mcp-scraper 0.19.0 → 0.20.0

package/dist/chunk-JWIE5NCR.js

@@ -0,0 +1,284 @@
+// src/lib/media-extractor.ts
+import { createWriteStream, mkdirSync } from "fs";
+import { homedir } from "os";
+import { join, extname, basename } from "path";
+import { pipeline } from "stream/promises";
+import { Readable } from "stream";
+var AD_PATTERNS = [
+  "doubleclick.net",
+  "googlesyndication.com",
+  "googletagmanager.com",
+  "google-analytics.com",
+  "googletagservices.com",
+  "adservice.google",
+  "googletag.",
+  "pagead2.googlesyndication",
+  "facebook.net/tr",
+  "connect.facebook.net",
+  "fbcdn.net/rsrc",
+  "analytics.twitter.com",
+  "static.ads-twitter.com",
+  "ads.twitter.com",
+  "t.co/i/adsct",
+  "pixel.advertising.com",
+  "hotjar.com",
+  "clarity.ms",
+  "quantserve.com",
+  "scorecardresearch.com",
+  "newrelic.com",
+  "nr-data.net",
+  "segment.io",
+  "segment.com",
+  "amplitude.com",
+  "mixpanel.com",
+  "heap.io",
+  "fullstory.com",
+  "moatads.com",
+  "criteo.com",
+  "adsrvr.org",
+  "rubiconproject.com",
+  "pubmatic.com",
+  "openx.net",
+  "appnexus.com",
+  "amazon-adsystem.com",
+  "media.net",
+  "yieldmo.com",
+  "triplelift.com",
+  "sharethrough.com",
+  "prebid.",
+  "smaato.net",
+  "indexworm.com",
+  "casalemedia.com",
+  "outbrain.com",
+  "taboola.com",
+  "revcontent.com",
+  "mgid.com",
+  "tawk.to",
+  "intercom.io",
+  "drift.com",
+  "hs-scripts.com",
+  "zopim.com",
+  "livechatinc.com",
+  "userlike.com",
+  "onetrust.com",
+  "cookielaw.org",
+  "cookieinformation.com",
+  "trustarc.com",
+  "/ads/",
+  "/ad/",
+  "/banner/",
+  "/banners/",
+  "/pixel/",
+  "/beacon/",
+  "/tracking/",
+  "/tracker/",
+  "/remarketing/",
+  "/conversion/",
+  "1x1.gif",
+  "spacer.gif",
+  "blank.gif",
+  "transparent.gif"
+];
+var IMAGE_EXTS = /* @__PURE__ */ new Set([".jpg", ".jpeg", ".png", ".webp", ".gif", ".avif", ".svg", ".tiff"]);
+var VIDEO_EXTS = /* @__PURE__ */ new Set([".mp4", ".webm", ".mov", ".avi", ".m4v", ".ogv", ".mkv"]);
+var AUDIO_EXTS = /* @__PURE__ */ new Set([".mp3", ".wav", ".ogg", ".aac", ".m4a", ".flac", ".opus"]);
+function isAdUrl(url) {
+  const lower = url.toLowerCase();
+  return AD_PATTERNS.some((p) => lower.includes(p));
+}
+function isDataUri(url) {
+  return url.startsWith("data:");
+}
+function typeFromUrl(url) {
+  try {
+    const ext = extname(new URL(url).pathname).toLowerCase();
+    if (IMAGE_EXTS.has(ext)) return "image";
+    if (VIDEO_EXTS.has(ext)) return "video";
+    if (AUDIO_EXTS.has(ext)) return "audio";
+  } catch {
+  }
+  return null;
+}
+function typeFromMime(mime) {
+  const lower = mime.toLowerCase();
+  if (lower.startsWith("image/")) return "image";
+  if (lower.startsWith("video/")) return "video";
+  if (lower.startsWith("audio/")) return "audio";
+  return null;
+}
+function resolveUrl(raw, base) {
+  if (!raw || isDataUri(raw)) return null;
+  try {
+    return new URL(raw, base).href;
+  } catch {
+    return null;
+  }
+}
+function safeFilename(url, index) {
+  try {
+    const u = new URL(url);
+    const base = basename(u.pathname).replace(/[^a-zA-Z0-9._-]/g, "_").slice(0, 80);
+    return base || `asset-${index}`;
+  } catch {
+    return `asset-${index}`;
+  }
+}
+function extractMediaUrls(html, baseUrl) {
+  const seen = /* @__PURE__ */ new Set();
+  const urls = [];
+  const add = (raw) => {
+    const resolved = resolveUrl(raw.trim(), baseUrl);
+    if (!resolved || seen.has(resolved) || isAdUrl(resolved)) return;
+    seen.add(resolved);
+    urls.push(resolved);
+  };
+  for (const m of html.matchAll(/<img\s[^>]*>/gi)) {
+    const tag = m[0];
+    for (const attr of ["src", "data-src", "data-lazy-src", "data-original"]) {
+      const v = (tag.match(new RegExp(`\\b${attr}\\s*=\\s*["']([^"']+)["']`, "i")) ?? [])[1];
+      if (v) add(v);
+    }
+    const srcset = (tag.match(/\bsrcset\s*=\s*["']([^"']+)["']/i) ?? [])[1];
+    if (srcset) {
+      for (const part of srcset.split(",")) {
+        const u = part.trim().split(/\s+/)[0];
+        if (u) add(u);
+      }
+    }
+  }
+  for (const m of html.matchAll(/<source\s[^>]*>/gi)) {
