@canonry/canonry 7.71.0 → 7.72.0

package/dist/{chunk-XSRKTGRG.js → chunk-VH2KHJGT.js}

@@ -1,7 +1,7 @@
 import {
   answerServiceAdapters,
   answerServiceDescriptors
-} from "./chunk-QREFWO3Q.js";
+} from "./chunk-XR6ZVMYK.js";
 import {
   AeroLlmUsageFeatures,
   aeroTurnStatus,
@@ -27,14 +27,14 @@ import {
   resolveConfiguredAgentProvider,
   resolveSessionProviderAndModel,
   setAeroSystemPrompt
-} from "./chunk-C3ZH2HLV.js";
+} from "./chunk-7OJQWLDI.js";
 import {
   AeroToolProfiles,
   AeroToolScopes,
   buildAeroStateTools,
   isAeroToolProfile,
   truncateToolResult
-} from "./chunk-OOXR7SZ5.js";
+} from "./chunk-H6IQ4YCJ.js";
 import {
   getGoogleAuthConfig,
   getGoogleConnection,
@@ -53,7 +53,7 @@ import {
   removeCloudflareTrafficConnection,
   removeCloudflareTrafficConnectionBySourceId,
   upsertCloudflareTrafficConnection
-} from "./chunk-JB4NDUWJ.js";
+} from "./chunk-BZZDFIY2.js";
 import {
   claudeAdapter,
   embedQueries,
@@ -64,7 +64,7 @@ import {
   openaiAdapter,
   openrouterAdapter,
   perplexityAdapter
-} from "./chunk-K4S2T6C5.js";
+} from "./chunk-JO73L677.js";
 import {
   classifyUsageSurface,
   createApiUsageTelemetry,
@@ -76,7 +76,7 @@ import {
   startOutcomeTimer,
   trackFeatureCompleted,
   trackInstallState
-} from "./chunk-JCECURGF.js";
+} from "./chunk-4LTL5RNY.js";
 import {
   isLoopbackBindHost
 } from "./chunk-TSENU4GG.js";
@@ -84,7 +84,7 @@ import {
   isApiProviderRegistrable,
   isCdpProviderRegistrable,
   registeredProviderNames
-} from "./chunk-ALXQPMYR.js";
+} from "./chunk-6MYHGP46.js";
 import {
   checkLatestVersionForServer,
   getOrCreateAnonymousId,
@@ -95,17 +95,17 @@ import {
   recordDashboardEvent,
   setTelemetryPreference,
   trackEvent
-} from "./chunk-DEMNY4YE.js";
+} from "./chunk-MJYO6UK2.js";
 import {
   createCanonryMcpServer
-} from "./chunk-ILLWCY2P.js";
+} from "./chunk-IV6VZ63E.js";
 import {
   CANONRY_MCP_TIERS,
   CANONRY_MCP_TOOLKIT_NAMES
 } from "./chunk-NDTU7FGL.js";
 import {
   ApiClient
-} from "./chunk-T6IV35KG.js";
+} from "./chunk-ATT3ZCSS.js";
 import {
   AGENT_PROVIDERS,
   adapterSupportsBatch,
@@ -136,21 +136,21 @@ import {
   saveConfigPatch,
   sentimentInstallReadiness,
   updateSharedProviderExecutionGate
-} from "./chunk-3NOFNFGV.js";
+} from "./chunk-JE5QURTR.js";
 import {
   getBundledSkillSnapshots
-} from "./chunk-NPHBHVAY.js";
+} from "./chunk-PPM6Z7CM.js";
 import {
   computeFactorAverages,
   deleteSiteCrawlGraphLayout,
   persistSiteCrawlGraphLayout,
   prepareSiteCrawlGraphLayout,
   recordedSharePct
-} from "./chunk-FNVUPIMW.js";
+} from "./chunk-IZMT4WLZ.js";
 import {
   CanonryMcpToolNames,
   canonryMcpTools
-} from "./chunk-WE3SK5GQ.js";
+} from "./chunk-GA2EHZTI.js";
 import {
   PACKAGE_VERSION
 } from "./chunk-AOJU5LKX.js";
@@ -162,11 +162,11 @@ import {
   addLogListener,
   createFastifyLogger,
   createLogger
-} from "./chunk-JEC5Z22M.js";
+} from "./chunk-L6WMECR5.js";
 import {
   computeCitedCompetitorDomains,
   determineCitationState
-} from "./chunk-ETMB45WP.js";
+} from "./chunk-DRGCVTYH.js";
 import {
   BingApiError,
   CC_CACHE_DIR,
@@ -286,7 +286,7 @@ import {
   webhookResponseReason,
   webhookTargetRefusalReason,
   writeAuditLog
-} from "./chunk-SZSJALQM.js";
+} from "./chunk-VJFQNFOR.js";
 import {
   OperationalLogStore,
   SentimentIdempotencyConflict,
@@ -372,7 +372,7 @@ import {
   siteLivenessState,
   trafficSources,
   usageCounters
-} from "./chunk-EDQ4V4PJ.js";
+} from "./chunk-O7WOCTBD.js";
 import {
   AGENT_MEMORY_VALUE_MAX_BYTES,
   AI_ENGINE_DOMAINS,
@@ -624,7 +624,7 @@ import {
   winnabilityClassLabel,
   withRetry,
   withSkippedProviders
-} from "./chunk-A3NSYJJL.js";
+} from "./chunk-YXAKL5HM.js";
 
