 carryover 분리본을 표시한다. GAS `runDataFeed`를 다시 돌리면 native 분리본으로 다시 물린다.\n"
+        "- 이 분리는 월 1회 갱신 하네스의 대상이며, 섹터별 대표 ETF 구성비 증빙은 `Source_URL`과 `Source_AsOf`가 유효해야만 인정한다.\n"
+    )
+    rows_data = result.get("rows") if isinstance(result.get("rows"), list) else []
+    if rows_data:
+        md += "\n\n**섹터 갱신 상세**\n\n" + _tbl(rows_data, [
+            "sector", "proxy_ticker", "proxy_name", "proxy_type", "source_kind", "transport_mode",
+            "source_url", "source_asof", "age_days", "constituent_count",
+            "stock_count", "etf_count", "weight_sum", "status", "refresh_reason",
+        ], max_rows=20)
+    return md
+
+
 def _etf_representative_monitor_v1(data_root: dict, hctx: dict, se: list) -> str:
     inner_data = data_root.get("data", {}) if isinstance(data_root.get("data"), dict) else {}
     payload = {"data": inner_data, "data_root": data_root, "_harness_context": hctx}
@@ -784,6 +832,11 @@ def _etf_representative_monitor_v1(data_root: dict, hctx: dict, se: list) -> str
     ])
     md += "\n\n**ETF 대표 종목 추출 원칙**\n\n"
     md += (
+        "- 섹터 프록시는 ETF 우선을 기본으로 두고, ETF가 실제로 있는 섹터는 ETF를 대표값으로 씁니다.\n"
+        "- 은행/증권/지주회사는 하나로 뭉치지 않고 각각 별도 섹터로 분리해 구성비 상위 종목을 증빙합니다.\n"
+        "- 방산/원전/건설/플랜트-EPC/로보틱스처럼 ETF 프록시가 있는 섹터는 ETF를 쓰고, 대표주 바스켓은 섹터별 기본 3종, 로보틱스는 5종으로 별도 모니터합니다.\n"
+        "- 로보틱스는 `RISE 현대차고정피지컬AI`를 섹터 프록시로 사용하고, 대표주는 해당 ETF의 실제 구성비 상위 5개 종목에서 뽑습니다.\n"
+        "- `Universe_Source=DEFAULT_TEMPLATE`인 행은 템플릿 경로이므로, 실제 시트 입력으로 바꿔 provenance를 완성해야 합니다.\n"
         "- 대표 종목은 우선 ETF 구성비중이 가장 큰 종목을 선택하고, 그 종목이 현재 유동성/호가/추세 조건을 충족하는지로 계속 모니터링합니다.\n"
         "- 구성비중 데이터가 비어 있거나 비정상일 때만 같은 섹터의 유동성 우선 후보로 대체합니다.\n"
         "- BUY_REVIEW는 ETF 수급이 대표 종목의 추세와 같이 붙을 때만 후보로 승격합니다.\n"
@@ -796,7 +849,7 @@ def _etf_representative_monitor_v1(data_root: dict, hctx: dict, se: list) -> str
             rep_states = []
             rep_weights = []
             if isinstance(reps, list):
-                for rep in reps[:3]:
+                for rep in reps[:5]:
                     if isinstance(rep, dict):
                         rep_names.append(f"{rep.get('name', '')}({rep.get('ticker', '')})")
                         rep_states.append(str(rep.get("monitor_state", "")))
@@ -805,6 +858,7 @@ def _etf_representative_monitor_v1(data_root: dict, hctx: dict, se: list) -> str
                 "sector": row.get("sector", ""),
                 "etf_proxy_ticker": row.get("etf_proxy_ticker", ""),
                 "etf_proxy_name": row.get("etf_proxy_name", ""),
+                "universe_source": row.get("universe_source", ""),
                 "representative_basket": " / ".join(rep_names),
                 "representative_count": row.get("representative_count", ""),
                 "basket_weights": ", ".join(rep_weights),
@@ -813,8 +867,8 @@ def _etf_representative_monitor_v1(data_root: dict, hctx: dict, se: list) -> str
                 "representative_basis_detail": row.get("representative_basis_detail", ""),
                 "basket_quality_state": row.get("basket_quality_state", ""),
                 "basket_coverage_pct": row.get("basket_coverage_pct", ""),
-                "selection_source": ", ".join(str(rep.get("selection_source", "")) for rep in reps[:3] if isinstance(rep, dict)),
-                "selection_score": ", ".join(str(rep.get("selection_score", "")) for rep in reps[:3] if isinstance(rep, dict)),
+                "selection_source": ", ".join(str(rep.get("selection_source", "")) for rep in reps[:5] if isinstance(rep, dict)),
+                "selection_score": ", ".join(str(rep.get("selection_score", "")) for rep in reps[:5] if isinstance(rep, dict)),
                 "basket_state": row.get("monitor_state", ""),
                 "basket_buy_review_count": row.get("basket_buy_review_count", ""),
                 "basket_caution_count": row.get("basket_caution_count", ""),
@@ -823,7 +877,7 @@ def _etf_representative_monitor_v1(data_root: dict, hctx: dict, se: list) -> str
             })
         md += "\n\n**대표 종목 모니터 테이블**\n\n"
         md += _tbl(display_rows, [
-            "sector", "etf_proxy_ticker", "etf_proxy_name", "representative_basket",
+            "sector", "etf_proxy_ticker", "etf_proxy_name", "universe_source", "representative_basket",
             "representative_count", "basket_weights", "basket_states", "representative_basis",
             "representative_basis_detail", "basket_quality_state", "basket_coverage_pct",
             "selection_source", "selection_score", "basket_state", "basket_buy_review_count",
@@ -1538,6 +1592,7 @@ def main() -> int:
         "market_context_learning_note":           lambda: _market_context_learning_note(hctx, se),
         "portfolio_performance_summary":          lambda: _portfolio_performance_summary(data_root, hctx, se),
         "portfolio_sector_exposure_summary":      lambda: _portfolio_sector_exposure_summary(data_root, hctx, se),
+        "sector_universe_refresh_audit_v1":       lambda: _sector_universe_refresh_audit_v1(data_root, hctx, se),
         "sector_trend_analysis_v1":               lambda: _sector_trend_analysis_v1(data_root, hctx, se),
         "investment_quality_headline":            lambda: _investment_quality_headline(hctx, se),
         "operational_truth_score":                lambda: _operational_truth_score(hctx, se),
diff --git a/tools/update_sector_universe_from_naver.py b/tools/update_sector_universe_from_naver.py
new file mode 100644
index 0000000..1a74021
--- /dev/null
+++ b/tools/update_sector_universe_from_naver.py
@@ -0,0 +1,616 @@
+from __future__ import annotations
+
+import argparse
+import datetime as dt
+import json
+import re
+import shutil
+import sys
+from collections import OrderedDict
+from pathlib import Path
+from typing import Any
+from urllib.parse import urljoin, urlparse, parse_qs
+
+import requests
