mcp-scraper 0.33.2 → 0.33.4

package/dist/bin/mcp-stdio-server.js

@@ -7,7 +7,7 @@ import {
   registerMemoryMcpTools,
   registerPaaExtractorMcpTools,
   registerSerpIntelligenceCaptureTools
-} from "../chunk-CC4VK5CR.js";
+} from "../chunk-ACBNIIBU.js";
 import "../chunk-R7EETU7Z.js";
 import "../chunk-MTSBI7ZH.js";
 import {
@@ -16,7 +16,7 @@ import {
 import "../chunk-V73MPRU6.js";
 import {
   PACKAGE_VERSION
-} from "../chunk-ZGEEL2LR.js";
+} from "../chunk-IQ742H22.js";
 import "../chunk-4767BA2O.js";
 import "../chunk-M2S27J6Z.js";
 import "../chunk-ICT7DDHL.js";

package/dist/{chunk-CC4VK5CR.js → chunk-ACBNIIBU.js}

@@ -19,7 +19,7 @@ import {
 } from "./chunk-V73MPRU6.js";
 import {
   PACKAGE_VERSION
-} from "./chunk-ZGEEL2LR.js";
+} from "./chunk-IQ742H22.js";
 import {
   MC_PER_CREDIT
 } from "./chunk-4767BA2O.js";
@@ -3481,9 +3481,9 @@ var HarvestPaaInputSchema = {
   gl: z.string().length(2).default("us").describe("Google country code inferred from location or user language."),
   hl: z.string().default("en").describe("Google interface/content language inferred from the user request."),
   device: z.enum(["desktop", "mobile"]).default("desktop").describe("SERP device context. Use mobile only for mobile rankings."),
-  proxyMode: z.enum(["configured", "none"]).default(DEFAULT_PROXY_MODE).describe("Leave unset for clean egress (the default). Country/region localization comes from gl/hl plus the city or region in the query \u2014 not from a proxy."),
-  proxyZip: z.string().regex(/^\d{5}$/).optional().describe('US ZIP for residential geo-IP targeting. Only meaningful with proxyMode "location".'),
-  debug: z.boolean().default(false).describe("Include sanitized diagnostics for debugging localization, CAPTCHA, or proxy behavior.")
+  proxyMode: z.enum(["configured", "none"]).default(DEFAULT_PROXY_MODE).describe("Leave unset for the default route. Country/region localization comes from gl/hl plus the city or region in the query."),
+  proxyZip: z.string().regex(/^\d{5}$/).optional().describe("Optional US ZIP override."),
+  debug: z.boolean().default(false).describe("Include sanitized diagnostics for debugging.")
 };
 var ExtractUrlInputSchema = {
   url: z.string().url().describe("Public http/https URL to extract."),
@@ -3508,7 +3508,7 @@ var MapSiteUrlsInputSchema = {
 var ExtractSiteInputSchema = {
   url: z.string().url().describe("Public website URL or domain to crawl for page CONTENT (map + scrape). For a technical SEO audit use audit_site instead \u2014 this returns content only, not analysis."),
   maxPages: z.number().int().min(1).max(1e4).optional().describe("Maximum pages to extract. Bulk crawls (over 25 pages) switch to folder mode: each page saved as its own Markdown file, with a summary plus folder path returned instead of inlining content."),
-  rotateProxies: z.boolean().optional().describe("Route page fetches through rotating residential proxies to defeat rate-limiting and bot blocks (403/429). Slower and pricier \u2014 use only when a site blocks normal crawling."),
+  rotateProxies: z.boolean().optional().describe("Use extra measures to get past sites that block normal crawling (403/429). Slower and pricier \u2014 use only when a site blocks normal crawling."),
   rotateProxyEvery: z.number().int().min(1).max(100).optional().describe("When rotateProxies is on, pages fetched per proxy before rotating. Default 30."),
   formats: z.array(z.enum(["markdown", "links", "json", "images", "branding"])).optional().describe("Per-page output formats: markdown, links, json, images are captured cheaply from HTML; branding (site-level logo/colors/fonts) requires a browser and adds time. Defaults to markdown+links."),
   background: z.boolean().default(false).describe("Run the crawl as a background job instead of blocking this call, returning a jobId immediately \u2014 poll it with check_site_export to get a downloadable zip (all page content, plus real image files if downloadImages is set) once ready. Use for large sites where a synchronous call would be slow."),
@@ -3517,7 +3517,7 @@ var ExtractSiteInputSchema = {
 var AuditSiteInputSchema = {
   url: z.string().url().describe("Public website URL or domain for a full technical SEO audit (issues, link graph, indexability, headings, images). For plain content use extract_site instead."),
