mcp-scraper 0.33.2 → 0.33.4

package/README.md

@@ -88,7 +88,7 @@ Build the branded one-click bundle:
 npm run build:mcpb
 ```
 
-The generated bundle is written to `build/mcpb/mcp-scraper-<version>.mcpb` and copied to `public/downloads/` for the hosted download. The current public bundle is `https://mcpscraper.dev/downloads/mcp-scraper.mcpb` (`0.33.2`, SHA-256 `3b754cbebacb52749e41a141f6b94e8082861586b60e81f0433ca30cb676949e`). Install it by opening or dragging it into Claude Desktop. Claude displays the `MCP Scraper` install card, icon, and API-key configuration field from the bundle manifest.
+The generated bundle is written to `build/mcpb/mcp-scraper-<version>.mcpb` and copied to `public/downloads/` for the hosted download. The current public bundle is `https://mcpscraper.dev/downloads/mcp-scraper.mcpb` (`0.33.4`, SHA-256 `9444edeec9e9bde429936ed0f9953a84569e17ad37ebbc1b66f7ac05fb40eb96`). Install it by opening or dragging it into Claude Desktop. Claude displays the `MCP Scraper` install card, icon, and API-key configuration field from the bundle manifest.
 
 The MCPB install exposes every tool — web-intelligence plus all `browser_*` tools — through the one `mcp-scraper` server.
 

package/dist/bin/api-server.cjs

@@ -29815,7 +29815,7 @@ var PACKAGE_VERSION;
 var init_version = __esm({
   "src/version.ts"() {
     "use strict";
-    PACKAGE_VERSION = "0.33.2";
+    PACKAGE_VERSION = "0.33.4";
   }
 });
 
@@ -30279,9 +30279,9 @@ var init_mcp_tool_schemas = __esm({
       gl: import_zod35.z.string().length(2).default("us").describe("Google country code inferred from location or user language."),
       hl: import_zod35.z.string().default("en").describe("Google interface/content language inferred from the user request."),
       device: import_zod35.z.enum(["desktop", "mobile"]).default("desktop").describe("SERP device context. Use mobile only for mobile rankings."),
-      proxyMode: import_zod35.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: import_zod35.z.string().regex(/^\d{5}$/).optional().describe('US ZIP for residential geo-IP targeting. Only meaningful with proxyMode "location".'),
-      debug: import_zod35.z.boolean().default(false).describe("Include sanitized diagnostics for debugging localization, CAPTCHA, or proxy behavior.")
+      proxyMode: import_zod35.z.enum(["configured", "none"]).default(DEFAULT_PROXY_MODE).describe("Leave unset for the default route. Country/region localization comes from gl/hl plus the city or region in the query."),
+      proxyZip: import_zod35.z.string().regex(/^\d{5}$/).optional().describe("Optional US ZIP override."),
+      debug: import_zod35.z.boolean().default(false).describe("Include sanitized diagnostics for debugging.")
     };
     ExtractUrlInputSchema = {
       url: import_zod35.z.string().url().describe("Public http/https URL to extract."),
@@ -30306,7 +30306,7 @@ var init_mcp_tool_schemas = __esm({
     ExtractSiteInputSchema = {
       url: import_zod35.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: import_zod35.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: import_zod35.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: import_zod35.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: import_zod35.z.number().int().min(1).max(100).optional().describe("When rotateProxies is on, pages fetched per proxy before rotating. Default 30."),
       formats: import_zod35.z.array(import_zod35.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: import_zod35.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."),
@@ -30315,7 +30315,7 @@ var init_mcp_tool_schemas = __esm({
     AuditSiteInputSchema = {
       url: import_zod35.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: import_zod35.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: import_zod35.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: import_zod35.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: import_zod35.z.number().int().min(1).max(100).optional().describe("When rotateProxies is on, pages fetched per proxy before rotating. Default 30."),
       background: import_zod35.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: import_zod35.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.")
@@ -30448,8 +30448,8 @@ var init_mcp_tool_schemas = __esm({
       hl: import_zod35.z.string().length(2).default("en").describe("Language inferred from user request."),
       maxResults: import_zod35.z.number().int().min(1).max(50).default(10).describe("Number of candidates to return. Default 10, maximum 50."),
       includeServices: import_zod35.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: import_zod35.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: import_zod35.z.string().regex(/^\d{5}$/).optional().describe("Optional US ZIP override, only used when proxyMode is location."),
+      proxyMode: import_zod35.z.enum(["configured", "none"]).default(DEFAULT_MAPS_PROXY_MODE).describe("Leave unset for the default route. Country/region localization comes from the city or region in the query plus gl/hl."),
+      proxyZip: import_zod35.z.string().regex(/^\d{5}$/).optional().describe("Optional US ZIP override."),
       debug: import_zod35.z.boolean().default(false).describe("Include sanitized browser/proxy diagnostics.")
