@canonry/canonry 7.70.0 → 7.72.0

package/dist/{chunk-OQR4F7WF.js → chunk-UWBJWKDQ.js}

@@ -1,6 +1,6 @@
 import {
   createApiClient
-} from "./chunk-BGTJTKOJ.js";
+} from "./chunk-ATT3ZCSS.js";
 import {
   emitJsonl
 } from "./chunk-Z7IUAJOF.js";
@@ -19,7 +19,7 @@ import {
   formatMicros,
   formatRunErrorOneLine,
   resolveProviderInput
-} from "./chunk-6JGQT4OH.js";
+} from "./chunk-YXAKL5HM.js";
 
 // src/commands/run.ts
 function getClient() {

package/dist/{chunk-3QHKN3JY.js → chunk-VH2KHJGT.js}

@@ -1,7 +1,7 @@
 import {
   answerServiceAdapters,
   answerServiceDescriptors
-} from "./chunk-SFMCFMJR.js";
+} from "./chunk-XR6ZVMYK.js";
 import {
   AeroLlmUsageFeatures,
   aeroTurnStatus,
@@ -27,14 +27,14 @@ import {
   resolveConfiguredAgentProvider,
   resolveSessionProviderAndModel,
   setAeroSystemPrompt
-} from "./chunk-BRT5GE3Z.js";
+} from "./chunk-7OJQWLDI.js";
 import {
   AeroToolProfiles,
   AeroToolScopes,
   buildAeroStateTools,
   isAeroToolProfile,
   truncateToolResult
-} from "./chunk-DW3KF33I.js";
+} from "./chunk-H6IQ4YCJ.js";
 import {
   getGoogleAuthConfig,
   getGoogleConnection,
@@ -53,7 +53,7 @@ import {
   removeCloudflareTrafficConnection,
   removeCloudflareTrafficConnectionBySourceId,
   upsertCloudflareTrafficConnection
-} from "./chunk-RPYSCFFA.js";
+} from "./chunk-BZZDFIY2.js";
 import {
   claudeAdapter,
   embedQueries,
@@ -64,7 +64,7 @@ import {
   openaiAdapter,
   openrouterAdapter,
   perplexityAdapter
-} from "./chunk-EIRIZPR6.js";
+} from "./chunk-JO73L677.js";
 import {
   classifyUsageSurface,
   createApiUsageTelemetry,
@@ -76,7 +76,7 @@ import {
   startOutcomeTimer,
   trackFeatureCompleted,
   trackInstallState
-} from "./chunk-3LTLLHCD.js";
+} from "./chunk-4LTL5RNY.js";
 import {
   isLoopbackBindHost
 } from "./chunk-TSENU4GG.js";
@@ -84,7 +84,7 @@ import {
   isApiProviderRegistrable,
   isCdpProviderRegistrable,
   registeredProviderNames
-} from "./chunk-3ZF3T3OG.js";
+} from "./chunk-6MYHGP46.js";
 import {
   checkLatestVersionForServer,
   getOrCreateAnonymousId,
@@ -95,17 +95,17 @@ import {
   recordDashboardEvent,
   setTelemetryPreference,
   trackEvent
-} from "./chunk-RVEXUQHP.js";
+} from "./chunk-MJYO6UK2.js";
 import {
   createCanonryMcpServer
-} from "./chunk-PHHQVV3G.js";
+} from "./chunk-IV6VZ63E.js";
 import {
   CANONRY_MCP_TIERS,
   CANONRY_MCP_TOOLKIT_NAMES
 } from "./chunk-NDTU7FGL.js";
 import {
   ApiClient
-} from "./chunk-BGTJTKOJ.js";
+} from "./chunk-ATT3ZCSS.js";
 import {
   AGENT_PROVIDERS,
   adapterSupportsBatch,
@@ -136,21 +136,21 @@ import {
   saveConfigPatch,
   sentimentInstallReadiness,
   updateSharedProviderExecutionGate
-} from "./chunk-6RI7JFFI.js";
+} from "./chunk-JE5QURTR.js";
 import {
   getBundledSkillSnapshots
-} from "./chunk-JNRMODJS.js";
+} from "./chunk-PPM6Z7CM.js";