   maxPages: z.number().int().min(1).max(1e4).optional().describe("Maximum pages to crawl and audit. Always writes a folder of analysis files plus per-page content, returning a summary plus the folder path."),
-  rotateProxies: z.boolean().optional().describe("Route page fetches through rotating residential proxies to defeat rate-limiting and bot blocks. Slower/pricier \u2014 use only when a site blocks normal crawling."),
+  rotateProxies: z.boolean().optional().describe("Use extra measures to get past sites that block normal crawling. Slower/pricier \u2014 use only when a site blocks normal crawling."),
   rotateProxyEvery: z.number().int().min(1).max(100).optional().describe("When rotateProxies is on, pages fetched per proxy before rotating. Default 30."),
   background: z.boolean().default(false).describe("Run the audit as a background job instead of blocking this call, returning a jobId immediately \u2014 poll it with check_site_export to get a downloadable zip (full audit report, all page content, plus real image files if downloadImages is set) once ready. Use for large sites where a synchronous call would be slow."),
   downloadImages: z.boolean().default(false).describe("Download every discovered image as a real file into the export bundle (not just image URLs/stats). OFF by default \u2014 must be explicitly set true. Implies background regardless of the background flag, since downloading a whole site's images is too slow to run synchronously. Capped at 20 images/page and 500 images/site.")
@@ -3650,8 +3650,8 @@ var MapsSearchInputSchema = {
   hl: z.string().length(2).default("en").describe("Language inferred from user request."),
   maxResults: z.number().int().min(1).max(50).default(10).describe("Number of candidates to return. Default 10, maximum 50."),
   includeServices: z.boolean().default(false).describe("Open each returned business profile to include its configured services and areas served when available. Adds a page visit per business; does not collect review cards."),
-  proxyMode: z.enum(["configured", "none"]).default(DEFAULT_MAPS_PROXY_MODE).describe("Leave unset for the default route (stealth browser on the managed ISP proxy, retried on a fresh session when Google blocks). Country/region localization comes from the city or region in the query plus gl/hl \u2014 not from a proxy."),
-  proxyZip: z.string().regex(/^\d{5}$/).optional().describe("Optional US ZIP override, only used when proxyMode is location."),
+  proxyMode: z.enum(["configured", "none"]).default(DEFAULT_MAPS_PROXY_MODE).describe("Leave unset for the default route. Country/region localization comes from the city or region in the query plus gl/hl."),
+  proxyZip: z.string().regex(/^\d{5}$/).optional().describe("Optional US ZIP override."),
   debug: z.boolean().default(false).describe("Include sanitized browser/proxy diagnostics.")
 };
 var DirectoryWorkflowInputSchema = {
@@ -3665,8 +3665,8 @@ var DirectoryWorkflowInputSchema = {
   includeZipGroups: z.boolean().default(true).describe("Attach ZIP groups from a configured US ZIPS CSV when available (MCP_SCRAPER_USZIPS_CSV_PATH or usZipsCsvPath)."),
   usZipsCsvPath: z.string().optional().describe("Local/test-only path to a US ZIPS CSV (state_abbr, zipcode, county, city columns). Deployed APIs should use MCP_SCRAPER_USZIPS_CSV_PATH instead. For ZIP enrichment, set MCP_SCRAPER_USZIPS_CSV_PATH on the server, or pass this in local/test mode."),
   saveCsv: z.boolean().default(true).describe("Save a directory-ready CSV of results to the MCP Scraper output directory and return its path."),
-  proxyMode: z.enum(["configured", "none"]).default(DEFAULT_MAPS_PROXY_MODE).describe("Proxy behavior per city search. Leave unset for the default route (stealth browser on the managed ISP proxy, retried fresh on a Google block). Country/region localization comes from the city or region in the query plus gl/hl \u2014 not from a proxy."),
-  proxyZip: z.string().regex(/^\d{5}$/).optional().describe("Optional ZIP override for proxy targeting; normally omitted."),
+  proxyMode: z.enum(["configured", "none"]).default(DEFAULT_MAPS_PROXY_MODE).describe("Proxy behavior per city search. Leave unset for the default route. Country/region localization comes from the city or region in the query plus gl/hl."),
+  proxyZip: z.string().regex(/^\d{5}$/).optional().describe("Optional US ZIP override."),
   debug: z.boolean().default(false).describe("Include sanitized browser/proxy diagnostics.")