+    const tag = m[0];
+    const src = (tag.match(/\bsrc\s*=\s*["']([^"']+)["']/i) ?? [])[1];
+    if (src) add(src);
+    const srcset = (tag.match(/\bsrcset\s*=\s*["']([^"']+)["']/i) ?? [])[1];
+    if (srcset) {
+      for (const part of srcset.split(",")) {
+        const u = part.trim().split(/\s+/)[0];
+        if (u) add(u);
+      }
+    }
+  }
+  for (const m of html.matchAll(/<video\s[^>]*>/gi)) {
+    const tag = m[0];
+    for (const attr of ["src", "poster"]) {
+      const v = (tag.match(new RegExp(`\\b${attr}\\s*=\\s*["']([^"']+)["']`, "i")) ?? [])[1];
+      if (v) add(v);
+    }
+  }
+  for (const m of html.matchAll(/<audio\s[^>]*>/gi)) {
+    const tag = m[0];
+    const v = (tag.match(/\bsrc\s*=\s*["']([^"']+)["']/i) ?? [])[1];
+    if (v) add(v);
+  }
+  for (const m of html.matchAll(/<meta\s[^>]*>/gi)) {
+    const tag = m[0];
+    const prop = (tag.match(/\b(?:property|name)\s*=\s*["']([^"']+)["']/i) ?? [])[1]?.toLowerCase();
+    if (prop === "og:image" || prop === "twitter:image") {
+      const content = (tag.match(/\bcontent\s*=\s*["']([^"']+)["']/i) ?? [])[1];
+      if (content) add(content);
+    }
+  }
+  for (const m of html.matchAll(/background(?:-image)?\s*:\s*url\(\s*["']?([^"')]+)["']?\s*\)/gi)) {
+    add(m[1]);
+  }
+  return urls;
+}
+async function downloadAsset(url, destDir, filename) {
+  const res = await fetch(url, {
+    headers: { "User-Agent": "Mozilla/5.0 (compatible; ThorbitBot/1.0)" },
+    signal: AbortSignal.timeout(15e3)
+  });
+  if (!res.ok) throw new Error(`HTTP ${res.status}`);
+  if (!res.body) throw new Error("Empty response body");
+  const mimeType = res.headers.get("content-type")?.split(";")[0].trim() ?? null;
+  let dest = join(destDir, filename);
+  if (mimeType && !extname(filename)) {
+    const mimeExt = {
+      "image/jpeg": ".jpg",
+      "image/png": ".png",
+      "image/webp": ".webp",
+      "image/gif": ".gif",
+      "image/svg+xml": ".svg",
+      "image/avif": ".avif",
+      "video/mp4": ".mp4",
+      "video/webm": ".webm",
+      "audio/mpeg": ".mp3",
+      "audio/ogg": ".ogg",
+      "audio/wav": ".wav"
+    };
+    const ext = mimeExt[mimeType];
+    if (ext) dest = dest + ext;
+  }
+  const writer = createWriteStream(dest);
+  await pipeline(Readable.fromWeb(res.body), writer);
+  const { statSync } = await import("fs");
+  const sizeBytes = statSync(dest).size;
+  return { savedPath: dest, sizeBytes, mimeType };
+}
+async function harvestPageMedia(html, pageUrl, options = {}) {
+  const types = options.types ?? ["image", "video", "audio"];
+  const typesSet = new Set(types);
+  const rawUrls = extractMediaUrls(html, pageUrl);
+  const totalFound = rawUrls.length;
+  const filteredUrls = [];
+  const kept = [];
+  for (const url of rawUrls) {
+    const type = typeFromUrl(url);
+    if (!type || !typesSet.has(type)) {
+      filteredUrls.push(url);
+      continue;
+    }
+    kept.push({ url, type });
+  }
+  if (options.outputDir === null) {
+    return {
+      pageUrl,
+      outputDir: null,
+      assets: kept.map(({ url, type }, i) => ({
+        url,
+        type,
+        mimeType: null,
+        filename: safeFilename(url, i),
+        savedPath: null,
+        sizeBytes: null
+      })),
+      filteredCount: filteredUrls.length,
+      totalFound
+    };
+  }
+  const domain = (() => {
+    try {
+      return new URL(pageUrl).hostname.replace(/^www\./, "");
+    } catch {
+      return "unknown";
+    }
+  })();
+  const stamp = (/* @__PURE__ */ new Date()).toISOString().replace(/[:.]/g, "-").slice(0, 19);
+  const outDir = options.outputDir ?? join(homedir(), "Downloads", "mcp-scraper", "media", `${stamp}-${domain}`);
+  mkdirSync(outDir, { recursive: true });
+  const assets = [];
+  await Promise.allSettled(
+    kept.map(async ({ url, type }, i) => {
+      const filename = safeFilename(url, i);
+      try {
+        const { savedPath, sizeBytes, mimeType } = await downloadAsset(url, outDir, filename);
+        const resolvedType = mimeType ? typeFromMime(mimeType) ?? type : type;