 import {
   computeFactorAverages,
   deleteSiteCrawlGraphLayout,
   persistSiteCrawlGraphLayout,
   prepareSiteCrawlGraphLayout,
   recordedSharePct
-} from "./chunk-7W4KNDAW.js";
+} from "./chunk-IZMT4WLZ.js";
 import {
   CanonryMcpToolNames,
   canonryMcpTools
-} from "./chunk-QNSMMQJY.js";
+} from "./chunk-GA2EHZTI.js";
 import {
   PACKAGE_VERSION
 } from "./chunk-AOJU5LKX.js";
@@ -162,11 +162,11 @@ import {
   addLogListener,
   createFastifyLogger,
   createLogger
-} from "./chunk-42JQQZZQ.js";
+} from "./chunk-L6WMECR5.js";
 import {
   computeCitedCompetitorDomains,
   determineCitationState
-} from "./chunk-R7RTWOOV.js";
+} from "./chunk-DRGCVTYH.js";
 import {
   BingApiError,
   CC_CACHE_DIR,
@@ -286,7 +286,7 @@ import {
   webhookResponseReason,
   webhookTargetRefusalReason,
   writeAuditLog
-} from "./chunk-JGO7LGVN.js";
+} from "./chunk-VJFQNFOR.js";
 import {
   OperationalLogStore,
   SentimentIdempotencyConflict,
@@ -372,7 +372,7 @@ import {
   siteLivenessState,
   trafficSources,
   usageCounters
-} from "./chunk-XOCF2U54.js";
+} from "./chunk-O7WOCTBD.js";
 import {
   AGENT_MEMORY_VALUE_MAX_BYTES,
   AI_ENGINE_DOMAINS,
@@ -446,6 +446,7 @@ import {
   SITE_AUDIT_DEFAULT_PAGE_LIMIT,
   SITE_AUDIT_MAX_EDGE_LIMIT,
   SITE_AUDIT_MAX_PAGE_LIMIT,
+  SITE_CRAWL_URL_VERSION_AUDIT_STATE,
   STORED_ONLY_HEADER,
   SchedulableRunKinds,
   SentimentOutcomes,
@@ -606,6 +607,7 @@ import {
   serializeRunError,
   shiftIsoCalendarDate,
   siteAuditRequestIdentity,
+  siteCrawlUrlVersionOf,
   splitList,
   startOfDayHourInTimeZone,
   statusClassOf,
@@ -622,7 +624,7 @@ import {
   winnabilityClassLabel,
   withRetry,
   withSkippedProviders
-} from "./chunk-6JGQT4OH.js";
+} from "./chunk-YXAKL5HM.js";
 
 // src/google-marketing-runtime.ts
 import crypto2 from "crypto";
@@ -5805,7 +5807,7 @@ function nonBlank(value) {
   return trimmed ? trimmed : void 0;
 }
 function resolveBuildCommit(env = process.env) {
-  const embedded = true ? "248bb6e43737a9fdd7b9fc5f05a1e150c58843a4" : void 0;
+  const embedded = true ? "5316fa88fb0c94357b949d9442ad4ce1b8307880" : void 0;
   return nonBlank(embedded) ?? nonBlank(env.CANONRY_COMMIT);
 }
 function resolveInstanceIdentity(env = process.env) {
@@ -11882,6 +11884,7 @@ var log14 = createLogger("SiteAudit");
 var SiteAuditCancelledError = class extends Error {
   name = "SiteAuditCancelledError";
 };
+var SITE_AUDIT_SKIP_URL_VERSIONS_AFTER = 3;
 var NON_PAGE_MEDIA_EXTENSIONS = /\.(?:jpe?g|png|gif|webp|avif|bmp|ico|tiff?|svg|mp4|webm|mov|avi|mp3|wav|ogg|woff2?|ttf|otf|eot)(?:$|\?)/i;
 var NON_PAGE_MEDIA_CONTENT_TYPE = /^(?:image|video|audio|font)\//i;
 function isNonPageMedia(page) {
@@ -11948,7 +11951,14 @@ function pageDiscoverySource(page) {
   if (page.provenance.sitemapSources.length > 0) return "sitemap";