 // src/google-marketing-runtime.ts
 import crypto2 from "crypto";
@@ -5807,7 +5807,7 @@ function nonBlank(value) {
   return trimmed ? trimmed : void 0;
 }
 function resolveBuildCommit(env = process.env) {
-  const embedded = true ? "9caa25b0ba175e5a81276cef6fe4c541f3f40073" : void 0;
+  const embedded = true ? "5316fa88fb0c94357b949d9442ad4ce1b8307880" : void 0;
   return nonBlank(embedded) ?? nonBlank(env.CANONRY_COMMIT);
 }
 function resolveInstanceIdentity(env = process.env) {
@@ -11884,6 +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)(?:$|\?)/i;
 var NON_PAGE_MEDIA_CONTENT_TYPE = /^(?:image|video|audio|font)\//i;
 function isNonPageMedia(page) {
@@ -12317,6 +12318,8 @@ async function executeSiteAudit(db, runId, projectId, opts = {}) {
           pagesFetched: progress.pagesFetched,
           pagesEligible: event.type === "summary" ? observedPages.size === 0 ? 0 : [...observedPages.values()].filter(isInventoryEligible).length : sql6`${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();
@@ -12361,6 +12364,7 @@ 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
     });
@@ -12560,7 +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,

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

@@ -9,14 +9,14 @@ import {
   normalizeCloudflareWorkerEvent,
   sessionizeJourneyRequests,
   verifyRequestSignature
-} from "./chunk-WNBN4FYE.js";
+} from "./chunk-JEBY6MS6.js";
 import {
   currentOutcomeAttribution,
   getRequestContext,
   registerRequestContext,
   requestAuditIdentity,
   sanitizeRequestContext
-} from "./chunk-U6EUBP4B.js";
+} from "./chunk-BADJCT6U.js";
 import {
   SentimentIdempotencyConflict,
   SentimentRepository,
@@ -156,7 +156,7 @@ import {
   userSessions,
   users,
   writeNegativeReviewMaxStars
-} from "./chunk-EDQ4V4PJ.js";
+} from "./chunk-O7WOCTBD.js";
 import {
   ADS_ACTIVATE_SCOPE,
   ADS_ACTIVATION_ABSOLUTE_MAX_ENTITIES,
@@ -1194,7 +1194,7 @@ import {
   wordpressSchemaDeployResultDtoSchema,
   wordpressSchemaStatusResultDtoSchema,
   wordpressStatusDtoSchema
-} from "./chunk-A3NSYJJL.js";
+} from "./chunk-YXAKL5HM.js";
 
 // ../api-routes/src/user-session.ts
 import crypto2 from "crypto";
@@ -42312,7 +42312,7 @@ var routeCatalog = [
       { 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. 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`. Scans made before URL versions were set aside have none.", schema: { type: "string", enum: ["include", "exclude", "only"] } },
+      { 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
@@ -72580,13 +72580,19 @@ async function technicalAeoRoutes(app, opts) {
     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)),
-      urlVersions
+      urlVersions,
+      urlVersionsNotFetched
     };
   };
   const completenessOf = (target) => ({

package/dist/{chunk-QREFWO3Q.js → chunk-XR6ZVMYK.js}

@@ -28,7 +28,7 @@ import {
   unknownUsageObservation,
   usdToMicros,
   withRetry
-} from "./chunk-A3NSYJJL.js";
+} from "./chunk-YXAKL5HM.js";
 
 // ../provider-answer-services/src/core/http.ts
 var MAX_RESPONSE_BYTES = 8 * 1024 * 1024;

package/dist/{chunk-A3NSYJJL.js → chunk-YXAKL5HM.js}

@@ -7911,7 +7911,15 @@ var siteCrawlInventorySummarySchema = z20.object({
    * 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)
+  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. */