mcp-scraper 0.3.6 → 0.3.7

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

@@ -2,10 +2,10 @@
 import {
   HttpMcpToolExecutor,
   buildPaaExtractorMcpServer
-} from "../chunk-ROZEDHAA.js";
+} from "../chunk-2O7RAK6E.js";
 import "../chunk-M2S27J6Z.js";
 import "../chunk-LFATOGDF.js";
-import "../chunk-IHXHQL53.js";
+import "../chunk-2NEVRPK3.js";
 
 // bin/mcp-stdio-server.ts
 import { readFileSync } from "fs";

package/dist/chunk-2NEVRPK3.js

@@ -0,0 +1,7 @@
+// src/version.ts
+var PACKAGE_VERSION = "0.3.7";
+
+export {
+  PACKAGE_VERSION
+};
+//# sourceMappingURL=chunk-2NEVRPK3.js.map

package/dist/chunk-2NEVRPK3.js.map

@@ -0,0 +1 @@
+{"version":3,"sources":["../src/version.ts"],"sourcesContent":["export const PACKAGE_VERSION = '0.3.7'\n"],"mappings":";AAAO,IAAM,kBAAkB;","names":[]}

package/dist/{chunk-ROZEDHAA.js → chunk-2O7RAK6E.js}

@@ -7,7 +7,7 @@ import {
 } from "./chunk-LFATOGDF.js";
 import {
   PACKAGE_VERSION
-} from "./chunk-IHXHQL53.js";
+} from "./chunk-2NEVRPK3.js";
 