   return "link";
 }
+function urlVersionOf(page) {
+  return siteCrawlUrlVersionOf({ fetchState: page.state, url: page.requestedUrl, canonicalUrl: page.canonicalUrl });
+}
+function isScoredPage(page) {
+  return page.audit != null && urlVersionOf(page) === null;
+}
 function pageAuditState(page) {
+  if (urlVersionOf(page) !== null) return SITE_CRAWL_URL_VERSION_AUDIT_STATE;
   if (page.audit) return "success";
   if (page.state === "fetch-error") return "error";
   return "not-applicable";
@@ -11962,7 +11972,7 @@ function isCancelled(error, signal) {
   return error instanceof DOMException && error.name === "AbortError";
 }
 function isLegacyScorecardPage(page) {
-  return page.audit != null || page.state === "fetch-error";
+  return isScoredPage(page) || page.state === "fetch-error";
 }
 function emptyCompleteCrawlError(rootUrl, finalRootUrl, pagesObserved) {
   if (!finalRootUrl) return new Error(`Site audit could not successfully audit any of ${pagesObserved} observed page(s).`);
@@ -12110,7 +12120,8 @@ async function executeSiteAudit(db, runId, projectId, opts = {}) {
     }
   };
   const persistPage = (tx, page, now) => {
-    const audit = page.audit;
+    const versionOf = urlVersionOf(page);
+    const audit = versionOf === null ? page.audit : null;
     const canonicalNodeKey = page.canonicalUrl ? nodeKeyByUrl.get(page.canonicalUrl) ?? null : null;
     const healthState = deriveSiteHealthState({
       fetchState: page.state,
@@ -12147,6 +12158,7 @@ async function executeSiteAudit(db, runId, projectId, opts = {}) {
       indexabilityState: page.indexability.state,
       indexabilityReasons: page.indexability.reasons,
       healthState,
+      urlVersionOf: versionOf,
       auditState: pageAuditState(page),
       auditScore: audit?.overallScore ?? null,
       auditFields: audit ? crawlAuditFields(audit) : {},
@@ -12181,6 +12193,7 @@ async function executeSiteAudit(db, runId, projectId, opts = {}) {
         indexabilityState: page.indexability.state,
         indexabilityReasons: page.indexability.reasons,
         healthState,
+        urlVersionOf: versionOf,
         auditState: pageAuditState(page),
         auditScore: audit?.overallScore ?? null,
         auditFields: audit ? crawlAuditFields(audit) : {},
@@ -12305,6 +12318,8 @@ async function executeSiteAudit(db, runId, projectId, opts = {}) {
           pagesFetched: progress.pagesFetched,
           pagesEligible: event.type === "summary" ? observedPages.size === 0 ? 0 : [...observedPages.values()].filter(isInventoryEligible).length : sql6`${siteCrawlAttempts.pagesEligible}`,
           pagesErrored: event.type === "summary" ? observedErrorCount(observedPages.values()) : sql6`${siteCrawlAttempts.pagesErrored}`,
+          // Only the summary knows how many URL versions the crawl left unfetched.