+        assets.push({ url, type: resolvedType, mimeType, filename: basename(savedPath), savedPath, sizeBytes });
+      } catch {
+        assets.push({ url, type, mimeType: null, filename, savedPath: null, sizeBytes: null });
+      }
+    })
+  );
+  assets.sort((a, b) => {
+    if (a.savedPath && !b.savedPath) return -1;
+    if (!a.savedPath && b.savedPath) return 1;
+    return a.url.localeCompare(b.url);
+  });
+  return { pageUrl, outputDir: outDir, assets, filteredCount: filteredUrls.length, totalFound };
+}
+
+export {
+  downloadAsset,
+  harvestPageMedia
+};
+//# sourceMappingURL=chunk-JWIE5NCR.js.map

package/dist/chunk-JWIE5NCR.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../src/lib/media-extractor.ts"],"sourcesContent":["import { createWriteStream, mkdirSync } from 'node:fs'\nimport { homedir } from 'node:os'\nimport { join, extname, basename } from 'node:path'\nimport { pipeline } from 'node:stream/promises'\nimport { Readable } from 'node:stream'\n\nconst AD_PATTERNS: string[] = [\n  'doubleclick.net', 'googlesyndication.com', 'googletagmanager.com',\n  'google-analytics.com', 'googletagservices.com', 'adservice.google',\n  'googletag.', 'pagead2.googlesyndication',\n  'facebook.net/tr', 'connect.facebook.net', 'fbcdn.net/rsrc',\n  'analytics.twitter.com', 'static.ads-twitter.com', 'ads.twitter.com',\n  't.co/i/adsct', 'pixel.advertising.com',\n  'hotjar.com', 'clarity.ms', 'quantserve.com', 'scorecardresearch.com',\n  'newrelic.com', 'nr-data.net', 'segment.io', 'segment.com',\n  'amplitude.com', 'mixpanel.com', 'heap.io', 'fullstory.com',\n  'moatads.com', 'criteo.com', 'adsrvr.org', 'rubiconproject.com',\n  'pubmatic.com', 'openx.net', 'appnexus.com', 'amazon-adsystem.com',\n  'media.net', 'yieldmo.com', 'triplelift.com', 'sharethrough.com',\n  'prebid.', 'smaato.net', 'indexworm.com', 'casalemedia.com',\n  'outbrain.com', 'taboola.com', 'revcontent.com', 'mgid.com',\n  'tawk.to', 'intercom.io', 'drift.com', 'hs-scripts.com',\n  'zopim.com', 'livechatinc.com', 'userlike.com',\n  'onetrust.com', 'cookielaw.org', 'cookieinformation.com', 'trustarc.com',\n  '/ads/', '/ad/', '/banner/', '/banners/', '/pixel/', '/beacon/',\n  '/tracking/', '/tracker/', '/remarketing/', '/conversion/',\n  '1x1.gif', 'spacer.gif', 'blank.gif', 'transparent.gif',\n]\n\nconst IMAGE_EXTS  = new Set(['.jpg', '.jpeg', '.png', '.webp', '.gif', '.avif', '.svg', '.tiff'])\nconst VIDEO_EXTS  = new Set(['.mp4', '.webm', '.mov', '.avi', '.m4v', '.ogv', '.mkv'])\nconst AUDIO_EXTS  = new Set(['.mp3', '.wav', '.ogg', '.aac', '.m4a', '.flac', '.opus'])\n\nexport type MediaType = 'image' | 'video' | 'audio'\n\nexport interface MediaAsset {\n  url:        string\n  type:       MediaType\n  mimeType:   string | null\n  filename:   string\n  savedPath:  string | null\n  sizeBytes:  number | null\n}\n\nexport interface MediaManifest {\n  pageUrl:       string\n  outputDir:     string | null\n  assets:        MediaAsset[]\n  filteredCount: number\n  totalFound:    number\n}\n\nexport interface MediaExtractOptions {\n  types?:      MediaType[]\n  outputDir?:  string | null\n}\n\nfunction isAdUrl(url: string): boolean {\n  const lower = url.toLowerCase()\n  return AD_PATTERNS.some(p => lower.includes(p))\n}\n\nfunction isDataUri(url: string): boolean {\n  return url.startsWith('data:')\n}\n\nfunction typeFromUrl(url: string): MediaType | null {\n  try {\n    const ext = extname(new URL(url).pathname).toLowerCase()\n    if (IMAGE_EXTS.has(ext)) return 'image'\n    if (VIDEO_EXTS.has(ext)) return 'video'\n    if (AUDIO_EXTS.has(ext)) return 'audio'\n  } catch { /* non-parseable URL */ }\n  return null\n}\n\nfunction typeFromMime(mime: string): MediaType | null {\n  const lower = mime.toLowerCase()\n  if (lower.startsWith('image/')) return 'image'\n  if (lower.startsWith('video/')) return 'video'\n  if (lower.startsWith('audio/')) return 'audio'\n  return null\n}\n\nfunction resolveUrl(raw: string, base: string): string | null {\n  if (!raw || isDataUri(raw)) return null\n  try { return new URL(raw, base).href } catch { return null }\n}\n\nfunction safeFilename(url: string, index: number): string {\n  try {\n    const u = new URL(url)\n    const base = basename(u.pathname).replace(/[^a-zA-Z0-9._-]/g, '_').slice(0, 80)\n    return base || `asset-${index}`\n  } catch {\n    return `asset-${index}`\n  }\n}\n\nexport function extractMediaUrls(html: string, baseUrl: string): string[] {\n  const seen = new Set<string>()\n  const urls: string[] = []\n\n  const add = (raw: string) => {\n    const resolved = resolveUrl(raw.trim(), baseUrl)\n    if (!resolved || seen.has(resolved) || isAdUrl(resolved)) return\n    seen.add(resolved)\n    urls.push(resolved)\n  }\n\n  for (const m of html.matchAll(/<img\\s[^>]*>/gi)) {\n    const tag = m[0]\n    for (const attr of ['src', 'data-src', 'data-lazy-src', 'data-original']) {\n      const v = (tag.match(new RegExp(`\\\\b${attr}\\\\s*=\\\\s*[\"']([^\"']+)[\"']`, 'i')) ?? [])[1]\n      if (v) add(v)\n    }\n    const srcset = (tag.match(/\\bsrcset\\s*=\\s*[\"']([^\"']+)[\"']/i) ?? [])[1]\n    if (srcset) {\n      for (const part of srcset.split(',')) {\n        const u = part.trim().split(/\\s+/)[0]\n        if (u) add(u)\n      }\n    }\n  }\n\n  for (const m of html.matchAll(/<source\\s[^>]*>/gi)) {\n    const tag = m[0]\n    const src = (tag.match(/\\bsrc\\s*=\\s*[\"']([^\"']+)[\"']/i) ?? [])[1]\n    if (src) add(src)\n    const srcset = (tag.match(/\\bsrcset\\s*=\\s*[\"']([^\"']+)[\"']/i) ?? [])[1]\n    if (srcset) {\n      for (const part of srcset.split(',')) {\n        const u = part.trim().split(/\\s+/)[0]\n        if (u) add(u)\n      }\n    }\n  }\n\n  for (const m of html.matchAll(/<video\\s[^>]*>/gi)) {\n    const tag = m[0]\n    for (const attr of ['src', 'poster']) {\n      const v = (tag.match(new RegExp(`\\\\b${attr}\\\\s*=\\\\s*[\"']([^\"']+)[\"']`, 'i')) ?? [])[1]\n      if (v) add(v)\n    }\n  }\n\n  for (const m of html.matchAll(/<audio\\s[^>]*>/gi)) {\n    const tag = m[0]\n    const v = (tag.match(/\\bsrc\\s*=\\s*[\"']([^\"']+)[\"']/i) ?? [])[1]\n    if (v) add(v)\n  }\n\n  for (const m of html.matchAll(/<meta\\s[^>]*>/gi)) {\n    const tag = m[0]\n    const prop = (tag.match(/\\b(?:property|name)\\s*=\\s*[\"']([^\"']+)[\"']/i) ?? [])[1]?.toLowerCase()\n    if (prop === 'og:image' || prop === 'twitter:image') {\n      const content = (tag.match(/\\bcontent\\s*=\\s*[\"']([^\"']+)[\"']/i) ?? [])[1]\n      if (content) add(content)\n    }\n  }\n\n  for (const m of html.matchAll(/background(?:-image)?\\s*:\\s*url\\(\\s*[\"']?([^\"')]+)[\"']?\\s*\\)/gi)) {\n    add(m[1])\n  }\n\n  return urls\n}\n\nexport async function downloadAsset(\n  url: string,\n  destDir: string,\n  filename: string,\n): Promise<{ savedPath: string; sizeBytes: number; mimeType: string | null }> {\n  const res = await fetch(url, {\n    headers: { 'User-Agent': 'Mozilla/5.0 (compatible; ThorbitBot/1.0)' },\n    signal:  AbortSignal.timeout(15_000),\n  })\n  if (!res.ok) throw new Error(`HTTP ${res.status}`)\n  if (!res.body) throw new Error('Empty response body')\n\n  const mimeType = res.headers.get('content-type')?.split(';')[0].trim() ?? null\n\n  let dest = join(destDir, filename)\n  if (mimeType && !extname(filename)) {\n    const mimeExt: Record<string, string> = {\n      'image/jpeg': '.jpg', 'image/png': '.png', 'image/webp': '.webp',\n      'image/gif': '.gif', 'image/svg+xml': '.svg', 'image/avif': '.avif',\n      'video/mp4': '.mp4', 'video/webm': '.webm',\n      'audio/mpeg': '.mp3', 'audio/ogg': '.ogg', 'audio/wav': '.wav',\n    }\n    const ext = mimeExt[mimeType]\n    if (ext) dest = dest + ext\n  }\n\n  const writer = createWriteStream(dest)\n  await pipeline(Readable.fromWeb(res.body as import('stream/web').ReadableStream), writer)\n\n  const { statSync } = await import('node:fs')\n  const sizeBytes = statSync(dest).size\n\n  return { savedPath: dest, sizeBytes, mimeType }\n}\n\nexport async function harvestPageMedia(\n  html: string,\n  pageUrl: string,\n  options: MediaExtractOptions = {},\n): Promise<MediaManifest> {\n  const types    = options.types ?? ['image', 'video', 'audio']\n  const typesSet = new Set(types)\n  const rawUrls  = extractMediaUrls(html, pageUrl)\n  const totalFound = rawUrls.length\n\n  