+from bs4 import BeautifulSoup
+from openpyxl import load_workbook
+from openpyxl.styles import Alignment, Font, PatternFill
+from openpyxl.utils import get_column_letter
+
+ROOT = Path(__file__).resolve().parents[1]
+if str(ROOT) not in sys.path:
+    sys.path.insert(0, str(ROOT))
+
+from src.quant_engine.sector_universe_refresh import build_sector_universe_refresh_audit
+
+DEFAULT_INPUT_XLSX = ROOT / "GatherTradingData.xlsx"
+DEFAULT_OUTPUT_XLSX = ROOT / "outputs" / "sector_universe_refresh" / "GatherTradingData_sector_universe.xlsx"
+DEFAULT_USER_AGENT = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/126.0 Safari/537.36"
+NAVER_BASE = "https://finance.naver.com"
+NAVER_ITEM_CODE_RE = re.compile(r"(?:https?:)?//finance\.naver\.com(?P<path>/item/[^\"'\s<>]+code=(?P<code>\d+)[^\"'\s<>]*)", re.I)
+NAVER_REL_CODE_RE = re.compile(r"(?P<path>/item/[^\"'\s<>]+code=(?P<code>\d+)[^\"'\s<>]*)", re.I)
+
+TITLE_FILL = PatternFill("solid", fgColor="1F4E78")
+HEADER_FILL = PatternFill("solid", fgColor="1F4E78")
+SUBHEADER_FILL = PatternFill("solid", fgColor="D9EAF7")
+WHITE_FONT = Font(color="FFFFFF", bold=True)
+BOLD_FONT = Font(bold=True)
+NOTE_FONT = Font(italic=True, color="666666")
+
+
+def _kst_now() -> dt.datetime:
+    return dt.datetime.now(dt.timezone(dt.timedelta(hours=9)))
+
+
+def _kst_today() -> str:
+    return _kst_now().strftime("%Y-%m-%d")
+
+
+def _clean_text(value: Any) -> str:
+    if value is None:
+        return ""
+    return str(value).strip()
+
+
+def _normalize_code(value: Any) -> str:
+    text = _clean_text(value)
+    if not text:
+        return ""
+    text = text.replace(",", "")
+    if text.endswith(".0"):
+        text = text[:-2]
+    if text.isdigit():
+        return text.zfill(6)
+    if re.fullmatch(r"\d+\.\d+", text):
+        return str(int(float(text))).zfill(6)
+    return text
+
+
+def _parse_weight(value: str) -> float | None:
+    text = _clean_text(value).replace("%", "").replace(",", "")
+    if not text:
+        return None
+    try:
+        return float(text)
+    except Exception:
+        return None
+
+
+def _discover_naver_candidate_urls(soup: BeautifulSoup, proxy_ticker: str) -> list[str]:
+    candidates: list[str] = []
+    seen: set[str] = set()
+
+    def add(url: str) -> None:
+        url = _clean_text(url)
+        if not url or url in seen:
+            return
+        seen.add(url)
+        candidates.append(url)
+
+    expected_code = _normalize_code(proxy_ticker)
+
+    for script in soup.find_all("script"):
+        src = _clean_text(script.get("src"))
+        if src:
+            if expected_code and expected_code in src:
+                if src.startswith("//"):
+                    add(f"https:{src}")
+                elif src.startswith("/"):
+                    add(urljoin(NAVER_BASE, src))
+                else:
+                    add(src)
+            continue
+        text = script.get_text(" ", strip=True) or ""
+        if not text:
+            continue
+        for regex in (NAVER_ITEM_CODE_RE, NAVER_REL_CODE_RE):
+            for match in regex.finditer(text):
+                code = _normalize_code(match.groupdict().get("code") or "")
+                if expected_code and code and code != expected_code:
+                    continue
+                path = match.groupdict().get("path") or ""
+                if path:
+                    add(urljoin(NAVER_BASE, path))
+
+    return candidates
+
+
+def _parse_naver_etf_holdings(session: requests.Session, proxy_ticker: str, limit: int) -> dict[str, Any]:
+    url_candidates = [
+        f"{NAVER_BASE}/item/main.naver?code={proxy_ticker}",
+        f"{NAVER_BASE}/item/coinfo.naver?code={proxy_ticker}&target=cu_more",
+    ]
+
+    last_message = ""
+    for url in url_candidates:
+        response = session.get(url, timeout=20)
+        response.raise_for_status()
+        soup = BeautifulSoup(response.text, "html.parser")
+        discovered = _discover_naver_candidate_urls(soup, proxy_ticker)
+        for candidate in discovered:
+            if candidate not in url_candidates:
+                url_candidates.append(candidate)
+        section = soup.select_one("div.section.etf_asset")
+        table = section.select_one("table.tb_type1_a") if section is not None else None
+        if table is None:
+            # layout changed or this endpoint does not expose the constituent table
+            last_message = "ETF constituent table missing; page structure may have changed"
+            continue
+
+        holdings: list[dict[str, Any]] = []
+        for tr in table.select("tbody tr"):
+            tds = tr.find_all("td")
+            if len(tds) < 3:
+                continue
+            name_link = tr.find("a", href=re.compile(r"code=\d+"))
+            if name_link is None:
+                continue
+            name = _clean_text(name_link.get_text(" ", strip=True))
+            href = _clean_text(name_link.get("href"))
+            m = re.search(r"code=(\d+)", href)
+            code = _normalize_code(m.group(1) if m else "")
+            if not code or not name:
+                continue
+            weight = _parse_weight(tds[2].get_text(" ", strip=True))
+            if weight is None:
+                continue
+            holdings.append({
+                "Constituent_Code": code,
+                "Constituent_Name": name,
+                "Weight": round(weight / 100.0, 6),
+                "Source": "NAVER_ETF_PAGE",
+            })
+            if len(holdings) >= limit:
+                break
+
+        if holdings:
+            return {
+                "source_url": url,
+                "source_kind": "NAVER_ETF_PAGE",
+                "holdings": holdings,
+                "discovered_urls": discovered,
+                "message": "",
+            }
+        last_message = "no holdings parsed"
+