+          urlVersionsSkipped: event.type === "summary" ? event.summary.urlVersions?.skipped ?? 0 : sql6`${siteCrawlAttempts.urlVersionsSkipped}`,
           edgesDiscovered: progress.edgesObserved,
           updatedAt: now
         }).where(eq18(siteCrawlAttempts.id, attemptId)).run();
@@ -12349,12 +12364,12 @@ async function executeSiteAudit(db, runId, projectId, opts = {}) {
       checkDeadLinks: opts.checkDeadLinks ?? false,
       includeGeo: opts.includeGeo ?? false,
       includeAgentSkills: opts.includeAgentSkills ?? true,
+      skipUrlVersionsAfter: SITE_AUDIT_SKIP_URL_VERSIONS_AFTER,
       signal: opts.signal,
       onEvent: persistEvent
     });
     const crawlSummary = report.summary;
-    const hasAuditedPages = crawlSummary.auditRollup.auditedPages > 0;
-    if (crawlSummary.complete && !hasAuditedPages) {
+    if (crawlSummary.complete && crawlSummary.auditRollup.auditedPages === 0) {
       throw emptyCompleteCrawlError(crawlSummary.rootUrl, crawlSummary.finalRootUrl, crawlSummary.pagesObserved);
     }
     const placementRulesetVersion = crawlSummary.linkPlacementRulesetVersion ?? null;
@@ -12379,7 +12394,11 @@ async function executeSiteAudit(db, runId, projectId, opts = {}) {
       rootUrl: crawlSummary.finalRootUrl ?? crawlSummary.rootUrl
     }, { signal: opts.signal, timeoutMs: opts.graphLayoutTimeoutMs });
     const finishedAt = (/* @__PURE__ */ new Date()).toISOString();
-    const factors = computeFactorAverages([...observedPages.values()]);
+    const pageObservations = [...observedPages.values()].filter((page) => urlVersionOf(page) === null);
+    const scoredPages = pageObservations.filter(isScoredPage);
+    const pagesAudited = scoredPages.length;
+    const aggregateScore = pagesAudited > 0 ? Math.round(scoredPages.reduce((total, page) => total + page.audit.overallScore, 0) / pagesAudited) : null;
+    const factors = computeFactorAverages(pageObservations);
     const errorCount = observedErrorCount(observedPages.values());
     const terminalStatus = crawlSummary.complete ? "completed" : "partial";
     const deadLinksChecked = opts.checkDeadLinks ? deadLinkCheckedCount(observedEdges.values(), observedPages.values()) : 0;
@@ -12387,8 +12406,8 @@ async function executeSiteAudit(db, runId, projectId, opts = {}) {
     const unverifiedLinks = report.deadLinks.unverified;
     const deadLinksFound = deadLinkFindings.length;
     const deadLinksUnverified = new Set(unverifiedLinks.map((finding) => finding.to)).size;
-    const legacyIssues = factors.map((factor) => toLegacyIssue(factor, crawlSummary.auditRollup.auditedPages)).filter((issue) => issue !== null);
-    const issueEvidence = buildSiteCrawlIssueEvidence(crawlSummary.siteIssues, observedPages.values());
+    const legacyIssues = factors.map((factor) => toLegacyIssue(factor, pagesAudited)).filter((issue) => issue !== null);
+    const issueEvidence = buildSiteCrawlIssueEvidence(crawlSummary.siteIssues, pageObservations);
     try {
       persistSiteCrawlGraphLayout(db, { projectId, runId, attemptId }, graphLayout, finishedAt);
     } catch (error) {
@@ -12420,16 +12439,16 @@ async function executeSiteAudit(db, runId, projectId, opts = {}) {
     const published = db.transaction((tx) => {
       const claim2 = tx.update(runs).set({ status: terminalStatus, finishedAt }).where(and14(eq18(runs.id, runId), eq18(runs.status, "running"))).run();
       if (claim2.changes === 0) return false;
-      if (hasAuditedPages) {
+      if (aggregateScore !== null) {
         tx.insert(siteAuditSnapshots).values({
           id: crypto19.randomUUID(),
           projectId,
           runId,
           sitemapUrl: opts.sitemapUrl ?? new URL("/sitemap.xml", crawlRootUrl).href,
           auditedAt: crawlSummary.completedAt,