const filteredUrls: string[] = []\n  const kept: Array<{ url: string; type: MediaType }> = []\n\n  for (const url of rawUrls) {\n    const type = typeFromUrl(url)\n    if (!type || !typesSet.has(type)) { filteredUrls.push(url); continue }\n    kept.push({ url, type })\n  }\n\n  if (options.outputDir === null) {\n    return {\n      pageUrl,\n      outputDir: null,\n      assets: kept.map(({ url, type }, i) => ({\n        url, type, mimeType: null, filename: safeFilename(url, i), savedPath: null, sizeBytes: null,\n      })),\n      filteredCount: filteredUrls.length,\n      totalFound,\n    }\n  }\n\n  const domain  = (() => { try { return new URL(pageUrl).hostname.replace(/^www\\./, '') } catch { return 'unknown' } })()\n  const stamp   = new Date().toISOString().replace(/[:.]/g, '-').slice(0, 19)\n  const outDir  = options.outputDir ?? join(homedir(), 'Downloads', 'mcp-scraper', 'media', `${stamp}-${domain}`)\n  mkdirSync(outDir, { recursive: true })\n\n  const assets: MediaAsset[] = []\n\n  await Promise.allSettled(\n    kept.map(async ({ url, type }, i) => {\n      const filename = safeFilename(url, i)\n      try {\n        const { savedPath, sizeBytes, mimeType } = await downloadAsset(url, outDir, filename)\n        const resolvedType = mimeType ? (typeFromMime(mimeType) ?? type) : type\n        assets.push({ url, type: resolvedType, mimeType, filename: basename(savedPath), savedPath, sizeBytes })\n      } catch {\n        assets.push({ url, type, mimeType: null, filename, savedPath: null, sizeBytes: null })\n      }\n    })\n  )\n\n  assets.sort((a, b) => {\n    if (a.savedPath && !b.savedPath) return -1\n    if (!a.savedPath && b.savedPath) return 1\n    return a.url.localeCompare(b.url)\n  })\n\n  return { pageUrl, outputDir: outDir, assets, filteredCount: filteredUrls.length, totalFound }\n}\n"],"mappings":";AAAA,SAAS,mBAAmB,iBAAiB;AAC7C,SAAS,eAAe;AACxB,SAAS,MAAM,SAAS,gBAAgB;AACxC,SAAS,gBAAgB;AACzB,SAAS,gBAAgB;AAEzB,IAAM,cAAwB;AAAA,EAC5B;AAAA,EAAmB;AAAA,EAAyB;AAAA,EAC5C;AAAA,EAAwB;AAAA,EAAyB;AAAA,EACjD;AAAA,EAAc;AAAA,EACd;AAAA,EAAmB;AAAA,EAAwB;AAAA,EAC3C;AAAA,EAAyB;AAAA,EAA0B;AAAA,EACnD;AAAA,EAAgB;AAAA,EAChB;AAAA,EAAc;AAAA,EAAc;AAAA,EAAkB;AAAA,EAC9C;AAAA,EAAgB;AAAA,EAAe;AAAA,EAAc;AAAA,EAC7C;AAAA,EAAiB;AAAA,EAAgB;AAAA,EAAW;AAAA,EAC5C;AAAA,EAAe;AAAA,EAAc;AAAA,EAAc;AAAA,EAC3C;AAAA,EAAgB;AAAA,EAAa;AAAA,EAAgB;AAAA,EAC7C;AAAA,EAAa;AAAA,EAAe;AAAA,EAAkB;AAAA,EAC9C;AAAA,EAAW;AAAA,EAAc;AAAA,EAAiB;AAAA,EAC1C;AAAA,EAAgB;AAAA,EAAe;AAAA,EAAkB;AAAA,EACjD;AAAA,EAAW;AAAA,EAAe;AAAA,EAAa;AAAA,EACvC;AAAA,EAAa;AAAA,EAAmB;AAAA,EAChC;AAAA,EAAgB;AAAA,EAAiB;AAAA,EAAyB;AAAA,EAC1D;AAAA,EAAS;AAAA,EAAQ;AAAA,EAAY;AAAA,EAAa;AAAA,EAAW;AAAA,EACrD;AAAA,EAAc;AAAA,EAAa;AAAA,EAAiB;AAAA,EAC5C;AAAA,EAAW;AAAA,EAAc;AAAA,EAAa;AACxC;AAEA,IAAM,aAAc,oBAAI,IAAI,CAAC,QAAQ,SAAS,QAAQ,SAAS,QAAQ,SAAS,QAAQ,OAAO,CAAC;AAChG,IAAM,aAAc,oBAAI,IAAI,CAAC,QAAQ,SAAS,QAAQ,QAAQ,QAAQ,QAAQ,MAAM,CAAC;AACrF,IAAM,aAAc,oBAAI,IAAI,CAAC,QAAQ,QAAQ,QAAQ,QAAQ,QAAQ,SAAS,OAAO,CAAC;AA0BtF,SAAS,QAAQ,KAAsB;AACrC,QAAM,QAAQ,IAAI,YAAY;AAC9B,SAAO,YAAY,KAAK,OAAK,MAAM,SAAS,CAAC,CAAC;AAChD;AAEA,SAAS,UAAU,KAAsB;AACvC,SAAO,IAAI,WAAW,OAAO;AAC/B;AAEA,SAAS,YAAY,KAA+B;AAClD,MAAI;AACF,UAAM,MAAM,QAAQ,IAAI,IAAI,GAAG,EAAE,QAAQ,EAAE,YAAY;AACvD,QAAI,WAAW,IAAI,GAAG,EAAG,QAAO;AAChC,QAAI,WAAW,IAAI,GAAG,EAAG,QAAO;AAChC,QAAI,WAAW,IAAI,GAAG,EAAG,QAAO;AAAA,EAClC,QAAQ;AAAA,EAA0B;AAClC,SAAO;AACT;AAEA,SAAS,aAAa,MAAgC;AACpD,QAAM,QAAQ,KAAK,YAAY;AAC/B,MAAI,MAAM,WAAW,QAAQ,EAAG,QAAO;AACvC,MAAI,MAAM,WAAW,QAAQ,EAAG,QAAO;AACvC,MAAI,MAAM,WAAW,QAAQ,EAAG,QAAO;AACvC,SAAO;AACT;AAEA,SAAS,WAAW,KAAa,MAA6B;AAC5D,MAAI,CAAC,OAAO,UAAU,GAAG,EAAG,QAAO;AACnC,MAAI;AAAE,WAAO,IAAI,IAAI,KAAK,IAAI,EAAE;AAAA,EAAK,QAAQ;AAAE,WAAO;AAAA,EAAK;AAC7D;AAEA,SAAS,aAAa,KAAa,OAAuB;AACxD,MAAI;AACF,UAAM,IAAI,IAAI,IAAI,GAAG;AACrB,UAAM,OAAO,SAAS,EAAE,QAAQ,EAAE,QAAQ,oBAAoB,GAAG,EAAE,MAAM,GAAG,EAAE;AAC9E,WAAO,QAAQ,SAAS,KAAK;AAAA,EAC/B,QAAQ;AACN,WAAO,SAAS,KAAK;AAAA,EACvB;AACF;AAEO,SAAS,iBAAiB,MAAc,SAA2B;AACxE,QAAM,OAAO,oBAAI,IAAY;AAC7B,QAAM,OAAiB,CAAC;AAExB,QAAM,MAAM,CAAC,QAAgB;AAC3B,UAAM,WAAW,WAAW,IAAI,KAAK,GAAG,OAAO;AAC/C,QAAI,CAAC,YAAY,KAAK,IAAI,QAAQ,KAAK,QAAQ,QAAQ,EAAG;