+    return {
+        "source_url": url_candidates[0],
+        "source_kind": "NAVER_ETF_PAGE_FAIL_LAYOUT_CHANGED",
+        "holdings": [],
+        "discovered_urls": [],
+        "message": last_message or "page structure changed; no expected values were inferred",
+    }
+
+
+def _extract_sector_seed_rows(ws) -> list[dict[str, Any]]:
+    headers = [ws.cell(2, c).value for c in range(1, ws.max_column + 1)]
+    headers = [str(h).strip() if h is not None else "" for h in headers]
+    idx = {name: i for i, name in enumerate(headers)}
+    rows: list[dict[str, Any]] = []
+    for r in range(3, ws.max_row + 1):
+        row = {name: ws.cell(r, c + 1).value for c, name in enumerate(headers) if name}
+        if not any(v not in (None, "") for v in row.values()):
+            continue
+        rows.append(row)
+    return rows
+
+
+def _group_seed_rows(rows: list[dict[str, Any]]) -> OrderedDict[str, dict[str, Any]]:
+    grouped: OrderedDict[str, dict[str, Any]] = OrderedDict()
+    for row in rows:
+        sector = _clean_text(row.get("Sector"))
+        if not sector:
+            continue
+        if sector not in grouped:
+            grouped[sector] = {
+                "meta": row,
+                "rows": [],
+            }
+        grouped[sector]["rows"].append(row)
+    return grouped
+
+
+def _build_refreshed_rows(seed_rows: list[dict[str, Any]], limit: int) -> tuple[list[dict[str, Any]], dict[str, Any]]:
+    session = requests.Session()
+    session.headers.update({"User-Agent": DEFAULT_USER_AGENT})
+    grouped = _group_seed_rows(seed_rows)
+    refreshed: list[dict[str, Any]] = []
+    sector_stats: list[dict[str, Any]] = []
+    today = _kst_today()
+
+    for sector, bundle in grouped.items():
+        meta = bundle["meta"]
+        proxy_ticker = _normalize_code(meta.get("Proxy_Ticker"))
+        proxy_name = _clean_text(meta.get("Proxy_Name"))
+        proxy_type = _clean_text(meta.get("Proxy_Type")) or "ETF"
+        base_ticker = _normalize_code(meta.get("Base_Ticker")) or "069500"
+
+        if sector == "금융/은행":
+            split_specs = [
+                {"sector": "은행", "proxy_ticker": "091170", "proxy_name": "KODEX 은행", "proxy_type": "ETF"},
+                {"sector": "증권", "proxy_ticker": "0111J0", "proxy_name": "HANARO 증권고배당TOP3플러스", "proxy_type": "ETF"},
+                {"sector": "지주회사", "proxy_ticker": "307520", "proxy_name": "TIGER 지주회사", "proxy_type": "ETF"},
+            ]
+            for spec in split_specs:
+                split_proxy_ticker = _normalize_code(spec["proxy_ticker"])
+                split_proxy_name = _clean_text(spec["proxy_name"])
+                split_proxy_type = _clean_text(spec["proxy_type"]) or "ETF"
+                split_source = "SHEET_INPUT"
+                split_source_url = ""
+                split_message = ""
+                split_source_kind = "SHEET_INPUT"
+                try:
+                    scraped = _parse_naver_etf_holdings(session, split_proxy_ticker, limit)
+                    split_source_url = scraped.get("source_url", "")
+                    split_source_kind = scraped.get("source_kind", "NAVER_ETF_PAGE_FAIL")
+                    holdings = scraped.get("holdings", [])
+                    split_message = scraped.get("message", "")
+                    if holdings:
+                        split_source = "NAVER_ETF_PAGE"
+                        weight_sum = round(sum(float(h["Weight"]) for h in holdings), 6)
+                        for h in holdings:
+                            refreshed.append({
+                                "Sector": spec["sector"],
+                                "Proxy_Ticker": split_proxy_ticker,
+                                "Proxy_Name": split_proxy_name,
+                                "Proxy_Type": split_proxy_type,
+                                "Base_Ticker": base_ticker,
+                                "Constituent_Code": h["Constituent_Code"],
+                                "Constituent_Name": h["Constituent_Name"],
+                                "Weight": h["Weight"],
+                                "Is_ETF": "N",
+                                "Enabled": "Y",
+                                "Effective_Date": today,
+                                "Source": split_source,
+                                "Transport_Mode": "HTML_SERVER_RENDERED",
+                                "Source_URL": split_source_url,
+                                "Source_AsOf": today,
+                                "Sector_Check": spec["sector"],
+                                "Weight_Sum_All": weight_sum,
+                                "Weight_Sum_Stocks_Only": weight_sum,
+                                "ETF_Rows": 0,
+                                "Status": "OK",
+                            })
+                        sector_stats.append({
+                            "sector": spec["sector"],
+                            "proxy_ticker": split_proxy_ticker,
+                            "proxy_name": split_proxy_name,
+                            "proxy_type": split_proxy_type,
+                            "source_kind": split_source,
+                            "transport_mode": "HTML_SERVER_RENDERED",
+                            "source_url": split_source_url,
+                            "source_asof": today,
+                            "constituent_count": len(holdings),
+                            "weight_sum": weight_sum,
+                            "status": "CURRENT",
+                            "refresh_reason": "NAVER_ETF_PAGE_SPLIT",
+                        })
+                        continue
+                except Exception as exc:
+                    split_message = str(exc)
+                    split_source_kind = "NAVER_ETF_PAGE_FAIL"
+
+                # 실패 시는 투명하게 남기고, 섹터 누락은 그대로 드러낸다.