-          aggregateScore: crawlSummary.auditRollup.aggregateScore ?? 0,
+          aggregateScore,
           pagesDiscovered: crawlSummary.pagesDiscovered,
-          pagesAudited: crawlSummary.auditRollup.auditedPages,
+          pagesAudited,
           pagesSkipped: Math.max(0, crawlSummary.pagesDiscovered - crawlSummary.pagesObserved),
           pagesErrored: errorCount,
           factorAverages: factors,
@@ -12545,7 +12564,20 @@ async function executeSiteAudit(db, runId, projectId, opts = {}) {
       if (current?.status === "cancelled") throw new SiteAuditCancelledError(`Site audit cancelled before publication: ${runId}`);
       throw new Error(`Site audit run was no longer running before publication: ${runId}`);
     }
-    log14.info("completed", { runId, projectId, status: terminalStatus, complete: crawlSummary.complete, pages: crawlSummary.pagesObserved });
+    log14.info("completed", {
+      runId,
+      projectId,
+      status: terminalStatus,
+      complete: crawlSummary.complete,
+      pages: crawlSummary.pagesObserved,
+      urlVersionsSkipped: crawlSummary.urlVersions?.skipped ?? 0,
+      urlVersionPatterns: (crawlSummary.urlVersions?.patterns ?? []).map((pattern) => ({
+        section: pattern.section,
+        parameters: pattern.parameters,
+        copies: pattern.copies,
+        skipped: pattern.skipped
+      }))
+    });
     trackSiteAuditOutcome(db, {
       runId,
       projectId,
@@ -12558,9 +12590,9 @@ async function executeSiteAudit(db, runId, projectId, opts = {}) {
         termination: crawlSummary.terminationReason,
         pagesDiscovered: crawlSummary.pagesDiscovered,
         pagesFetched: crawlSummary.pagesFetched,
-        pagesAudited: crawlSummary.auditRollup.auditedPages,
+        pagesAudited,
         pagesErrored: errorCount,
-        aggregateScore: crawlSummary.auditRollup.aggregateScore,
+        aggregateScore,
         pageBudget: maxPages,
         checkDeadLinks: opts.checkDeadLinks ?? false,
         deadLinksFound

package/dist/{chunk-JGO7LGVN.js → chunk-VJFQNFOR.js}

@@ -9,14 +9,14 @@ import {
   normalizeCloudflareWorkerEvent,
   sessionizeJourneyRequests,
   verifyRequestSignature
-} from "./chunk-CNSLBHFP.js";
+} from "./chunk-JEBY6MS6.js";
 import {
   currentOutcomeAttribution,
   getRequestContext,
   registerRequestContext,
   requestAuditIdentity,
   sanitizeRequestContext
-} from "./chunk-KDV5NQZH.js";
+} from "./chunk-BADJCT6U.js";
 import {
   SentimentIdempotencyConflict,
   SentimentRepository,
@@ -156,7 +156,7 @@ import {
   userSessions,
   users,
   writeNegativeReviewMaxStars
-} from "./chunk-XOCF2U54.js";
+} from "./chunk-O7WOCTBD.js";
 import {
   ADS_ACTIVATE_SCOPE,
   ADS_ACTIVATION_ABSOLUTE_MAX_ENTITIES,
@@ -378,6 +378,7 @@ import {
   SiteAuditTrendDirections,
   SiteCrawlFetchedStates,
   SiteCrawlIndexabilityReasons,
+  SiteCrawlUrlVersionFilters,
   SiteHealthLinkKinds,
   SiteHealthTemplateDetections,
   TrafficEventConfidences,
@@ -1111,6 +1112,7 @@ import {
   siteCrawlPagesResponseSchema,
   siteCrawlStructureResponseSchema,
   siteCrawlSummarySchema,
+  siteCrawlUrlVersionFilterSchema,
   siteHealthChangesResponseSchema,
   siteHealthLinkKindSchema,
   siteHealthPathResponseSchema,
@@ -1192,7 +1194,7 @@ import {
   wordpressSchemaDeployResultDtoSchema,
   wordpressSchemaStatusResultDtoSchema,
   wordpressStatusDtoSchema
-} from "./chunk-6JGQT4OH.js";
+} from "./chunk-YXAKL5HM.js";
 
 // ../api-routes/src/user-session.ts
 import crypto2 from "crypto";
@@ -42309,13 +42311,14 @@ var routeCatalog = [
       { name: "indexabilityState", in: "query", description: "Filter by crawler-derived indexability state; this is not Google index coverage.", schema: stringSchema },