     };
     DirectoryWorkflowInputSchema = {
@@ -30463,8 +30463,8 @@ var init_mcp_tool_schemas = __esm({
       includeZipGroups: import_zod35.z.boolean().default(true).describe("Attach ZIP groups from a configured US ZIPS CSV when available (MCP_SCRAPER_USZIPS_CSV_PATH or usZipsCsvPath)."),
       usZipsCsvPath: import_zod35.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: import_zod35.z.boolean().default(true).describe("Save a directory-ready CSV of results to the MCP Scraper output directory and return its path."),
-      proxyMode: import_zod35.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: import_zod35.z.string().regex(/^\d{5}$/).optional().describe("Optional ZIP override for proxy targeting; normally omitted."),
+      proxyMode: import_zod35.z.enum(["configured", "none"]).default(DEFAULT_MAPS_PROXY_MODE).describe("Proxy behavior per city search. Leave unset for the default route. Country/region localization comes from the city or region in the query plus gl/hl."),
+      proxyZip: import_zod35.z.string().regex(/^\d{5}$/).optional().describe("Optional US ZIP override."),
       debug: import_zod35.z.boolean().default(false).describe("Include sanitized browser/proxy diagnostics.")
     };
     ArtifactPointerOutputSchema = import_zod35.z.object({
@@ -31303,9 +31303,9 @@ var init_mcp_tool_schemas = __esm({
       gl: import_zod35.z.string().length(2).default("us").describe("Google country code inferred from location or user language."),
       hl: import_zod35.z.string().default("en").describe("Google interface/content language inferred from user request."),
       device: import_zod35.z.enum(["desktop", "mobile"]).default("desktop").describe("SERP device context. Use mobile only for mobile rankings."),
-      proxyMode: import_zod35.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: import_zod35.z.string().regex(/^\d{5}$/).optional().describe('US ZIP for residential geo-IP targeting. Only meaningful with proxyMode "location".'),
-      debug: import_zod35.z.boolean().default(false).describe("Include sanitized diagnostics for debugging localization, CAPTCHA, or proxy behavior."),
+      proxyMode: import_zod35.z.enum(["configured", "none"]).default(DEFAULT_PROXY_MODE).describe("Leave unset for the default route. Country/region localization comes from gl/hl plus the city or region in the query."),
+      proxyZip: import_zod35.z.string().regex(/^\d{5}$/).optional().describe("Optional US ZIP override."),
+      debug: import_zod35.z.boolean().default(false).describe("Include sanitized diagnostics for debugging."),
       pages: import_zod35.z.number().int().min(1).max(2).default(1).describe("Number of result pages to fetch (1\u20132).")
     };
     CaptureSerpSnapshotInputSchema = {
@@ -31314,8 +31314,8 @@ var init_mcp_tool_schemas = __esm({
       gl: import_zod35.z.string().length(2).default("us").describe("Google country code inferred from the requested market."),
       hl: import_zod35.z.string().default("en").describe("Google interface/content language inferred from the user request."),
       device: import_zod35.z.enum(["desktop", "mobile"]).default("desktop").describe("SERP device context. Use mobile only for mobile rankings/evidence."),
-      proxyMode: import_zod35.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: import_zod35.z.string().regex(/^\d{5}$/).optional().describe('US ZIP for residential geo-IP targeting. Only meaningful with proxyMode "location".'),
+      proxyMode: import_zod35.z.enum(["configured", "none"]).default(DEFAULT_PROXY_MODE).describe("Leave unset for the default route. Country/region localization comes from gl/hl plus the city or region in the query."),
+      proxyZip: import_zod35.z.string().regex(/^\d{5}$/).optional().describe("Optional US ZIP override."),
       pages: import_zod35.z.number().int().min(1).max(2).default(1).describe("Google result pages to capture. Use 2 only for deeper ranking evidence."),
       debug: import_zod35.z.boolean().default(false).describe("Include sanitized browser/proxy/location diagnostics."),
       includePageSnapshots: import_zod35.z.boolean().default(false).describe("Also capture ranking-page snapshots for selected SERP URLs. Each attempted snapshot adds 1 Credit."),