mcp-scraper 0.2.15 → 0.2.16

package/README.md

@@ -8,7 +8,7 @@ Use the MCPB Desktop Extension for the branded Claude Desktop install, or use th
 
 MCP Scraper ships three local stdio entrypoints plus human-facing helper CLIs:
 
-- `mcp-scraper` — live web intelligence, SERP, PAA, site extraction, YouTube, Facebook, Maps, directory, rank tracker blueprint, and credit tools.
+- `mcp-scraper` — live web intelligence, SERP, PAA, site extraction, YouTube, Facebook ads and organic video transcripts, Maps, directory, rank tracker blueprint, and credit tools.
 - `browser-agent` — an agent-controlled live cloud browser with screenshots, clicks, typing, scrolling, live watch URLs, replay links, and MP4 replay download.
 - `mcp-scraper-combined` — one local MCP server that exposes both tool sets. This is the entrypoint used by the MCPB Desktop Extension.
 - `mcp-scraper-install` — a human-facing terminal installer card with the branded ASCII intro and copyable install commands. This command is safe to print because it is not an MCP stdio server.
@@ -81,7 +81,7 @@ 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` (`0.2.13`, SHA-256 `93c3c35b6a3315a1b9319761891cf0e805465b1ab19efa0252c7b8f2d9383030`). 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` (`0.2.16`, SHA-256 `a581f0ddb12d3cebd03d90dfb657cc9127a1f7ef995fb0934d5428faf8aec58b`). 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 the same web-intelligence tools as `mcp-scraper` plus all `browser_*` tools from `browser-agent` through one server.
 
@@ -154,8 +154,8 @@ env = { MCP_SCRAPER_API_KEY = "sk_live_your_key" }
 - `youtube_transcribe`
 - `facebook_ad_search`
 - `facebook_page_intel`
-- `facebook_ad_transcribe`
-- `facebook_video_transcribe` — transcribe an organic Facebook reel, video, watch, or share URL. The tool renders the page, extracts the best public Facebook CDN MP4 URL, then returns transcript text, timestamped chunks, and the extracted MP4 URL for follow-up download.
+- `facebook_ad_transcribe` — transcribe a direct Facebook ad video URL returned by `facebook_page_intel`.
+- `facebook_video_transcribe` — transcribe an organic Facebook reel, video, watch, post, or share URL, including `fb.watch` links. The tool renders the page, extracts the best matching public Facebook CDN MP4 URL, then returns transcript text, timestamped chunks, selected quality, video metadata, and the extracted MP4 URL for follow-up download.
 - `maps_search` — search Google Maps for multiple business/profile candidates. Use for GMB/GBP prospect lists, competitors, categories, and anything needing more than the Google 3-pack. In default `proxyMode: "location"`, retryable failures rotate to a new residential proxy and new browser session for up to 5 attempts. `maxResults` defaults to 10 and is capped at 50.
 - `maps_place_intel` — hydrate one known/named Google Maps business with profile details and optional reviews. Use after `maps_search` when a selected candidate needs full details.
 - `directory_workflow` — build city-by-city directory/prospecting datasets from Census place selection plus Google Maps searches. Use it for requests like "all cities over 100k population in Tennessee, then get 20 roofers from Maps." In default `proxyMode: "location"`, each city search rotates retryable failures to a new residential proxy and new browser session for up to 5 attempts. The saved CSV includes `source_location`, `result_position`, `business_name`, `review_stars`, `review_count`, `category`, `address`, `phone`, `hours_status`, `website_url`, `directions_url`, `place_url`, `cid`, `cid_decimal`, Census population, and ZIP groups.

package/dist/bin/api-server.cjs

@@ -14356,7 +14356,8 @@ ${adBlocks}`,
     `
 ---
 \u{1F4A1} **Tips**
-- Transcribe video ads: use \`facebook_ad_transcribe\` with the \`videoUrl\` above
+- Transcribe video ads: use \`facebook_ad_transcribe\` with the direct \`videoUrl\` above
+- Transcribe organic Facebook reels/posts: use \`facebook_video_transcribe\` with the public Facebook URL
 - Find other advertisers: use \`facebook_ad_search\``
   ].filter(Boolean).join("\n");
   return {
@@ -14802,7 +14803,7 @@ ${text}`,
 ${chunkRows}` : "",
     `
 ---
-\u{1F4A1} Get more ads from this advertiser: use \`facebook_page_intel\``
+\u{1F4A1} Get more ads from this advertiser: use \`facebook_page_intel\`. For public Facebook reel/post URLs, use \`facebook_video_transcribe\`.`
   ].filter(Boolean).join("\n");
   return oneBlock(full);
 }
@@ -14838,7 +14839,7 @@ ${text}`,
 ${chunkRows}` : "",
     `
 ---
-\u{1F4A1} To download the extracted MP4, call the Facebook media endpoint with the extracted MP4 URL.`
+\u{1F4A1} Use \`videoUrl\` as the extracted MP4 for download or follow-up processing. Use \`facebook_ad_transcribe\` only for Ad Library videoUrl values.`
   ].filter(Boolean).join("\n");
   return {
     ...oneBlock(full),
@@ -20145,7 +20146,7 @@ var PACKAGE_VERSION;
 var init_version = __esm({
   "src/version.ts"() {
     "use strict";
-    PACKAGE_VERSION = "0.2.15";
+    PACKAGE_VERSION = "0.2.16";
   }
 });
 