       { name: "nodeKey", in: "query", description: "Return only the page with this node key. Combined with a filter it answers whether that exact page is in the filtered set, without paging through the list.", schema: stringSchema },
       { name: "healthState", in: "query", description: "Filter by the derived Site Health state shared with the dashboard and agents. Unlike `indexabilityState` this folds fetch state, canonical identity, and the crawler reasons into one decision. `hidden` means the SITE told answer engines not to index the page (noindex, canonical-away, robots.txt); a non-HTML file is `resource` and a redirect is `redirect`, because neither is suppressed. Values: `eligible`, `hidden`, `resource`, `redirect`, `failed`, `unchecked`.", schema: { type: "string", enum: ["eligible", "hidden", "resource", "redirect", "failed", "unchecked"] } },
-      { name: "auditState", in: "query", description: "Filter by audit state.", schema: stringSchema },
+      { name: "auditState", in: "query", description: "Filter by audit state. A URL version has `url-version`: it was fetched and classified but not scored.", schema: stringSchema },
+      { name: "urlVersions", in: "query", description: "URL versions are parameter addresses (filters, views, cache settings) whose canonical names the same path, so the site marks them as copies of that page. The audit does not score them or count them in fixes, and `inventorySummary.urlVersions` counts them. `include` (default) keeps every crawled address, `exclude` lists pages only, as the dashboard does, and `only` lists the URL versions, each with `urlVersionOf`. Once a few addresses with the same parameter pattern prove to be copies, the crawl stops fetching that pattern: those have no rows and `inventorySummary.urlVersionsNotFetched` counts them. Scans made before URL versions were set aside have none.", schema: { type: "string", enum: ["include", "exclude", "only"] } },
       { name: "sort", in: "query", description: "Sort order. Defaults to `url`.", schema: { type: "string", enum: ["url", "path", "score-asc", "score-desc"] } },
       crawlCursorParameter,
       crawlLimitParameter
     ],
     responses: {
-      400: errorResponse("Invalid date, conflicting runId/date, or invalid cursor identity."),
+      400: errorResponse("Invalid date, conflicting runId/date, invalid urlVersions, or invalid cursor identity."),
       200: jsonResponse("Crawl pages returned.", "SiteCrawlPagesResponseDto"),
       404: errorResponse("Project or site-audit run not found. A known run that published no crawl returns 200 with the no-crawl state instead.")
     }
@@ -72106,6 +72109,14 @@ function parseSiteHealthState(value) {
   }
   return allowed.data;
 }
+function parseUrlVersionFilter(value) {
+  if (value === void 0 || value === "") return SiteCrawlUrlVersionFilters.include;
+  const allowed = siteCrawlUrlVersionFilterSchema.safeParse(value);
+  if (!allowed.success) {
+    throw validationError(`"urlVersions" must be one of: ${siteCrawlUrlVersionFilterSchema.options.join(", ")}`);
+  }
+  return allowed.data;
+}
 function parseBoolean2(value) {
   if (value === true || value === "true" || value === "1") return true;
   if (value === false || value === "false" || value === "0") return false;
@@ -72188,7 +72199,8 @@ function mapCrawlPage(row) {
     linkScoreRaw: row.linkScoreRaw,
     linkScoreNormalized: row.linkScoreNormalized,
     healthState: deriveSiteHealthState(row),
-    healthReason: deriveSiteHealthReason(row)
+    healthReason: deriveSiteHealthReason(row),
+    urlVersionOf: row.urlVersionOf ?? null
   };
 }
 function withRecordedShare(factor) {
@@ -72559,19 +72571,28 @@ async function technicalAeoRoutes(app, opts) {
   };
   const inventorySummaryFor = (scope) => {
     if (!scope) return null;
-    const groups = crawlReasonGroups(and83(
+    const inScope = and83(