 // src/harvest-timeout.ts
 var VERCEL_FUNCTION_MAX_MS = 3e5;
@@ -1482,16 +1482,13 @@ ${chunkRows}` : "",
 }
 function structuredInstagramBrowser(raw) {
   const browser = raw && typeof raw === "object" ? raw : {};
-  const mode = browser.mode === "local" ? "local" : "hosted";
-  const requestedMode = browser.requestedMode === "local" || browser.requestedMode === "hosted" || browser.requestedMode === "auto" ? browser.requestedMode : "auto";
-  const profileSource = browser.profileSource === "managed_profile" || browser.profileSource === "direct_profile_dir" ? browser.profileSource : "hosted";
   return {
-    mode,
-    requestedMode,
+    mode: "hosted",
+    requestedMode: "hosted",
     profileName: typeof browser.profileName === "string" ? browser.profileName : null,
-    profileSource,
-    profileDirConfigured: browser.profileDirConfigured === true,
-    executablePathConfigured: browser.executablePathConfigured === true
+    profileSource: "hosted",
+    profileDirConfigured: false,
+    executablePathConfigured: false
   };
 }
 function structuredInstagramPagination(raw, input) {
@@ -1533,7 +1530,7 @@ function formatInstagramProfileContent(raw, input) {
   const itemRows = items.slice(0, 100).map(
     (item, i) => `| ${i + 1} | ${item.type} | \`${item.shortcode}\` | ${item.url} | ${cell(item.firstSeenStage ?? "")} |`
   ).join("\n");
-  const browserLabel = browser.mode === "local" ? `local profile ${browser.profileName ? `\`${browser.profileName}\`` : "(unnamed)"}` : "hosted browser";
+  const browserLabel = "hosted browser";
   const full = [
     `# Instagram Profile Content: ${d.handle ?? input.handle ?? input.url ?? "profile"}`,
     `**Collected:** ${items.length} items \xB7 posts ${typeCounts?.post ?? 0} \xB7 reels ${typeCounts?.reel ?? 0} \xB7 tv ${typeCounts?.tv ?? 0}`,
@@ -1608,7 +1605,7 @@ function formatInstagramMediaDownload(raw, input) {
   const transcriptText = transcript?.text ?? "";
   const chunks = transcript?.chunks ?? [];
   const browser = structuredInstagramBrowser(d.browser);
-  const browserLabel = browser.mode === "local" ? `local profile ${browser.profileName ? `\`${browser.profileName}\`` : "(unnamed)"}` : "hosted browser";
+  const browserLabel = "hosted browser";
   const downloadRows = downloads.map((download, i) => {
     const status = download.error ? `error: ${cell(download.error)}` : `${download.sizeBytes ?? 0} bytes`;
     return `| ${i + 1} | ${download.kind} | ${download.savedPath ? `\`${download.savedPath}\`` : "\u2014"} | ${status} |`;
@@ -1841,14 +1838,12 @@ var FacebookVideoTranscribeInputSchema = {
   quality: 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.")
 };
 var InstagramProfileContentInputSchema = {
-  handle: z.string().min(1).optional().describe("Instagram handle, with or without @. Provide handle or url. Use only public profiles unless a configured browser profile is available server-side."),
+  handle: z.string().min(1).optional().describe("Instagram handle, with or without @. Provide handle or url."),
   url: z.string().url().optional().describe("Instagram profile URL, e.g. https://www.instagram.com/nasaartemis/. Provide handle or url."),
-  maxItems: z.number().int().min(1).max(500).default(50).describe("Maximum profile grid post/reel/tv URLs to collect. Default 50, maximum 500. Use higher values with an authenticated browser profile when the user asks for a fuller archive."),
-  maxScrolls: z.number().int().min(0).max(100).default(10).describe("Maximum pagination scroll attempts. Default 10, maximum 100. Increase when using authenticated Chromium and the profile has a long history."),
+  maxItems: z.number().int().min(1).max(500).default(50).describe("Maximum profile grid post/reel/tv URLs to collect. Default 50, maximum 500. Use higher values only when the user asks for a fuller archive."),
+  maxScrolls: z.number().int().min(0).max(100).default(10).describe("Maximum pagination scroll attempts. Default 10, maximum 100. Increase for long profiles when Instagram continues loading more grid links."),
   scrollDelayMs: z.number().int().min(250).max(5e3).default(1200).describe("Delay after each pagination scroll before collecting newly loaded links. Default 1200ms. Increase to 2000-3000ms when Instagram loads slowly."),
-  stableScrollLimit: z.number().int().min(1).max(10).default(4).describe("Stop after this many consecutive scrolls with no new links or scroll progress. Default 4."),
-  browserMode: z.enum(["auto", "hosted", "local"]).default("auto").describe("Browser backend. auto uses the hosted browser unless local mode is configured or browserProfile is passed. local uses an imported/authenticated Chrome profile from browser_profile_import/sync. hosted forces the remote browser path."),
-  browserProfile: z.string().min(1).optional().describe("Managed local browser profile name created by browser_profile_import/sync. Passing this opts the Instagram tool into local authenticated Chromium for this call.")
+  stableScrollLimit: z.number().int().min(1).max(10).default(4).describe("Stop after this many consecutive scrolls with no new links or scroll progress. Default 4.")
 };
 var InstagramMediaDownloadInputSchema = {
   url: z.string().url().describe("Instagram post, reel, or tv URL, e.g. https://www.instagram.com/reel/SHORTCODE/. The tool renders the page, extracts text, image metadata, and Instagram CDN media tracks."),