@@ -20205,10 +20206,10 @@ var init_mcp_tool_schemas = __esm({
       maxResults: import_zod26.z.number().int().min(1).max(20).default(10)
     };
     FacebookAdTranscribeInputSchema = {
-      videoUrl: import_zod26.z.string().url().describe("Facebook CDN video URL from a facebook_page_intel result")
+      videoUrl: import_zod26.z.string().url().describe("Direct Facebook CDN video URL from a facebook_page_intel ad result. Do not pass a public Facebook reel/post/share URL here; use facebook_video_transcribe for organic Facebook URLs.")
     };
     FacebookVideoTranscribeInputSchema = {
-      url: import_zod26.z.string().url().describe("Organic Facebook reel, video, watch, or share URL. The tool renders the page, extracts the best public Facebook CDN MP4 URL, then transcribes it."),
+      url: import_zod26.z.string().url().describe("Organic Facebook reel, video, watch, post, or share URL from facebook.com, m.facebook.com, or fb.watch. The tool renders the page, extracts the best matching public Facebook CDN MP4 URL, then transcribes it. Use this when the user pastes a normal Facebook video page URL and asks for the transcript or downloadable MP4."),
       quality: import_zod26.z.enum(["best", "hd", "sd"]).default("best").describe("Preferred progressive MP4 quality. Use best by default; hd prefers the highest HD progressive URL; sd forces the SD URL.")
     };
     MapsPlaceIntelInputSchema = {
@@ -21083,7 +21084,7 @@ function registerPaaExtractorMcpTools(server, executor, options = {}) {
   }, async (input) => formatYoutubeTranscribe(await executor.youtubeTranscribe(input), input));
   server.registerTool("facebook_page_intel", {
     title: "Facebook Advertiser Ad Intel",
-    description: withReportNote("Harvest ads from a Facebook advertiser. Returns ad copy, headlines, CTAs, creative type, status, landing URLs, and video URLs ready for transcription. Accepts pageId, libraryId, or a brand/advertiser name as query. Use after facebook_ad_search when possible."),
+    description: withReportNote("Harvest ads from a Facebook advertiser. Returns ad copy, headlines, CTAs, creative type, status, landing URLs, and direct ad video URLs ready for facebook_ad_transcribe. Accepts pageId, libraryId, or a brand/advertiser name as query. Use after facebook_ad_search when possible. For normal public Facebook reels/posts/watch/share URLs, use facebook_video_transcribe instead."),
     inputSchema: FacebookPageIntelInputSchema,
     outputSchema: FacebookPageIntelOutputSchema,
     annotations: liveWebToolAnnotations("Facebook Advertiser Ad Intel")
@@ -21097,13 +21098,13 @@ function registerPaaExtractorMcpTools(server, executor, options = {}) {
   }, async (input) => formatFacebookAdSearch(await executor.facebookAdSearch(input), input));
   server.registerTool("facebook_ad_transcribe", {
     title: "Facebook Ad Transcription",
-    description: "Transcribe audio from a Facebook ad video. Returns full transcript and timestamped chunks. Use the videoUrl value from facebook_page_intel results.",
+    description: "Transcribe audio from a Facebook ad video CDN URL. Returns full transcript and timestamped chunks. Use only with the direct videoUrl value from facebook_page_intel results, not public Facebook post/reel/share URLs.",
     inputSchema: FacebookAdTranscribeInputSchema,
     annotations: liveWebToolAnnotations("Facebook Ad Transcription")
   }, async (input) => formatFacebookAdTranscribe(await executor.facebookAdTranscribe(input), input));
   server.registerTool("facebook_video_transcribe", {
     title: "Facebook Organic Video Transcription",
-    description: withReportNote("Transcribe audio from an organic Facebook reel, video, watch, or share URL. Renders the Facebook page, extracts the best public Facebook CDN MP4 URL from the page state, then returns the full transcript, timestamped chunks, and extracted MP4 URL."),
+    description: withReportNote("Transcribe audio from an organic Facebook reel, video, watch, post, or share URL, including fb.watch links. Use this when the user pastes a normal Facebook video page URL and wants the transcript or downloadable MP4. Renders the Facebook page in a browser, selects the best matching public Facebook CDN MP4 URL from page state, then returns sourceUrl, resolved pageUrl, videoId, ownerName, selectedQuality, bitrate, videoDurationSec, extracted MP4 URL, full transcript, and timestamped chunks."),
     inputSchema: FacebookVideoTranscribeInputSchema,
     outputSchema: FacebookVideoTranscribeOutputSchema,
     annotations: liveWebToolAnnotations("Facebook Organic Video Transcription")
@@ -22914,7 +22915,7 @@ __export(server_exports, {
   app: () => app
 });
 function configuredOrigins() {
-  const origins = /* @__PURE__ */ new Set(["https://mcpscraper.dev"]);
+  const origins = /* @__PURE__ */ new Set(["https://mcpscraper.dev", "https://www.mcpscraper.dev"]);
   for (const raw of (process.env.ALLOWED_ORIGINS ?? process.env.APP_ORIGIN ?? "").split(",")) {
     const trimmed = raw.trim();
     if (trimmed) origins.add(trimmed);