       eq100(siteCrawlPages.projectId, scope.projectId),
       eq100(siteCrawlPages.runId, scope.runId),
       eq100(siteCrawlPages.attemptId, scope.attemptId)
-    ));
+    );
+    const groups = crawlReasonGroups(inScope);
     const total = groups.reduce((sum, group) => sum + group.pages, 0);
     const eligible = groups.reduce((sum, group) => sum + (group.inventoryEligible ? group.pages : 0), 0);
+    const urlVersions = app.db.select({ value: count8() }).from(siteCrawlPages).where(and83(inScope, isNotNull10(siteCrawlPages.urlVersionOf))).get()?.value ?? 0;
+    const urlVersionsNotFetched = app.db.select({ value: siteCrawlAttempts.urlVersionsSkipped }).from(siteCrawlAttempts).where(and83(
+      eq100(siteCrawlAttempts.projectId, scope.projectId),
+      eq100(siteCrawlAttempts.runId, scope.runId),
+      eq100(siteCrawlAttempts.id, scope.attemptId)
+    )).get()?.value ?? 0;
     return {
       scope: "selected-snapshot",
       total,
       eligible,
       excluded: total - eligible,
-      excludedReasons: reasonCounts(groups.filter((group) => !group.inventoryEligible))
+      excludedReasons: reasonCounts(groups.filter((group) => !group.inventoryEligible)),
+      urlVersions,
+      urlVersionsNotFetched
     };
   };
   const completenessOf = (target) => ({
@@ -73739,6 +73760,9 @@ async function technicalAeoRoutes(app, opts) {
     if (request.query.fetchState) filters.push(eq100(siteCrawlPages.fetchState, request.query.fetchState));
     if (request.query.indexabilityState) filters.push(eq100(siteCrawlPages.indexabilityState, request.query.indexabilityState));
     if (request.query.auditState) filters.push(eq100(siteCrawlPages.auditState, request.query.auditState));
+    const urlVersions = parseUrlVersionFilter(request.query.urlVersions);
+    if (urlVersions === SiteCrawlUrlVersionFilters.exclude) filters.push(isNull17(siteCrawlPages.urlVersionOf));
+    if (urlVersions === SiteCrawlUrlVersionFilters.only) filters.push(isNotNull10(siteCrawlPages.urlVersionOf));
     const healthState = parseSiteHealthState(request.query.healthState);
     const limit = parseBoundedLimit(request.query.limit, 100, 200);
     const sort = ["score-desc", "score-asc", "path"].includes(request.query.sort ?? "") ? request.query.sort : "url";
@@ -73752,6 +73776,8 @@ async function technicalAeoRoutes(app, opts) {
       fetchState: request.query.fetchState || null,
       indexabilityState: request.query.indexabilityState || null,
       auditState: request.query.auditState || null,
+      // Absent for the default, so a cursor issued before this filter existed still matches.
+      ...urlVersions === SiteCrawlUrlVersionFilters.include ? {} : { urlVersions },
       healthState,
       sort
     })).digest("base64url");

package/dist/{chunk-SFMCFMJR.js → chunk-XR6ZVMYK.js}

@@ -28,7 +28,7 @@ import {
   unknownUsageObservation,
   usdToMicros,
   withRetry
-} from "./chunk-6JGQT4OH.js";
+} from "./chunk-YXAKL5HM.js";
 
 // ../provider-answer-services/src/core/http.ts
 var MAX_RESPONSE_BYTES = 8 * 1024 * 1024;

package/dist/{chunk-6JGQT4OH.js → chunk-YXAKL5HM.js}

@@ -7905,7 +7905,21 @@ var siteCrawlInventorySummarySchema = z20.object({
   eligible: z20.number().int().nonnegative(),
   excluded: z20.number().int().nonnegative(),
   /** Disjoint primary reasons: their page counts sum to excluded. */
-  excludedReasons: z20.array(siteCrawlHealthReasonCountSchema).max(siteHealthReasonSchema.options.length)
+  excludedReasons: z20.array(siteCrawlHealthReasonCountSchema).max(siteHealthReasonSchema.options.length),
+  /**
+   * URL versions the audit set aside (see `siteCrawlUrlVersionOf`): not scored
+   * and not counted in fixes. A subset of `excluded`; 0 for a scan made before
+   * the audit set them aside, which scored every address.