+                sector_stats.append({
+                    "sector": spec["sector"],
+                    "proxy_ticker": split_proxy_ticker,
+                    "proxy_name": split_proxy_name,
+                    "proxy_type": split_proxy_type,
+                    "source_kind": split_source_kind,
+                    "transport_mode": "LAYOUT_CHANGED" if split_source_kind == "NAVER_ETF_PAGE_FAIL_LAYOUT_CHANGED" else "UNKNOWN",
+                    "source_url": split_source_url,
+                    "source_asof": today,
+                    "constituent_count": 0,
+                    "weight_sum": 0.0,
+                    "status": "FAIL" if "FAIL" in split_source_kind else "WARN",
+                    "refresh_reason": split_message or "split_sector_fallback",
+                })
+            continue
+
+        source = "SHEET_INPUT"
+        source_url = ""
+        message = ""
+        source_kind = "SHEET_INPUT"
+
+        if proxy_type != "ETF":
+            source_kind = "REPRESENTATIVE_STOCK_PROXY"
+            source = source_kind
+            source_url = f"{NAVER_BASE}/item/main.naver?code={proxy_ticker}" if proxy_ticker else ""
+            fallback_rows = bundle["rows"][:limit] if bundle["rows"] else []
+            weight_sum = 0.0
+            for row in fallback_rows:
+                weight = row.get("Weight")
+                try:
+                    weight_sum += float(weight) if weight not in (None, "") else 0.0
+                except Exception:
+                    pass
+                refreshed.append({
+                    "Sector": sector,
+                    "Proxy_Ticker": proxy_ticker,
+                    "Proxy_Name": proxy_name,
+                    "Proxy_Type": proxy_type,
+                    "Base_Ticker": base_ticker,
+                    "Constituent_Code": _normalize_code(row.get("Constituent_Code")),
+                    "Constituent_Name": _clean_text(row.get("Constituent_Name")),
+                    "Weight": float(row.get("Weight") or 0),
+                    "Is_ETF": _clean_text(row.get("Is_ETF")) or "N",
+                    "Enabled": "Y",
+                    "Effective_Date": today,
+                    "Source": source_kind,
+                    "Transport_Mode": "HTML_SERVER_RENDERED" if source_kind == "REPRESENTATIVE_STOCK_PROXY" else "MANUAL_OR_TEMPLATE",
+                    "Source_URL": source_url,
+                    "Source_AsOf": today,
+                    "Sector_Check": sector,
+                    "Weight_Sum_All": weight_sum,
+                    "Weight_Sum_Stocks_Only": weight_sum,
+                    "ETF_Rows": 0,
+                    "Status": "CURRENT",
+                })
+            sector_stats.append({
+                "sector": sector,
+                "proxy_ticker": proxy_ticker,
+                "proxy_name": proxy_name,
+                "proxy_type": proxy_type,
+                "source_kind": source_kind,
+                "transport_mode": "HTML_SERVER_RENDERED" if source_kind == "REPRESENTATIVE_STOCK_PROXY" else "MANUAL_OR_TEMPLATE",
+                "source_url": source_url,
+                "source_asof": today,
+                "constituent_count": len(fallback_rows),
+                "weight_sum": round(weight_sum, 6),
+                "status": "CURRENT",
+                "refresh_reason": "REPRESENTATIVE_STOCK_PROXY",
+            })
+            continue
+
+        if proxy_ticker:
+            try:
+                scraped = _parse_naver_etf_holdings(session, proxy_ticker, limit)
+                source_url = scraped.get("source_url", "")
+                source_kind = scraped.get("source_kind", "NAVER_ETF_PAGE_FAIL")
+                holdings = scraped.get("holdings", [])
+                message = scraped.get("message", "")
+                if holdings:
+                    source = "NAVER_ETF_PAGE"
+                    weight_sum = round(sum(float(h["Weight"]) for h in holdings), 6)
+                    for h in holdings:
+                        refreshed.append({
+                            "Sector": sector,
+                            "Proxy_Ticker": proxy_ticker,
+                            "Proxy_Name": proxy_name,
+                            "Proxy_Type": proxy_type,
+                            "Base_Ticker": base_ticker,
+                            "Constituent_Code": h["Constituent_Code"],
+                            "Constituent_Name": h["Constituent_Name"],
+                            "Weight": h["Weight"],
+                            "Is_ETF": "N",
+                            "Enabled": "Y",
+                            "Effective_Date": today,
+                            "Source": source,
+                            "Transport_Mode": "HTML_SERVER_RENDERED",
+                            "Source_URL": source_url,
+                            "Source_AsOf": today,
+                            "Sector_Check": sector,
+                            "Weight_Sum_All": weight_sum,
+                            "Weight_Sum_Stocks_Only": weight_sum,
+                            "ETF_Rows": 0,
+                            "Status": "OK",
+                        })
+                    sector_stats.append({
+                        "sector": sector,
+                        "proxy_ticker": proxy_ticker,
+                        "proxy_name": proxy_name,
+                        "proxy_type": proxy_type,
+                        "source_kind": source,
+                        "transport_mode": "HTML_SERVER_RENDERED",
+                        "source_url": source_url,
+                        "source_asof": today,
+                        "constituent_count": len(holdings),
+                        "weight_sum": weight_sum,
+                        "status": "CURRENT",
+                        "refresh_reason": "NAVER_ETF_PAGE",
+                    })
+                    continue
+            except Exception as exc:
+                message = str(exc)
+                source_kind = "NAVER_ETF_PAGE_FAIL"
+
+        # fallback: preserve seed rows but expose the failure transparently
+        fallback_rows = bundle["rows"][:limit] if bundle["rows"] else []
+        weight_sum = 0.0
+        for row in fallback_rows:
+            weight = row.get("Weight")
+            try:
+                weight_sum += float(weight) if weight not in (None, "") else 0.0
+            except Exception:
+                pass
+            refreshed.append({
+                "Sector": sector,
+                "Proxy_Ticker": proxy_ticker,
+                "Proxy_Name": proxy_name,
