mcp-scraper 0.57.0 → 0.57.2

package/CHANGELOG.md

@@ -4,6 +4,22 @@ All notable changes to MCP Scraper are documented here. The format is based on [
 
 ## [Unreleased]
 
+## [0.57.2] - 2026-08-16
+
+### Fixed
+
+- Blob storage no longer falls back to the local filesystem on a serverless deployment. `BLOB_READ_WRITE_TOKEN` was empty in production, so `getBlobStore()` returned a local store and every write through it — the scrape-vault fallback document and Instagram media downloads — was saved to an ephemeral path inside the function and handed back as a `file://` URL no request could read. Without a token a serverless deployment now returns a store that refuses to write and says why.
+
+## [0.57.1] - 2026-08-16
+
+### Fixed
+
+- Commons image hosting resolves the dedicated public store token and refuses to store anything unless the resolved store is Vercel Blob. `BLOB_READ_WRITE_TOKEN` is empty in production, so the previous release would have written a `file://` path into a public wiki entity instead of failing; hosting now fails loudly with `image_storage_unconfigured`.
+
+### Added
+
+- `scripts/backfill-commons-images.ts` hosts the still-reachable featured images and article media on Commons storage, rewrites the entity, and records a contribution-ledger entry per rewrite. Dry run by default; `--apply` to write. Run against production: 15 featured images and 20 media sets moved onto Commons hosting; 26 sources were already dead 404s and one SVG is rejected by policy.
+
 ## [0.57.0] - 2026-08-16
 
 ### Fixed

package/README.md

@@ -90,7 +90,7 @@ 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` (`0.57.0`, SHA-256 `9d85d704c6c961f69b42bcdc015627abfb6178e2dc832712023a112ac858e53e`). 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` (`0.57.2`, SHA-256 `1c7ae1f1acf54fc91bc44a8dae29b1da20542a63b168f496ae91a67be9caaf33`). 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

@@ -37303,16 +37303,21 @@ function byteLength(data) {
 function localBaseDir2() {
   return process.env.MCP_SCRAPER_OUTPUT_DIR?.trim() || (0, import_node_path11.join)((0, import_node_os7.homedir)(), "Downloads", "mcp-scraper");
 }
+function runsWithoutLocalDisk(env = process.env) {
+  return Boolean(env.VERCEL || env.AWS_LAMBDA_FUNCTION_NAME);
+}
+function resolveBlobStore(env = process.env) {
+  const token6 = env.BLOB_READ_WRITE_TOKEN?.trim();
+  if (token6) return new VercelBlobStore(token6);
+  if (runsWithoutLocalDisk(env)) return new UnconfiguredBlobStore();
+  return new LocalBlobStore(localBaseDir2());
+}
 function getBlobStore() {
   if (cached) return cached;
-  if (process.env.BLOB_READ_WRITE_TOKEN) {
-    cached = new VercelBlobStore(process.env.BLOB_READ_WRITE_TOKEN);
-  } else {
-    cached = new LocalBlobStore(localBaseDir2());
-  }
+  cached = resolveBlobStore();
   return cached;
 }
-var import_node_fs7, import_node_os7, import_node_path11, LocalBlobStore, cached, VercelBlobStore;
+var import_node_fs7, import_node_os7, import_node_path11, LocalBlobStore, cached, UnconfiguredBlobStore, VercelBlobStore;
 var init_blob_store = __esm({
   "src/api/blob-store.ts"() {
     "use strict";
@@ -37342,6 +37347,15 @@ var init_blob_store = __esm({
       }
     };
     cached = null;
+    UnconfiguredBlobStore = class {
+      kind = "unconfigured";
+      async put() {
+        throw new Error("Blob storage is not configured on this deployment: set BLOB_READ_WRITE_TOKEN. Refusing to write to a local path that no request can read back.");
+      }
+      async get() {
+        return null;
+      }
+    };
     VercelBlobStore = class {
       constructor(token6) {
         this.token = token6;
@@ -44624,7 +44638,7 @@ var PACKAGE_VERSION;
 var init_version = __esm({
   "src/version.ts"() {
     "use strict";
-    PACKAGE_VERSION = "0.57.0";
+    PACKAGE_VERSION = "0.57.2";
   }
 });
 
@@ -64399,8 +64413,12 @@ function decodeBase64Image(value) {
   }
   return bytes;
 }
+function commonsImageBlobStore() {
+  const token6 = process.env.COMMONS_IMAGE_READ_WRITE_TOKEN?.trim() || process.env.COMMONS_IMAGE_BLOB_READ_WRITE_TOKEN?.trim() || process.env.BLOB_READ_WRITE_TOKEN?.trim();
+  return token6 ? new VercelBlobStore(token6) : getBlobStore();
+}
 function isCommonsHostedImageUrl(url) {
-  return /^https:\/\/[^/]*(\.public\.blob\.vercel-storage\.com|blob\.vercel-storage\.com)\//i.test(url) || url.startsWith("file://");
+  return /^https:\/\/[^/]*(\.public\.blob\.vercel-storage\.com|blob\.vercel-storage\.com)\//i.test(url);
 }
 async function hostCommonsImage(input) {
   if (hostOverride) return hostOverride(input);
@@ -64436,7 +64454,15 @@ async function hostCommonsImage(input) {
   }
   const extension = contentType.split("/")[1] === "jpeg" ? "jpg" : contentType.split("/")[1];
   const storageKey = `commons/images/${digest2.slice(0, 2)}/${digest2}.${extension}`;
-  const stored = await getBlobStore().put(storageKey, bytes, contentType);
+  const store = commonsImageBlobStore();
+  if (store.kind !== "vercel-blob") {
+    throw new CommonsImageError(
+      "image_storage_unconfigured",
+      "Commons image hosting needs a Vercel Blob token. Refusing to write a local file path into a public wiki entity.",
+      422
+    );
+  }
+  const stored = await store.put(storageKey, bytes, contentType);
   const id = `TPW-IMG-${digest2.slice(0, 16).toUpperCase()}`;
   await getDb().execute({
     sql: `