 };
 var ArtifactPointerOutputSchema = z.object({
@@ -4505,9 +4505,9 @@ var SearchSerpInputSchema = {
   gl: z.string().length(2).default("us").describe("Google country code inferred from location or user language."),
   hl: z.string().default("en").describe("Google interface/content language inferred from user request."),
   device: z.enum(["desktop", "mobile"]).default("desktop").describe("SERP device context. Use mobile only for mobile rankings."),
-  proxyMode: z.enum(["configured", "none"]).default(DEFAULT_PROXY_MODE).describe("Leave unset for clean egress (the default). Country/region localization comes from gl/hl plus the city or region in the query \u2014 not from a proxy."),
-  proxyZip: z.string().regex(/^\d{5}$/).optional().describe('US ZIP for residential geo-IP targeting. Only meaningful with proxyMode "location".'),
-  debug: z.boolean().default(false).describe("Include sanitized diagnostics for debugging localization, CAPTCHA, or proxy behavior."),
+  proxyMode: z.enum(["configured", "none"]).default(DEFAULT_PROXY_MODE).describe("Leave unset for the default route. Country/region localization comes from gl/hl plus the city or region in the query."),
+  proxyZip: z.string().regex(/^\d{5}$/).optional().describe("Optional US ZIP override."),
+  debug: z.boolean().default(false).describe("Include sanitized diagnostics for debugging."),
   pages: z.number().int().min(1).max(2).default(1).describe("Number of result pages to fetch (1\u20132).")
 };
 var CaptureSerpSnapshotInputSchema = {
@@ -4516,8 +4516,8 @@ var CaptureSerpSnapshotInputSchema = {
   gl: z.string().length(2).default("us").describe("Google country code inferred from the requested market."),
   hl: z.string().default("en").describe("Google interface/content language inferred from the user request."),
   device: z.enum(["desktop", "mobile"]).default("desktop").describe("SERP device context. Use mobile only for mobile rankings/evidence."),
-  proxyMode: z.enum(["configured", "none"]).default(DEFAULT_PROXY_MODE).describe("Leave unset for clean egress (the default). Country/region localization comes from gl/hl plus the city or region in the query \u2014 not from a proxy."),
-  proxyZip: z.string().regex(/^\d{5}$/).optional().describe('US ZIP for residential geo-IP targeting. Only meaningful with proxyMode "location".'),
+  proxyMode: z.enum(["configured", "none"]).default(DEFAULT_PROXY_MODE).describe("Leave unset for the default route. Country/region localization comes from gl/hl plus the city or region in the query."),
+  proxyZip: z.string().regex(/^\d{5}$/).optional().describe("Optional US ZIP override."),
   pages: z.number().int().min(1).max(2).default(1).describe("Google result pages to capture. Use 2 only for deeper ranking evidence."),
   debug: z.boolean().default(false).describe("Include sanitized browser/proxy/location diagnostics."),
   includePageSnapshots: z.boolean().default(false).describe("Also capture ranking-page snapshots for selected SERP URLs. Each attempted snapshot adds 1 Credit."),
@@ -10151,4 +10151,4 @@ export {
   MEMORY_TOOL_SCHEMAS,
   registerMemoryMcpTools
 };
-//# sourceMappingURL=chunk-CC4VK5CR.js.map
+//# sourceMappingURL=chunk-ACBNIIBU.js.map