+                "Proxy_Type": proxy_type,
+                "Base_Ticker": base_ticker,
+                "Constituent_Code": _normalize_code(row.get("Constituent_Code")),
+                "Constituent_Name": _clean_text(row.get("Constituent_Name")),
+                "Weight": float(row.get("Weight") or 0),
+                "Is_ETF": _clean_text(row.get("Is_ETF")) or "N",
+                "Enabled": "Y",
+                "Effective_Date": today,
+                "Source": source_kind,
+                "Transport_Mode": "LAYOUT_CHANGED" if source_kind == "NAVER_ETF_PAGE_FAIL_LAYOUT_CHANGED" else "UNKNOWN",
+                "Source_URL": source_url,
+                "Source_AsOf": today,
+                "Sector_Check": sector,
+                "Weight_Sum_All": weight_sum,
+                "Weight_Sum_Stocks_Only": weight_sum,
+                "ETF_Rows": 0,
+                "Status": "FAIL" if source_kind.endswith("FAIL") else "WARN",
+            })
+        sector_stats.append({
+            "sector": sector,
+            "proxy_ticker": proxy_ticker,
+            "proxy_name": proxy_name,
+            "proxy_type": proxy_type,
+            "source_kind": source_kind,
+            "transport_mode": "LAYOUT_CHANGED" if source_kind == "NAVER_ETF_PAGE_FAIL_LAYOUT_CHANGED" else "UNKNOWN",
+            "source_url": source_url,
+            "source_asof": today,
+            "constituent_count": len(fallback_rows),
+            "weight_sum": round(weight_sum, 6),
+            "status": "FAIL" if "FAIL" in source_kind else "WARN",
+            "refresh_reason": message or "seed_fallback",
+        })
+
+    audit_payload = build_sector_universe_refresh_audit({"data": {"sector_universe": refreshed}})
+    return refreshed, {
+        "sector_universe_refresh_audit": audit_payload,
+        "sector_stats": sector_stats,
+    }
+
+
+def _style_title(ws, title: str, subtitle: str) -> None:
+    ws.merge_cells(start_row=1, start_column=1, end_row=1, end_column=max(8, ws.max_column or 8))
+    ws["A1"] = title
+    ws["A1"].font = WHITE_FONT
+    ws["A1"].fill = TITLE_FILL
+    ws["A1"].alignment = Alignment(horizontal="left")
+    ws.merge_cells(start_row=2, start_column=1, end_row=2, end_column=max(8, ws.max_column or 8))
+    ws["A2"] = subtitle
+    ws["A2"].font = NOTE_FONT
+
+
+def _write_table(ws, start_row: int, start_col: int, headers: list[str], rows: list[list[Any]]) -> int:
+    for i, header in enumerate(headers, start=start_col):
+        cell = ws.cell(start_row, i)
+        cell.value = header
+        cell.font = WHITE_FONT
+        cell.fill = HEADER_FILL
+        cell.alignment = Alignment(horizontal="center")
+    for r_idx, row in enumerate(rows, start=start_row + 1):
+        for c_idx, value in enumerate(row, start=start_col):
+            ws.cell(r_idx, c_idx).value = value
+    return start_row + len(rows)
+
+
+def _write_sector_universe_sheet(wb, rows: list[dict[str, Any]]) -> None:
+    if "sector_universe" in wb.sheetnames:
+        del wb["sector_universe"]
+    ws = wb.create_sheet("sector_universe")
+    headers = [
+        "Sector", "Proxy_Ticker", "Proxy_Name", "Proxy_Type", "Base_Ticker",
+        "Constituent_Code", "Constituent_Name", "Weight", "Is_ETF", "Enabled",
+        "Effective_Date", "Source", "Transport_Mode", "Source_URL", "Source_AsOf", "Sector_Check",
+        "Weight_Sum_All", "Weight_Sum_Stocks_Only", "ETF_Rows", "Status",
+    ]
+    now = _kst_now().strftime("%Y-%m-%d %H:%M:%S")
+    ws["A1"] = f"updated: {now} KST"
+    ws["A1"].font = Font(bold=True)
+    _write_table(ws, 2, 1, headers, [[r.get(h, "") for h in headers] for r in rows])
+    for col_idx, header in enumerate(headers, start=1):
+        if header in {"Proxy_Ticker", "Base_Ticker", "Constituent_Code"}:
+            for r in range(3, ws.max_row + 1):
+                ws.cell(r, col_idx).number_format = "@"
+        if header in {"Weight", "Weight_Sum_All", "Weight_Sum_Stocks_Only"}:
+            for r in range(3, ws.max_row + 1):
+                ws.cell(r, col_idx).number_format = "0.0000"
+        width = 16
+        if header in {"Constituent_Name", "Proxy_Name"}:
+            width = 22
+        elif header in {"Source_URL"}:
+            width = 42
+        elif header in {"Status", "Source", "Sector_Check", "Proxy_Type", "Transport_Mode"}:
+            width = 16
+        ws.column_dimensions[get_column_letter(col_idx)].width = width
+    ws.freeze_panes = "A3"
+    ws.sheet_view.showGridLines = False
+
+
+def _write_audit_sheet(wb, audit_payload: dict[str, Any]) -> None:
+    audit = audit_payload["sector_universe_refresh_audit"]
+    if "sector_universe_refresh_audit" in wb.sheetnames:
+        del wb["sector_universe_refresh_audit"]
+    ws = wb.create_sheet("sector_universe_refresh_audit")
+    ws.sheet_view.showGridLines = False
+    _style_title(
+        ws,
+        "섹터 월간 갱신 감사",
+        "Naver ETF 페이지 기반 월간 갱신 상태와 provenance 분리 현황을 점검한다.",
+    )
+    summary = audit.get("summary", {})
+    summary_rows = [
+        ["formula_id", audit.get("formula_id", "")],
+        ["gate", audit.get("gate", "")],
+        ["sector_count", summary.get("sector_count", 0)],
+        ["current_count", summary.get("current_count", 0)],
+        ["due_count", summary.get("due_count", 0)],
+        ["overdue_count", summary.get("overdue_count", 0)],
+        ["missing_count", summary.get("missing_count", 0)],
+        ["template_count", summary.get("template_count", 0)],
+        ["sheet_input_count", summary.get("sheet_input_count", 0)],
+        ["naver_source_count", summary.get("naver_source_count", 0)],
+        ["missing_source_url_count", summary.get("missing_source_url_count", 0)],
+        ["stale_sector_count", summary.get("stale_sector_count", 0)],
+        ["oldest_source_asof", summary.get("oldest_source_asof", "")],
+        ["newest_source_asof", summary.get("newest_source_asof", "")],
+    ]
+    _write_table(ws, 4, 1, ["key", "value"], summary_rows)
+    rows = audit.get("rows", []) or []
+    if rows:
+        headers = [
+            "sector", "proxy_ticker", "proxy_name", "proxy_type", "source_kind",
+            "source_url", "source_asof", "age_days", "constituent_count",
+            "stock_count", "etf_count", "weight_sum", "status", "refresh_reason",
+        ]
+        _write_table(ws, 4, 4, headers, [[r.get(h, "") for h in headers] for r in rows])
+        for idx, header in enumerate(headers, start=4):
+            width = 16
+            if header in {"sector", "proxy_name", "refresh_reason"}:
+                width = 20
