mcp-scraper 0.38.0 → 0.38.2

package/dist/{chunk-P64PZUCA.js → chunk-NKCCGADE.js}

@@ -15,17 +15,17 @@ import {
   browserServiceProfileSaveChanges,
   recordVendorUsage,
   vendorCostUsd
-} from "./chunk-MA5JBAUZ.js";
+} from "./chunk-EGJKUB4Q.js";
 import {
   DEFAULT_MAPS_PROXY_MODE,
   DEFAULT_PROXY_MODE
 } from "./chunk-CB5C3BPB.js";
 import {
   PACKAGE_VERSION
-} from "./chunk-H3M3O457.js";
+} from "./chunk-YODBNTTN.js";
 import {
   MC_PER_CREDIT
-} from "./chunk-PWPUKR5U.js";
+} from "./chunk-AZRPG43B.js";
 import {
   sanitizeVendorName
 } from "./chunk-44HZLHDV.js";
@@ -1948,8 +1948,12 @@ async function formatExtractSite(raw, input, ctx) {
       schemaTypes: schemaTypesOf(p)
     })));
     const preview = pages.slice(0, BULK_PAGE_THRESHOLD).map(pageRow).join("\n");
+    const stoppedEarly = d.stoppedEarly;
     const summaryHeader = [
       `# Site Extract: ${input.url}`,
+      stoppedEarly ? `
+> **Partial crawl.** The crawl stopped before finishing, so this covers ${pages.length} page(s) rather than the whole site. You were only charged for the pages returned. Reason: ${stoppedEarly.message ?? "unknown"}
+` : "",
       durationLine,
       `
 ## Pages (first ${Math.min(BULK_PAGE_THRESHOLD, pages.length)} of ${pages.length})
@@ -4305,8 +4309,9 @@ var GoogleAdsSearchInputSchema = {
   maxResults: z2.number().int().min(1).max(20).default(10).describe("Maximum advertisers to return. Default 10, maximum 20.")
 };
 var GoogleAdsPageIntelInputSchema = {
-  advertiserId: z2.string().optional().describe("Google Ads Transparency advertiser ID (starts with AR...). Use one returned by google_ads_search; do not construct one yourself."),
-  domain: z2.string().optional().describe("A domain (e.g. getviktor.com) whose primary advertiser to inspect when advertiserId is unknown. One of advertiserId or domain is required."),
+  domain: z2.string().optional().describe("A website from google_ads_search.websites (e.g. getviktor.com). Preferred input: people know the website, not the advertiser's registered name. Resolves to whichever advertiser account runs ads for that domain."),
+  advertiserName: z2.string().optional().describe('An advertiser name exactly as returned in google_ads_search.advertisers[].name (e.g. "PPS Plumbing Services"). Use when the user picked an advertiser rather than a website.'),
+  advertiserId: z2.string().optional().describe("Google Ads Transparency advertiser ID (starts with AR...). Only available if you already have one from a prior page-intel result \u2014 google_ads_search does NOT return advertiser IDs. Do not construct one yourself."),
   region: z2.string().length(2).default("US").describe("Two-letter region code for where the ads are shown. Default US."),
   maxAds: z2.number().int().min(1).max(200).default(50).describe("Maximum creatives to inspect and hydrate. Default 50, maximum 200. Prefer 25-50 for focused scans.")
 };
@@ -4761,6 +4766,10 @@ var DiffPageOutputSchema = {
 };
 var ExtractSiteOutputSchema = {
   url: z2.string(),
+  stoppedEarly: z2.object({
+    code: z2.string(),
+    message: z2.string()
+  }).nullish().describe("Present when the crawl stopped before finishing. The returned pages are everything that was captured, and billing covers only those pages - treat the result as partial, not as the whole site."),
   pageCount: z2.number().int().min(0).optional().describe("Absent when background is true \u2014 the crawl has not finished yet."),
   pages: z2.array(z2.object({
     url: z2.string(),
@@ -5140,9 +5149,15 @@ var GoogleAdsSearchOutputSchema = {
   advertisers: z2.array(z2.object({
     advertiserId: NullableString,
     name: NullableString,
+    basedIn: NullableString,
     domain: NullableString,
     approxAdCount: z2.number().int().nullable(),
     detailUrl: NullableString
+  })),
+  websiteCount: z2.number().int().min(0),
+  websites: z2.array(z2.object({
+    domain: z2.string(),
+    detailUrl: z2.string()
   }))
 };
 var GoogleAdsPageIntelOutputSchema = {
@@ -6451,14 +6466,14 @@ function registerPaaExtractorMcpTools(server, executor, options = {}) {
   }, async (input) => formatFacebookAdTranscribe(await executor.facebookAdTranscribe(input), input));
   server.registerTool("google_ads_search", {
     title: "Google Ads Transparency Search",
-    description: "Search the Google Ads Transparency Center to find advertisers by domain or brand name. Returns advertisers with advertiser ID and approximate ad count, to pass to google_ads_page_intel.",
+    description: "Search the Google Ads Transparency Center by brand, person, or website. Returns two lists: advertisers (name, country, approximate ad count) and websites (domains). Neither carries an advertiser ID. Pass a websites[].domain (preferred - users know the website, not the registered advertiser name) or an advertisers[].name to google_ads_page_intel to pull that advertiser's actual ads.",
     inputSchema: GoogleAdsSearchInputSchema,
     outputSchema: recordOutputSchema("google_ads_search", GoogleAdsSearchOutputSchema),
     annotations: liveWebToolAnnotations("Google Ads Transparency Search")
   }, async (input) => formatGoogleAdsSearch(await executor.googleAdsSearch(input), input));
   server.registerTool("google_ads_page_intel", {
     title: "Google Ads Advertiser Intel",
-    description: "Harvest an advertiser's ad creatives from the Google Ads Transparency Center: format, image URLs, and \u2014 for video ads \u2014 a YouTube video ID or direct video URL. Accepts advertiserId (from google_ads_search) or a domain.",
+    description: "Harvest an advertiser's ad creatives from the Google Ads Transparency Center: format, image URLs, and \u2014 for video ads \u2014 a YouTube video ID or direct video URL. Accepts a domain or advertiserName from google_ads_search (google_ads_search does not return advertiser IDs). A domain may map to several advertiser accounts; this returns the primary one.",
     inputSchema: GoogleAdsPageIntelInputSchema,
     outputSchema: recordOutputSchema("google_ads_page_intel", GoogleAdsPageIntelOutputSchema),
     annotations: liveWebToolAnnotations("Google Ads Advertiser Intel")
@@ -11267,4 +11282,4 @@ export {
   MEMORY_TOOL_SCHEMAS,
   registerMemoryMcpTools
 };
-//# sourceMappingURL=chunk-P64PZUCA.js.map
+//# sourceMappingURL=chunk-NKCCGADE.js.map