AAC1D,SAAK,IAAI,QAAQ;AACjB,SAAK,KAAK,QAAQ;AAAA,EACpB;AAEA,aAAW,KAAK,KAAK,SAAS,gBAAgB,GAAG;AAC/C,UAAM,MAAM,EAAE,CAAC;AACf,eAAW,QAAQ,CAAC,OAAO,YAAY,iBAAiB,eAAe,GAAG;AACxE,YAAM,KAAK,IAAI,MAAM,IAAI,OAAO,MAAM,IAAI,6BAA6B,GAAG,CAAC,KAAK,CAAC,GAAG,CAAC;AACrF,UAAI,EAAG,KAAI,CAAC;AAAA,IACd;AACA,UAAM,UAAU,IAAI,MAAM,kCAAkC,KAAK,CAAC,GAAG,CAAC;AACtE,QAAI,QAAQ;AACV,iBAAW,QAAQ,OAAO,MAAM,GAAG,GAAG;AACpC,cAAM,IAAI,KAAK,KAAK,EAAE,MAAM,KAAK,EAAE,CAAC;AACpC,YAAI,EAAG,KAAI,CAAC;AAAA,MACd;AAAA,IACF;AAAA,EACF;AAEA,aAAW,KAAK,KAAK,SAAS,mBAAmB,GAAG;AAClD,UAAM,MAAM,EAAE,CAAC;AACf,UAAM,OAAO,IAAI,MAAM,+BAA+B,KAAK,CAAC,GAAG,CAAC;AAChE,QAAI,IAAK,KAAI,GAAG;AAChB,UAAM,UAAU,IAAI,MAAM,kCAAkC,KAAK,CAAC,GAAG,CAAC;AACtE,QAAI,QAAQ;AACV,iBAAW,QAAQ,OAAO,MAAM,GAAG,GAAG;AACpC,cAAM,IAAI,KAAK,KAAK,EAAE,MAAM,KAAK,EAAE,CAAC;AACpC,YAAI,EAAG,KAAI,CAAC;AAAA,MACd;AAAA,IACF;AAAA,EACF;AAEA,aAAW,KAAK,KAAK,SAAS,kBAAkB,GAAG;AACjD,UAAM,MAAM,EAAE,CAAC;AACf,eAAW,QAAQ,CAAC,OAAO,QAAQ,GAAG;AACpC,YAAM,KAAK,IAAI,MAAM,IAAI,OAAO,MAAM,IAAI,6BAA6B,GAAG,CAAC,KAAK,CAAC,GAAG,CAAC;AACrF,UAAI,EAAG,KAAI,CAAC;AAAA,IACd;AAAA,EACF;AAEA,aAAW,KAAK,KAAK,SAAS,kBAAkB,GAAG;AACjD,UAAM,MAAM,EAAE,CAAC;AACf,UAAM,KAAK,IAAI,MAAM,+BAA+B,KAAK,CAAC,GAAG,CAAC;AAC9D,QAAI,EAAG,KAAI,CAAC;AAAA,EACd;AAEA,aAAW,KAAK,KAAK,SAAS,iBAAiB,GAAG;AAChD,UAAM,MAAM,EAAE,CAAC;AACf,UAAM,QAAQ,IAAI,MAAM,6CAA6C,KAAK,CAAC,GAAG,CAAC,GAAG,YAAY;AAC9F,QAAI,SAAS,cAAc,SAAS,iBAAiB;AACnD,YAAM,WAAW,IAAI,MAAM,mCAAmC,KAAK,CAAC,GAAG,CAAC;AACxE,UAAI,QAAS,KAAI,OAAO;AAAA,IAC1B;AAAA,EACF;AAEA,aAAW,KAAK,KAAK,SAAS,gEAAgE,GAAG;AAC/F,QAAI,EAAE,CAAC,CAAC;AAAA,EACV;AAEA,SAAO;AACT;AAEA,eAAsB,cACpB,KACA,SACA,UAC4E;AAC5E,QAAM,MAAM,MAAM,MAAM,KAAK;AAAA,IAC3B,SAAS,EAAE,cAAc,2CAA2C;AAAA,IACpE,QAAS,YAAY,QAAQ,IAAM;AAAA,EACrC,CAAC;AACD,MAAI,CAAC,IAAI,GAAI,OAAM,IAAI,MAAM,QAAQ,IAAI,MAAM,EAAE;AACjD,MAAI,CAAC,IAAI,KAAM,OAAM,IAAI,MAAM,qBAAqB;AAEpD,QAAM,WAAW,IAAI,QAAQ,IAAI,cAAc,GAAG,MAAM,GAAG,EAAE,CAAC,EAAE,KAAK,KAAK;AAE1E,MAAI,OAAO,KAAK,SAAS,QAAQ;AACjC,MAAI,YAAY,CAAC,QAAQ,QAAQ,GAAG;AAClC,UAAM,UAAkC;AAAA,MACtC,cAAc;AAAA,MAAQ,aAAa;AAAA,MAAQ,cAAc;AAAA,MACzD,aAAa;AAAA,MAAQ,iBAAiB;AAAA,MAAQ,cAAc;AAAA,MAC5D,aAAa;AAAA,MAAQ,cAAc;AAAA,MACnC,cAAc;AAAA,MAAQ,aAAa;AAAA,MAAQ,aAAa;AAAA,IAC1D;AACA,UAAM,MAAM,QAAQ,QAAQ;AAC5B,QAAI,IAAK,QAAO,OAAO;AAAA,EACzB;AAEA,QAAM,SAAS,kBAAkB,IAAI;AACrC,QAAM,SAAS,SAAS,QAAQ,IAAI,IAA2C,GAAG,MAAM;AAExF,QAAM,EAAE,SAAS,IAAI,MAAM,OAAO,IAAS;AAC3C,QAAM,YAAY,SAAS,IAAI,EAAE;AAEjC,SAAO,EAAE,WAAW,MAAM,WAAW,SAAS;AAChD;AAEA,eAAsB,iBACpB,MACA,SACA,UAA+B,CAAC,GACR;AACxB,QAAM,QAAW,QAAQ,SAAS,CAAC,SAAS,SAAS,OAAO;AAC5D,QAAM,WAAW,IAAI,IAAI,KAAK;AAC9B,QAAM,UAAW,iBAAiB,MAAM,OAAO;AAC/C,QAAM,aAAa,QAAQ;AAE3B,QAAM,eAAyB,CAAC;AAChC,QAAM,OAAgD,CAAC;AAEvD,aAAW,OAAO,SAAS;AACzB,UAAM,OAAO,YAAY,GAAG;AAC5B,QAAI,CAAC,QAAQ,CAAC,SAAS,IAAI,IAAI,GAAG;AAAE,mBAAa,KAAK,GAAG;AAAG;AAAA,IAAS;AACrE,SAAK,KAAK,EAAE,KAAK,KAAK,CAAC;AAAA,EACzB;AAEA,MAAI,QAAQ,cAAc,MAAM;AAC9B,WAAO;AAAA,MACL;AAAA,MACA,WAAW;AAAA,MACX,QAAQ,KAAK,IAAI,CAAC,EAAE,KAAK,KAAK,GAAG,OAAO;AAAA,QACtC;AAAA,QAAK;AAAA,QAAM,UAAU;AAAA,QAAM,UAAU,aAAa,KAAK,CAAC;AAAA,QAAG,WAAW;AAAA,QAAM,WAAW;AAAA,MACzF,EAAE;AAAA,MACF,eAAe,aAAa;AAAA,MAC5B;AAAA,IACF;AAAA,EACF;AAEA,QAAM,UAAW,MAAM;AAAE,QAAI;AAAE,aAAO,IAAI,IAAI,OAAO,EAAE,SAAS,QAAQ,UAAU,EAAE;AAAA,IAAE,QAAQ;AAAE,aAAO;AAAA,IAAU;AAAA,EAAE,GAAG;AACtH,QAAM,SAAU,oBAAI,KAAK,GAAE,YAAY,EAAE,QAAQ,SAAS,GAAG,EAAE,MAAM,GAAG,EAAE;AAC1E,QAAM,SAAU,QAAQ,aAAa,KAAK,QAAQ,GAAG,aAAa,eAAe,SAAS,GAAG,KAAK,IAAI,MAAM,EAAE;AAC9G,YAAU,QAAQ,EAAE,WAAW,KAAK,CAAC;AAErC,QAAM,SAAuB,CAAC;AAE9B,QAAM,QAAQ;AAAA,IACZ,KAAK,IAAI,OAAO,EAAE,KAAK,KAAK,GAAG,MAAM;AACnC,YAAM,WAAW,aAAa,KAAK,CAAC;AACpC,UAAI;AACF,cAAM,EAAE,WAAW,WAAW,SAAS,IAAI,MAAM,cAAc,KAAK,QAAQ,QAAQ;AACpF,cAAM,eAAe,WAAY,aAAa,QAAQ,KAAK,OAAQ;AACnE,eAAO,KAAK,EAAE,KAAK,MAAM,cAAc,UAAU,UAAU,SAAS,SAAS,GAAG,WAAW,UAAU,CAAC;AAAA,MACxG,QAAQ;AACN,eAAO,KAAK,EAAE,KAAK,MAAM,UAAU,MAAM,UAAU,WAAW,MAAM,WAAW,KAAK,CAAC;AAAA,MACvF;AAAA,IACF,CAAC;AAAA,EACH;AAEA,SAAO,KAAK,CAAC,GAAG,MAAM;AACpB,QAAI,EAAE,aAAa,CAAC,EAAE,UAAW,QAAO;AACxC,QAAI,CAAC,EAAE,aAAa,EAAE,UAAW,QAAO;AACxC,WAAO,EAAE,IAAI,cAAc,EAAE,GAAG;AAAA,EAClC,CAAC;AAED,SAAO,EAAE,SAAS,WAAW,QAAQ,QAAQ,eAAe,aAAa,QAAQ,WAAW;AAC9F;","names":[]}

package/dist/{chunk-O2S5TOCG.js → chunk-KE7KE2Q2.js}