+            elif header == "source_url":
+                width = 42
+            ws.column_dimensions[get_column_letter(idx)].width = width
+    ws.freeze_panes = "A5"
+
+
+def main() -> int:
+    ap = argparse.ArgumentParser()
+    ap.add_argument("--input", default=str(DEFAULT_INPUT_XLSX))
+    ap.add_argument("--output", default=str(DEFAULT_OUTPUT_XLSX))
+    ap.add_argument("--limit", type=int, default=10, help="Per-sector holdings limit from Naver ETF pages")
+    ap.add_argument("--apply", action="store_true", help="Overwrite the input workbook in place as well")
+    args = ap.parse_args()
+
+    input_path = Path(args.input)
+    output_path = Path(args.output)
+    if not input_path.exists():
+        raise FileNotFoundError(input_path)
+
+    wb = load_workbook(input_path)
+    if "sector_universe" not in wb.sheetnames:
+        raise RuntimeError("sector_universe sheet not found")
+    seed_ws = wb["sector_universe"]
+    seed_rows = _extract_sector_seed_rows(seed_ws)
+    refreshed_rows, audit_payload = _build_refreshed_rows(seed_rows, max(1, args.limit))
+
+    _write_sector_universe_sheet(wb, refreshed_rows)
+    _write_audit_sheet(wb, audit_payload)
+
+    output_path.parent.mkdir(parents=True, exist_ok=True)
+    wb.save(output_path)
+    if args.apply and input_path.resolve() != output_path.resolve():
+        shutil.copy2(output_path, input_path)
+
+    print(json.dumps({
+        "status": "OK",
+        "input": str(input_path),
+        "output": str(output_path),
+        "rows": len(refreshed_rows),
+        "sectors": len(audit_payload["sector_stats"]),
+        "current_count": audit_payload["sector_universe_refresh_audit"]["summary"]["current_count"],
+        "overdue_count": audit_payload["sector_universe_refresh_audit"]["summary"]["overdue_count"],
+        "template_count": audit_payload["sector_universe_refresh_audit"]["summary"]["template_count"],
+    }, ensure_ascii=False, indent=2))
+    return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())
diff --git a/tools/update_workbook_sector_insights.py b/tools/update_workbook_sector_insights.py
index e793263..d5191b1 100644
--- a/tools/update_workbook_sector_insights.py
+++ b/tools/update_workbook_sector_insights.py
@@ -9,8 +9,14 @@ from openpyxl.chart import BarChart, LineChart, Reference
 from openpyxl.styles import Font, PatternFill, Alignment
 from openpyxl.utils import get_column_letter
 
+import sys
 
 ROOT = Path(__file__).resolve().parent.parent
+if str(ROOT) not in sys.path:
+    sys.path.insert(0, str(ROOT))
+
+from src.quant_engine.sector_universe_refresh import build_sector_universe_refresh_audit
+
 INPUT_XLSX = ROOT / "GatherTradingData.xlsx"
 OUTPUT_DIR = ROOT / "outputs" / "sector_insights_enhanced"
 OUTPUT_XLSX = OUTPUT_DIR / "GatherTradingData_sector_insights.xlsx"
@@ -593,10 +599,12 @@ def build_sector_summary(wb, data: dict) -> None:
     ws["A20"] = "Notes"
     ws["A20"].fill = SUBHEADER_FILL
     ws["A20"].font = BOLD_FONT
-    ws["A21"] = "섹터별 ETF 프록시와 스마트머니 방향이 다르면 매수 근거를 보수적으로 해석해야 합니다."
+    ws["A21"] = "섹터별 ETF 프록시를 기준으로 보고, 은행/증권/지주회사는 분리해서 구성비 상위 종목을 증빙해야 합니다. 대표주 모니터는 섹터 기본 3종, 로보틱스 5종 바스켓으로 함께 확인해야 합니다."
     ws["A21"].alignment = Alignment(wrap_text=True)
-    ws["A22"] = "데이터 결측은 하네스 업데이트가 필요합니다."
+    ws["A22"] = "Universe_Source가 DEFAULT_TEMPLATE인 행은 템플릿이며, 실제 시트 입력으로 전환되어야 provenance가 완성됩니다."
     ws["A22"].alignment = Alignment(wrap_text=True)
+    ws["A23"] = "다음 세분화 후보는 바이오/제약과 방산/우주처럼 현재 섹터를 더 세밀하게 나누는 방향입니다. 로보틱스는 RISE 현대차고정피지컬AI를 섹터 프록시로 사용하고, 대표주는 해당 ETF의 실제 구성비 상위 5개 종목에서 뽑습니다."
+    ws["A23"].alignment = Alignment(wrap_text=True)
 
     chart = LineChart()
     chart.title = "Average Sector Score / Breadth Trend"
@@ -622,11 +630,11 @@ def build_sector_analysis(wb, data: dict) -> None:
     style_title(
         ws,
         "섹터 동향 분석",
-        "섹터별 ETF 프록시, 스마트머니 유입, 수익률, 유동성 방향을 함께 보는 상세 시트",
+        "섹터별 ETF 프록시, 대표주 모니터, 스마트머니 유입, 수익률, 유동성 방향을 함께 보는 상세 시트",
         end_col=18,
     )
     headers = [
-        "sector", "proxy_ticker", "proxy_name", "proxy_type", "etf_code",
+        "sector", "proxy_ticker", "proxy_name", "proxy_type", "universe_source", "etf_code",
         "etf_execution_use", "etf_liquidity_score", "etf_liquidity_status", "etf_nav_risk",
         "proxy_confidence", "rank", "rank_delta_w1", "rank_delta_w2", "sector_score",
         "score_delta", "sector_ret5d", "sector_ret20d", "etf_return_5d", "etf_return_20d",
@@ -661,7 +669,7 @@ def build_sector_analysis(wb, data: dict) -> None:
     chart.x_axis.title = "20D Return"
     chart.height = 8
     chart.width = 14
-    data_ref = Reference(ws, min_col=17, min_row=4, max_row=4 + len(rows))
+    data_ref = Reference(ws, min_col=18, min_row=4, max_row=4 + len(rows))
     cats = Reference(ws, min_col=1, min_row=5, max_row=4 + len(rows))
     chart.add_data(data_ref, titles_from_data=True)
     chart.set_categories(cats)
@@ -817,6 +825,67 @@ def build_sector_timeline(wb, data: dict, source_data: dict | None = None) -> No
             ws.add_chart(money_chart, "L36")
 
 
+def build_sector_universe_refresh_audit_sheet(wb, source_data: dict) -> None:
+    ws = wb.create_sheet("sector_universe_refresh_audit")
+    style_sheet(ws)
+    style_title(
+        ws,
+        "섹터 월간 갱신 감사",
+        "Naver ETF 페이지 기반 구성종목 갱신 상태와 provenance 분리 상태를 점검하는 감사 시트. AJAX/XHR 전제는 두지 않고 HTML 서버렌더링 테이블을 우선한다.",
+        end_col=16,
+    )
+    payload = {"data": source_data}
+    audit = build_sector_universe_refresh_audit(payload)
+    summary = audit.get("summary") or {}
+    items = [
+        ("formula_id", audit.get("formula_id", "")),
+        ("gate", audit.get("gate", "")),
+        ("sector_count", summary.get("sector_count", 0)),
+        ("current_count", summary.get("current_count", 0)),
+        ("due_count", summary.get("due_count", 0)),
+        ("overdue_count", summary.get("overdue_count", 0)),
+        ("layout_changed_count", summary.get("layout_changed_count", 0)),
+        ("missing_count", summary.get("missing_count", 0)),
+        ("template_count", summary.get("template_count", 0)),
+        ("sheet_input_count", summary.get("sheet_input_count", 0)),
+        ("naver_source_count", summary.get("naver_source_count", 0)),
+        ("missing_source_url_count", summary.get("missing_source_url_count", 0)),
+        ("stale_sector_count", summary.get("stale_sector_count", 0)),
+    ]
+    add_kpi_block(ws, 4, items)
+    ws["D4"] = "Refresh policy"
+    ws["D4"].fill = SUBHEADER_FILL
+    ws["D4"].font = BOLD_FONT
+    ws["D5"] = "NAVER_ETF_PAGE rows are the monthly refreshed source."