@@ -1856,9 +1851,7 @@ var InstagramMediaDownloadInputSchema = {
   downloadMedia: z.boolean().default(true).describe("Download extracted text/media files to the MCP Scraper output directory when the API server can write files. Always returns extracted media URLs even when false."),
   downloadAllTracks: z.boolean().default(false).describe("Download every captured Instagram MP4 track instead of only the selected best video and audio tracks. Use false by default to avoid duplicate bitrates."),
   includeTranscript: z.boolean().default(false).describe("Transcribe the selected audio track when available. This adds transcription cost and may take longer."),
-  mux: z.boolean().default(true).describe("When video and audio tracks are downloaded separately, try to mux them into a single MP4 if ffmpeg is available. Returns separate tracks when muxing is unavailable."),
-  browserMode: z.enum(["auto", "hosted", "local"]).default("auto").describe("Browser backend. auto uses the hosted browser unless local mode is configured or browserProfile is passed. local uses an imported/authenticated Chrome profile from browser_profile_import/sync. hosted forces the remote browser path."),
-  browserProfile: z.string().min(1).optional().describe("Managed local browser profile name created by browser_profile_import/sync. Passing this opts the Instagram tool into local authenticated Chromium for this call.")
+  mux: z.boolean().default(true).describe("When video and audio tracks are downloaded separately, try to mux them into a single MP4 if ffmpeg is available. Returns separate tracks when muxing is unavailable.")
 };
 var MapsPlaceIntelInputSchema = {
   businessName: z.string().min(1).describe('Business name only. If user says "Elite Roofing Denver CO", use businessName="Elite Roofing" and location="Denver, CO".'),
@@ -2257,10 +2250,10 @@ var TranscriptChunkOutput = z.object({
   text: z.string()
 });
 var InstagramBrowserOutput = z.object({
-  mode: z.enum(["hosted", "local"]),
-  requestedMode: z.enum(["auto", "hosted", "local"]),
+  mode: z.literal("hosted"),
+  requestedMode: z.literal("hosted"),
   profileName: NullableString,
-  profileSource: z.enum(["hosted", "managed_profile", "direct_profile_dir"]),
+  profileSource: z.literal("hosted"),
   profileDirConfigured: z.boolean(),
   executablePathConfigured: z.boolean()
 });
@@ -3002,14 +2995,14 @@ function registerPaaExtractorMcpTools(server, executor, options = {}) {
   }, async (input) => formatFacebookVideoTranscribe(await executor.facebookVideoTranscribe(input), input));
   server.registerTool("instagram_profile_content", {
     title: "Instagram Profile Content Discovery",
-    description: withReportNote("Discover Instagram profile grid content links for a handle or profile URL. Use this when the user wants a person or brand account content inventory before selecting posts/reels to download. Returns profile stats, collected post/reel/tv URLs, shortcodes, type counts, browser mode, pagination attempts, stop reason, and limitations. For authenticated collection, first use browser_profile_import/sync, then pass browserProfile and raise maxItems/maxScrolls for longer profile pagination."),
+    description: withReportNote("Discover Instagram profile grid content links for a handle or profile URL. Use this when the user wants a person or brand account content inventory before selecting posts/reels to download. Returns profile stats, collected post/reel/tv URLs, shortcodes, type counts, browser details, pagination attempts, stop reason, and limitations."),
     inputSchema: InstagramProfileContentInputSchema,
     outputSchema: InstagramProfileContentOutputSchema,
     annotations: liveWebToolAnnotations("Instagram Profile Content Discovery")
   }, async (input) => formatInstagramProfileContent(await executor.instagramProfileContent(input), input));
   server.registerTool("instagram_media_download", {
     title: "Instagram Post/Reel Media Download",
-    description: withReportNote("Extract and download media from one Instagram post, reel, or tv URL. Use after instagram_profile_content or when the user gives a specific Instagram URL and wants the image, caption/text, reel audio/video tracks, optional muxed MP4, or optional transcript. Reels commonly expose separate video-only and audio-only MP4 tracks; this tool selects the best video and audio tracks and attempts muxing when ffmpeg is available. Pass browserProfile from browser_profile_import/sync when the media is login-gated."),
+    description: withReportNote("Extract and download media from one Instagram post, reel, or tv URL. Use after instagram_profile_content or when the user gives a specific Instagram URL and wants the image, caption/text, reel audio/video tracks, optional muxed MP4, or optional transcript. Reels commonly expose separate video-only and audio-only MP4 tracks; this tool selects the best video and audio tracks and attempts muxing when ffmpeg is available."),
     inputSchema: InstagramMediaDownloadInputSchema,
     outputSchema: InstagramMediaDownloadOutputSchema,
     annotations: liveWebToolAnnotations("Instagram Post/Reel Media Download")
@@ -3342,4 +3335,4 @@ export {
   registerPaaExtractorMcpTools,
   HttpMcpToolExecutor
 };
-//# sourceMappingURL=chunk-ROZEDHAA.js.map
+//# sourceMappingURL=chunk-2O7RAK6E.js.map