@@ -12,14 +12,14 @@ import {
   browserServiceProfileSaveChanges,
   recordVendorUsage,
   vendorCostUsd
-} from "./chunk-D7ZT27HY.js";
+} from "./chunk-UN7VMHZL.js";
 import {
   DEFAULT_MAPS_PROXY_MODE,
   DEFAULT_PROXY_MODE
 } from "./chunk-XGIPATLV.js";
 import {
   PACKAGE_VERSION
-} from "./chunk-OQHYDW4Q.js";
+} from "./chunk-PZB3TJWK.js";
 import {
   MC_PER_CREDIT
 } from "./chunk-HL33CGJF.js";
@@ -1448,9 +1448,24 @@ function buildSeoExport(siteUrl, pages, branding) {
     branding: branding ?? void 0
   };
 }
+function formatBackgroundJobStarted(toolLabel, data) {
+  if (data.status !== "pending" || typeof data.jobId !== "string") return null;
+  const jobId = data.jobId;
+  const full = [
+    `# ${toolLabel} \u2014 export started`,
+    `**Job ID:** \`${jobId}\``,
+    `
+Running in the background \u2014 this can take a while for large sites.`,
+    `
+Poll \`check_site_export\` with this jobId to get the download link once it's ready.`
+  ].join("\n");
+  return { content: [{ type: "text", text: full }], structuredContent: { jobId, status: "pending" } };
+}
 async function formatExtractSite(raw, input, ctx) {
   const parsed = parseData(raw);
   if ("error" in parsed) return { content: [{ type: "text", text: parsed.error }], isError: true };
+  const started = formatBackgroundJobStarted("Multi-Page Site Content Crawl", parsed.data);
+  if (started) return started;
   const d = parsed.data;
   const pages = d.pages ?? [];
   const schemaTypesOf = (p) => p.kpo?.type ?? (Array.isArray(p.schema) && p.schema.length ? [`${p.schema.length} block(s)`] : []);
@@ -1568,6 +1583,8 @@ ${pageDetails}${tips}`;
 async function formatAuditSite(raw, input, ctx) {
   const parsed = parseData(raw);
   if ("error" in parsed) return { content: [{ type: "text", text: parsed.error }], isError: true };
+  const started = formatBackgroundJobStarted("Technical SEO Audit", parsed.data);
+  if (started) return started;
   const d = parsed.data;
   const pages = d.pages ?? [];
   const schemaTypesOf = (p) => p.kpo?.type ?? (Array.isArray(p.schema) && p.schema.length ? [`${p.schema.length} block(s)`] : []);
@@ -1639,6 +1656,42 @@ ${imgLine}`
   ].join("\n");
   return { content: [{ type: "text", text: full }], structuredContent };
 }
+function formatCheckSiteExport(raw, input) {
+  const parsed = parseData(raw);
+  if ("error" in parsed) return { content: [{ type: "text", text: parsed.error }], isError: true };
+  const d = parsed.data;
+  const bundle = (d.artifacts ?? []).find((a) => a.key.endsWith("bundle.zip")) ?? null;
+  const progress = d.totalUrls ? `${d.doneUrls ?? 0}/${d.totalUrls} pages` : "starting";
+  const body = d.status === "complete" && bundle ? `
+## \u2705 Ready
+**Download:** ${bundle.url}
+**Size:** ${(bundle.bytes / 1e6).toFixed(1)} MB` : d.status === "complete" ? `
+## \u26A0\uFE0F Complete, but no bundle was produced
+Check the job's artifacts \u2014 nothing matched \`bundle.zip\`.` : d.status === "failed" ? `
+## \u274C Failed
+${d.error ?? "no error message recorded"}` : `
+## \u23F3 Not ready yet
+Status: ${d.status} (${progress}). Poll again shortly.`;
+  const full = [
+    `# Site Export: ${d.startUrl ?? input.jobId}`,
+    `**Job ID:** \`${d.jobId}\``,
+    `**Status:** ${d.status}`,
+    body
+  ].join("\n");
+  return {
+    ...oneBlock(full),
+    structuredContent: {
+      jobId: d.jobId,
+      status: d.status,
+      startUrl: d.startUrl,
+      totalUrls: d.totalUrls,
+      doneUrls: d.doneUrls,
+      bundleUrl: bundle?.url ?? null,
+      bundleBytes: bundle?.bytes ?? null,
+      error: d.error ?? null
+    }
+  };
+}
 function formatYoutubeHarvest(raw, input) {
   const parsed = parseData(raw);
   if ("error" in parsed) return { content: [{ type: "text", text: parsed.error }], isError: true };
@@ -3358,13 +3411,20 @@ var ExtractSiteInputSchema = {
   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."),
   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.")
+  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."),
+  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.")
