mcp-scraper 0.3.10 → 0.3.12

package/dist/{chunk-CED7X4WB.js → chunk-7R7VBQRV.js}

@@ -7,7 +7,7 @@ import {
 } from "./chunk-LFATOGDF.js";
 import {
   PACKAGE_VERSION
-} from "./chunk-NEGW2ZEJ.js";
+} from "./chunk-FB5ZJGFN.js";
 
 // src/harvest-timeout.ts
 var VERCEL_FUNCTION_MAX_MS = 3e5;
@@ -1530,7 +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 = "hosted browser";
+  const browserLabel = browser.profileName ? `hosted browser profile ${browser.profileName}` : "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}`,
@@ -1605,7 +1605,7 @@ function formatInstagramMediaDownload(raw, input) {
   const transcriptText = transcript?.text ?? "";
   const chunks = transcript?.chunks ?? [];
   const browser = structuredInstagramBrowser(d.browser);
-  const browserLabel = "hosted browser";
+  const browserLabel = browser.profileName ? `hosted browser profile ${browser.profileName}` : "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} |`;
@@ -1840,13 +1840,17 @@ var FacebookVideoTranscribeInputSchema = {
 var InstagramProfileContentInputSchema = {
   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 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."),
+  profile: z.string().min(1).optional().describe("Optional saved hosted browser profile name to load authenticated Instagram access. If omitted, the server uses its configured default profile when present."),
+  saveProfileChanges: z.boolean().optional().describe("Whether to save browser changes back to the hosted profile. Leave unset unless intentionally updating the saved login session."),
+  maxItems: z.number().int().min(1).max(2e3).default(50).describe("Maximum profile grid post/reel/tv URLs to collect. Default 50, maximum 2000. Use higher values only when the user asks for a fuller archive."),
+  maxScrolls: z.number().int().min(0).max(250).default(10).describe("Maximum pagination scroll attempts. Default 10, maximum 250. 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.")
 };
 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."),
+  profile: z.string().min(1).optional().describe("Optional saved hosted browser profile name to load authenticated Instagram access. If omitted, the server uses its configured default profile when present."),
+  saveProfileChanges: z.boolean().optional().describe("Whether to save browser changes back to the hosted profile. Leave unset unless intentionally updating the saved login session."),
   mediaTypes: z.array(z.enum(["image", "video", "audio"])).default(["image", "video", "audio"]).describe("Which media types to download when downloadMedia is true. Reels commonly expose separate video-only and audio-only MP4 tracks."),
   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."),
@@ -2258,8 +2262,8 @@ var InstagramBrowserOutput = z.object({
   executablePathConfigured: z.boolean()
 });
 var InstagramPaginationOutput = z.object({
-  maxItems: z.number().int().min(1).max(500),
-  maxScrolls: z.number().int().min(0).max(100),
+  maxItems: z.number().int().min(1).max(2e3),
+  maxScrolls: z.number().int().min(0).max(250),
   attemptedScrolls: z.number().int().min(0),
   stableScrolls: z.number().int().min(0),
   stableScrollLimit: z.number().int().min(1).max(10),
@@ -3114,6 +3118,29 @@ function youtubeVideoIdFromUrl(url) {
   }
   return null;
 }
+async function readResponseData(res) {
+  const text = await res.text();
+  if (!text.trim()) return null;
+  try {
+    return JSON.parse(text);
+  } catch {
+    return text;
+  }
+}
+function httpErrorPayload(path, res, data) {
+  const objectData = data && typeof data === "object" && !Array.isArray(data) ? data : null;
+  const bodyMessage = objectData ? objectData.message ?? objectData.error ?? objectData.error_code : data;
+  return {
+    ...objectData ?? { body: data },
+    error: objectData?.error ?? objectData?.error_code ?? "mcp_http_error",
+    error_type: "http",
+    retryable: res.status === 429 || res.status >= 500,
+    status: res.status,
+    statusText: res.statusText,
+    path,
+    message: typeof bodyMessage === "string" && bodyMessage.trim() ? bodyMessage : `MCP Scraper HTTP ${res.status}${res.statusText ? ` ${res.statusText}` : ""} for ${path}`
+  };
+}
 var HttpMcpToolExecutor = class {
   baseUrl;
   apiKey;
@@ -3141,9 +3168,9 @@ var HttpMcpToolExecutor = class {
         body: JSON.stringify(body),
         signal: AbortSignal.timeout(timeoutMs)
       });
-      const data = await res.json();
+      const data = await readResponseData(res);
       if (!res.ok) {
-        return { content: [{ type: "text", text: JSON.stringify(data) }], isError: true };
+        return { content: [{ type: "text", text: JSON.stringify(httpErrorPayload(path, res, data)) }], isError: true };
       }
       return { content: [{ type: "text", text: JSON.stringify(data) }] };
     } catch (err) {
@@ -3176,9 +3203,9 @@ var HttpMcpToolExecutor = class {
         },
         signal: AbortSignal.timeout(timeoutMs)
       });
-      const data = await res.json();
+      const data = await readResponseData(res);
       if (!res.ok) {
-        return { content: [{ type: "text", text: JSON.stringify(data) }], isError: true };
+        return { content: [{ type: "text", text: JSON.stringify(httpErrorPayload(path, res, data)) }], isError: true };
       }
       return { content: [{ type: "text", text: JSON.stringify(data) }] };
     } catch (err) {
@@ -3335,4 +3362,4 @@ export {
   registerPaaExtractorMcpTools,
   HttpMcpToolExecutor
 };
-//# sourceMappingURL=chunk-CED7X4WB.js.map
+//# sourceMappingURL=chunk-7R7VBQRV.js.map