+    ws["D6"] = "SHEET_INPUT rows are manual/provisional and must stay separate."
+    ws["D7"] = "DEFAULT_TEMPLATE rows are a fail in the monthly gate."
+    ws["D8"] = "Source_URL and Source_AsOf are required for provenance."
+    ws["D9"] = "This is HTML-server-rendered, not AJAX. JS is only a fallback probe for candidate URLs."
+    ws["D10"] = "No guessed holdings are written when the page layout changes."
+    ws["D11"] = "NAVER_ETF_PAGE_FAIL_LAYOUT_CHANGED is a separate layout-change failure state."
+    ws["D12"] = "Financial sectors are split as 은행 / 증권 / 지주회사 in sector_universe; sector_flow reflects carryover until GAS runDataFeed is rerun."
+    ws["D13"] = "This split is part of the monthly refresh harness; Source_URL and Source_AsOf must remain valid for provenance."
+    rows = audit.get("rows") or []
+    if rows:
+        headers = [
+            "sector", "proxy_ticker", "proxy_name", "proxy_type", "source_kind", "transport_mode",
+            "source_url", "source_asof", "age_days", "constituent_count",
+            "stock_count", "etf_count", "weight_sum", "status", "refresh_reason",
+        ]
+        write_table(ws, 14, 1, headers, [[row.get(h, "") for h in headers] for row in rows])
+        for col, width in {
+            "A": 16, "B": 12, "C": 18, "D": 12, "E": 16, "F": 18, "G": 42, "H": 14,
+            "I": 10, "J": 14, "K": 12, "L": 12, "M": 12, "N": 12, "O": 24,
+        }.items():
+            ws.column_dimensions[col].width = width
+    ws.freeze_panes = "A5"
+    ws["A11"] = "Notes"
+    ws["A11"].fill = SUBHEADER_FILL
+    ws["A11"].font = BOLD_FONT
+    ws["A12"] = "홈페이지 리뉴얼로 표 구조가 바뀌면, 파서는 추정하지 않고 실패 상태를 남겨 월간 게이트에서 잡는다."
+    ws["A12"].alignment = Alignment(wrap_text=True)
+
+
 def build_etf_summary(wb, data: dict) -> None:
     ws = wb.create_sheet("etf_representative_summary")
     style_sheet(ws)
@@ -847,6 +916,7 @@ def build_etf_summary(wb, data: dict) -> None:
     ws["D6"] = "2) Missing slots filled with same-sector live candidates"
     ws["D7"] = "3) Missing data stays explicit as DATA_MISSING"
     ws["D8"] = "4) Minimum 3 names per sector basket"
+    ws["D9"] = "5) Universe_Source=DEFAULT_TEMPLATE rows are provisional until sheet-backed data exists."
     ws["G4"] = "Top reps"
     ws["G4"].fill = SUBHEADER_FILL
     ws["G4"].font = BOLD_FONT
@@ -865,7 +935,7 @@ def build_etf_monitor(wb, data: dict) -> None:
         end_col=18,
     )
     headers = [
-        "sector", "etf_proxy_ticker", "etf_proxy_name", "etf_proxy_type", "sector_rank",
+        "sector", "etf_proxy_ticker", "etf_proxy_name", "etf_proxy_type", "universe_source", "sector_rank",
         "sector_score", "sector_smart_money_5d_krw", "sector_ret20d", "representative_count",
         "representative_ticker", "representative_name", "representative_basis",
         "representative_basis_detail", "constituent_weight", "basket_quality_state",
@@ -894,7 +964,7 @@ def build_etf_monitor(wb, data: dict) -> None:
     chart.x_axis.title = "Coverage %"
     chart.height = 8
     chart.width = 14
-    data_ref = Reference(ws, min_col=16, min_row=4, max_row=4 + len(rows))
+    data_ref = Reference(ws, min_col=17, min_row=4, max_row=4 + len(rows))
     cats = Reference(ws, min_col=1, min_row=5, max_row=4 + len(rows))
     chart.add_data(data_ref, titles_from_data=True)
     chart.set_categories(cats)
@@ -922,6 +992,7 @@ def main() -> None:
         "performance_readiness_summary",
         "operational_eval_queue_summary",
         "portfolio_sector_exposure",
+        "sector_universe_refresh_audit",
         "_portfolio_holdings_helper",
         "sector_trend_summary",
         "sector_trend_analysis",
@@ -936,6 +1007,7 @@ def main() -> None:
     build_performance_readiness_summary(wb)
     build_operational_eval_queue_summary(wb)
     build_portfolio_sector_exposure(wb)
+    build_sector_universe_refresh_audit_sheet(wb, raw_source)
     build_sector_timeline(wb, sector, raw_source)
     build_sector_analysis(wb, sector)
     build_sector_summary(wb, sector)
@@ -949,6 +1021,7 @@ def main() -> None:
         "performance_readiness_summary",
         "operational_eval_queue_summary",
         "portfolio_sector_exposure",
+        "sector_universe_refresh_audit",
         "sector_trend_summary",
         "sector_trend_analysis",
         "sector_trend_timeline",
diff --git a/tools/validate_report_section_completeness_v1.py b/tools/validate_report_section_completeness_v1.py
index 9f2b94f..dcff732 100644
--- a/tools/validate_report_section_completeness_v1.py
+++ b/tools/validate_report_section_completeness_v1.py
@@ -21,6 +21,7 @@ REPORT_SECTION_ORDER = [
     "single_conclusion", "immediate_execution_playbook", "market_context_learning_note",
     "portfolio_performance_summary",
     "portfolio_sector_exposure_summary",
+    "sector_universe_refresh_audit_v1",
     "sector_trend_analysis_v1",
     "etf_representative_monitor_v1",
     "performance_readiness_summary",
diff --git a/tools/validate_sector_universe_monthly_refresh_v1.py b/tools/validate_sector_universe_monthly_refresh_v1.py
new file mode 100644
index 0000000..52fa3a8
--- /dev/null
+++ b/tools/validate_sector_universe_monthly_refresh_v1.py
@@ -0,0 +1,173 @@
+#!/usr/bin/env python3
+from __future__ import annotations
+
+import argparse
+import datetime as dt
+import json
+import sys
+from pathlib import Path
+from typing import Any
+
+from openpyxl import load_workbook
+
+
+ROOT = Path(__file__).resolve().parents[1]
+DEFAULT_XLSX = ROOT / "GatherTradingData.xlsx"
+MAX_AGE_DAYS = 31
+
+
+def _txt(value: Any, default: str = "") -> str:
+    if value is None:
+        return default
+    if isinstance(value, str):
+        return value.strip() or default
+    return str(value).strip() or default
+
+
+def _parse_date(value: Any) -> dt.date | None:
+    text = _txt(value)
+    if not text:
+        return None
+    for fmt in ("%Y-%m-%d", "%Y.%m.%d", "%Y/%m/%d"):
+        try:
+            return dt.datetime.strptime(text[:10], fmt).date()
+        except Exception:
+            pass
+    try:
+        return dt.date.fromisoformat(text[:10])
+    except Exception:
+        return None
+
+
+def _age_days(value: Any) -> int | None:
+    parsed = _parse_date(value)
+    if parsed is None:
+        return None
+    today = dt.datetime.now(dt.timezone(dt.timedelta(hours=9))).date()
+    return (today - parsed).days
+
+
+def main() -> int:
+    ap = argparse.ArgumentParser()
+    ap.add_argument("--xlsx", default=str(DEFAULT_XLSX))
+    args = ap.parse_args()
+
+    xlsx = Path(args.xlsx)
+    if not xlsx.exists():
+        print(f"[오류] 워크북 없음: {xlsx}")
+        return 1
+
+    wb = load_workbook(xlsx, data_only=True)
+    if "sector_universe" not in wb.sheetnames:
+        print("[FAIL] sector_universe 시트가 없습니다.")