 };
 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."),
-  rotateProxyEvery: z.number().int().min(1).max(100).optional().describe("When rotateProxies is on, pages fetched per proxy before rotating. Default 30.")
+  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.")
+};
+var CheckSiteExportInputSchema = {
+  jobId: z.string().min(1).describe('The jobId returned by extract_site or audit_site when called with background (or downloadImages) set \u2014 poll this until status is "complete" (or "failed").')
 };
 var YoutubeHarvestInputSchema = {
   mode: z.enum(["search", "channel"]).describe("Use search for topic/keyword requests. Use channel when the user provides @handle, channel ID, or channel URL."),
@@ -3767,36 +3827,52 @@ var DiffPageOutputSchema = {
 };
 var ExtractSiteOutputSchema = {
   url: z.string(),
-  pageCount: z.number().int().min(0),
+  pageCount: z.number().int().min(0).optional().describe("Absent when background is true \u2014 the crawl has not finished yet."),
   pages: z.array(z.object({
     url: z.string(),
     title: NullableString,
     schemaTypes: z.array(z.string())
-  })),
-  durationMs: z.number().min(0),
+  })).optional().describe("Absent when background is true \u2014 the crawl has not finished yet."),
+  durationMs: z.number().min(0).optional().describe("Absent when background is true \u2014 the crawl has not finished yet."),
   truncatedCount: z.number().int().min(0).optional(),
-  artifact: ArtifactPointerOutputSchema.optional()
+  artifact: ArtifactPointerOutputSchema.optional(),
+  jobId: z.string().optional().describe("Present when background (or downloadImages) was set \u2014 poll with check_site_export."),
+  status: z.enum(["pending"]).optional().describe("Present when background (or downloadImages) was set."),
+  statusUrl: z.string().optional().describe("Present when background (or downloadImages) was set \u2014 informational; use check_site_export with jobId, not this URL directly.")
 };
 var AuditSiteOutputSchema = {
   url: z.string(),
-  pageCount: z.number().int().min(0),
-  durationMs: z.number().min(0),
-  bulkFolder: z.string().nullable(),
-  issues: z.record(z.string(), z.number()),
+  pageCount: z.number().int().min(0).optional().describe("Absent when background is true \u2014 the audit has not finished yet."),
+  durationMs: z.number().min(0).optional().describe("Absent when background is true \u2014 the audit has not finished yet."),
+  bulkFolder: z.string().nullable().optional().describe("Absent when background is true \u2014 the audit has not finished yet."),
+  issues: z.record(z.string(), z.number()).optional().describe("Absent when background is true \u2014 the audit has not finished yet."),
   images: z.object({
     unique: z.number().int().min(0),
     totalBytes: z.number().min(0),
     over100kb: z.number().int().min(0),
     legacyFormat: z.number().int().min(0)
-  }),
+  }).optional().describe("Absent when background is true \u2014 the audit has not finished yet."),
   links: z.object({
     internal: z.number().int().min(0),
     external: z.number().int().min(0),
     orphans: z.number().int().min(0),
     brokenInternal: z.number().int().min(0),
     externalDomains: z.number().int().min(0)
-  }),
-  artifact: ArtifactPointerOutputSchema.optional()
+  }).optional().describe("Absent when background is true \u2014 the audit has not finished yet."),
+  artifact: ArtifactPointerOutputSchema.optional(),
+  jobId: z.string().optional().describe("Present when background (or downloadImages) was set \u2014 poll with check_site_export."),
+  status: z.enum(["pending"]).optional().describe("Present when background (or downloadImages) was set."),
+  statusUrl: z.string().optional().describe("Present when background (or downloadImages) was set \u2014 informational; use check_site_export with jobId, not this URL directly.")
+};
+var CheckSiteExportOutputSchema = {
+  jobId: z.string(),
+  status: z.enum(["pending", "running", "complete", "failed"]),
+  startUrl: z.string().optional(),
+  totalUrls: z.number().int().min(0).optional(),
+  doneUrls: z.number().int().min(0).optional(),
+  bundleUrl: z.string().nullable().describe("Downloadable zip URL once status is complete; null otherwise."),
+  bundleBytes: z.number().int().min(0).nullable().describe("Zip size in bytes once status is complete; null otherwise."),
+  error: z.string().nullable().optional().describe("Present with a message when status is failed.")
 };
 var MapsPlaceIntelOutputSchema = {
   name: z.string(),
@@ -5100,6 +5176,13 @@ function registerPaaExtractorMcpTools(server, executor, options = {}) {
     outputSchema: recordOutputSchema("audit_site", AuditSiteOutputSchema),
     annotations: liveWebToolAnnotations("Technical SEO Audit")
   }, async (input) => formatAuditSite(await executor.auditSite(input), input, ctx));
+  server.registerTool("check_site_export", {
+    title: "Check Site Export",
+    description: "Poll the status of a background extract_site or audit_site job (one started with background or downloadImages set). Returns a downloadable zip URL (all page content, plus real image files if downloadImages was set) once status is complete.",
+    inputSchema: CheckSiteExportInputSchema,
+    outputSchema: recordOutputSchema("check_site_export", CheckSiteExportOutputSchema),
+    annotations: liveWebToolAnnotations("Check Site Export")
+  }, async (input) => formatCheckSiteExport(await executor.checkSiteExport(input), input));
   server.registerTool("youtube_harvest", {
     title: "YouTube Video Harvest",
     description: 'Harvest YouTube video metadata by topic search or channel library. Use mode "search" for keyword/topic requests, mode "channel" for @handles/channel IDs/URLs. Returns titles, views, durations, and videoIds.',
@@ -5592,6 +5675,9 @@ var HttpMcpToolExecutor = class {
   auditSite(input) {
     return this.call("/extract-site", input);
   }
+  checkSiteExport(input) {
+    return this.getJson(`/extract-site/status/${encodeURIComponent(input.jobId)}`);
+  }
   youtubeHarvest(input) {
     return this.call("/youtube/harvest", input);
   }
@@ -9693,4 +9779,4 @@ export {
   registerMemoryMcpTools,
   MemoryMcpToolExecutor
 };
-//# sourceMappingURL=chunk-O2S5TOCG.js.map
+//# sourceMappingURL=chunk-KE7KE2Q2.js.map