+   */
+  urlVersions: z20.number().int().nonnegative().default(0),
+  /**
+   * URL versions the crawl did not fetch at all: once a few addresses with
+   * the same parameter pattern proved to be copies, it skipped the rest. They
+   * have no page rows, so they are outside `total` and `urlVersions`; the URL
+   * versions a scan found are `urlVersions + urlVersionsNotFetched`. 0 for a
+   * scan that fetched every address.
+   */
+  urlVersionsNotFetched: z20.number().int().nonnegative().default(0)
 });
 var siteCrawlAvailableScanDatesSchema = z20.object({
   /** Most recent distinct visible crawl dates, newest first. Fewer rows than totalDates means the list is bounded. */
@@ -8082,6 +8096,24 @@ function deriveSiteHealthState(input) {
       return "unchecked";
   }
 }
+var siteCrawlUrlVersionFilterSchema = z20.enum(["include", "exclude", "only"]);
+var SiteCrawlUrlVersionFilters = siteCrawlUrlVersionFilterSchema.enum;
+var SITE_CRAWL_URL_VERSION_AUDIT_STATE = "url-version";
+function siteCrawlUrlVersionOf(input) {
+  if (input.fetchState !== SiteCrawlFetchStates.html || !input.canonicalUrl) return null;
+  let own;
+  let canonical;
+  try {
+    own = new URL(input.url);
+    canonical = new URL(input.canonicalUrl);
+  } catch {
+    return null;
+  }
+  if (own.search === "") return null;
+  if (hostOf(own.href) !== hostOf(canonical.href) || own.pathname !== canonical.pathname) return null;
+  if (own.search === canonical.search) return null;
+  return canonical.href;
+}
 var TEMPLATE_LINK_RATIO_THRESHOLD = 0.7;
 var TEMPLATE_LINK_MIN_FETCHED_PAGES = 15;
 var siteCrawlLinkPlacementSchema = z20.enum(["navigation", "content", "unknown"]);
@@ -8251,7 +8283,14 @@ var siteCrawlPageSchema = z20.object({
   linkScoreNormalized: percent().nullable(),
   healthState: siteHealthStateSchema,
   /** Concrete fetch/indexability reason; unknown never implies a canonical points elsewhere. */
-  healthReason: siteHealthReasonSchema.optional()
+  healthReason: siteHealthReasonSchema.optional(),
+  /**
+   * The page this address is a URL version of (its canonical), or null for a
+   * page. A URL version is not scored: its `auditState` is `url-version` and
+   * its `auditScore` null. Always null in a scan made before the audit set URL
+   * versions aside.
+   */
+  urlVersionOf: z20.string().nullable().default(null)
 });
 var siteCrawlAuditFindingSchema = z20.object({
   type: z20.enum(["found", "missing", "info", "timeout", "unreachable"]),
@@ -26709,6 +26748,10 @@ export {
   SiteCrawlIndexabilityReasons,
   deriveSiteHealthReason,
   deriveSiteHealthState,
+  siteCrawlUrlVersionFilterSchema,
+  SiteCrawlUrlVersionFilters,
+  SITE_CRAWL_URL_VERSION_AUDIT_STATE,
+  siteCrawlUrlVersionOf,
   TEMPLATE_LINK_RATIO_THRESHOLD,
   SiteHealthLinkClassificationSources,
   siteHealthTemplateDetectionSchema,