+        return 1
+
+    ws = wb["sector_universe"]
+    headers = [_txt(ws.cell(2, c).value) for c in range(1, ws.max_column + 1)]
+    idx = {name: i for i, name in enumerate(headers) if name}
+    required = ["Sector", "Proxy_Ticker", "Constituent_Code", "Weight", "Source", "Source_URL", "Source_AsOf"]
+    missing_headers = [h for h in required if h not in idx]
+
+    rows: list[dict[str, Any]] = []
+    for r in range(3, ws.max_row + 1):
+        row = {h: ws.cell(r, c + 1).value for c, h in enumerate(headers) if h}
+        if not any(v not in (None, "") for v in row.values()):
+            continue
+        rows.append(row)
+
+    sector_map: dict[str, list[dict[str, Any]]] = {}
+    for row in rows:
+        sector = _txt(row.get("Sector"))
+        if sector:
+            sector_map.setdefault(sector, []).append(row)
+
+    template_rows = 0
+    representative_rows = 0
+    sheet_input_rows = 0
+    naver_rows = 0
+    layout_changed_rows = 0
+    fail_rows = 0
+    missing_source_url = 0
+    stale_rows = 0
+    mixed_sector_count = 0
+    sector_status_rows: list[str] = []
+
+    for sector, sector_rows in sector_map.items():
+        source_kinds = {_txt(r.get("Source"), "SHEET_INPUT") or "SHEET_INPUT" for r in sector_rows}
+        if len(source_kinds) > 1:
+            mixed_sector_count += 1
+            sector_status_rows.append(f"{sector}:MIXED({','.join(sorted(source_kinds))})")
+
+        sector_template = any(src == "DEFAULT_TEMPLATE" for src in source_kinds)
+        sector_rep = any(src == "REPRESENTATIVE_STOCK_PROXY" for src in source_kinds)
+        sector_input = any(src == "SHEET_INPUT" for src in source_kinds)
+        sector_naver = any(src == "NAVER_ETF_PAGE" for src in source_kinds)
+        sector_layout_changed = any(src == "NAVER_ETF_PAGE_FAIL_LAYOUT_CHANGED" for src in source_kinds)
+        sector_fail = any("FAIL" in src for src in source_kinds)
+
+        if sector_template:
+            template_rows += len(sector_rows)
+        if sector_rep:
+            representative_rows += len(sector_rows)
+        if sector_input:
+            sheet_input_rows += len(sector_rows)
+        if sector_naver:
+            naver_rows += len(sector_rows)
+        if sector_layout_changed:
+            layout_changed_rows += len(sector_rows)
+        if sector_fail:
+            fail_rows += len(sector_rows)
+
+        source_urls = {_txt(r.get("Source_URL")) for r in sector_rows if _txt(r.get("Source_URL"))}
+        if not source_urls:
+            missing_source_url += len(sector_rows)
+
+        ages = [_age_days(r.get("Source_AsOf")) for r in sector_rows]
+        age_vals = [a for a in ages if a is not None]
+        if age_vals and max(age_vals) > MAX_AGE_DAYS:
+            stale_rows += sum(1 for a in age_vals if a is not None and a > MAX_AGE_DAYS)
+            sector_status_rows.append(f"{sector}:STALE(max={max(age_vals)})")
+
+    gate = "PASS"
+    if missing_headers:
+        gate = "FAIL"
+    elif template_rows > 0 or fail_rows > 0 or stale_rows > 0 or mixed_sector_count > 0:
+        gate = "FAIL"
+    elif sheet_input_rows > 0:
+        gate = "WARN"
+
+    print(f"[sector_universe_refresh] gate={gate}")
+    print(f"  rows={len(rows)} sectors={len(sector_map)}")
+    print(f"  naver_rows={naver_rows} representative_rows={representative_rows} layout_changed_rows={layout_changed_rows} sheet_input_rows={sheet_input_rows} template_rows={template_rows} fail_rows={fail_rows}")
+    print(f"  missing_source_url={missing_source_url} stale_rows={stale_rows} mixed_sector_count={mixed_sector_count}")
+    if missing_headers:
+        print(f"  missing_headers={missing_headers}")
+    if sector_status_rows:
+        print("  sector_flags=" + ", ".join(sector_status_rows[:20]))
+
+    result = {
+        "validator": "validate_sector_universe_monthly_refresh_v1",
+        "gate": gate,
+        "total_rows": len(rows),
+        "sector_count": len(sector_map),
+        "naver_rows": naver_rows,
+        "representative_rows": representative_rows,
+        "layout_changed_rows": layout_changed_rows,
+        "sheet_input_rows": sheet_input_rows,
+        "template_rows": template_rows,
+        "fail_rows": fail_rows,
+        "missing_source_url": missing_source_url,
+        "stale_rows": stale_rows,
+        "mixed_sector_count": mixed_sector_count,
+        "missing_headers": missing_headers,
+        "sector_flags": sector_status_rows,
+        "max_age_days": MAX_AGE_DAYS,
+    }
+    out = ROOT / "Temp" / "sector_universe_refresh_validation.json"
+    out.write_text(json.dumps(result, ensure_ascii=False, indent=2), encoding="utf-8")
+    print(f"OUTPUT: {out}")
+    return 0 if gate in {"PASS", "WARN"} else 1
+
+
+if __name__ == "__main__":
+    sys.exit(main())