mcp-scraper 0.13.0 → 0.14.0

package/README.md

@@ -88,7 +88,7 @@ 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` (`0.13.0`, SHA-256 `f94c993e9e3b0cb6d56809e134f556460dfad5835cc47a0bfa63c19dcdf7115b`). 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` (`0.14.0`, SHA-256 `01b37c065b803a19ac824b82058b311b115e272d9be01d870094737c74b58753`). 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.
 
@@ -174,6 +174,17 @@ env = { MCP_SCRAPER_API_KEY = "sk_live_your_key" }
 - `rank_tracker_workflow` — generate a database schema, cron/heartbeat plan, ingestion workflow, metrics list, and implementation prompt for building rank trackers. It has modes for Maps rankings via `directory_workflow`/`maps_search`, organic rankings via `search_serp`, AI Overview citation tracking, and PAA source presence tracking. This planning tool does not spend credits.
 - `credits_info`
 
+### Connected-account tools
+
+- `list_service_connections` — list this caller's tenant-owned OAuth connections and the exact live read/action tools available on each account. Connections are never shared between customers.
+- `export_connected_service_data` — fetch a bounded Gmail, Google Calendar, or Zoom time range in one MCP call. The server handles provider pagination, Gmail hydration, normalization, and continuation. Small exports return inline; larger exports become private JSONL retained for seven days with a 15-minute signed URL. Attachments remain metadata-only and oversized individual bodies/transcripts are truncated with warnings.
+- `renew_connected_data_download` — issue a fresh 15-minute signed URL for an unexpired private export artifact without pulling the provider again.
+- `read_service_connection` — run one small live read by exact allowlisted name. Do not loop it over a time range; use `export_connected_service_data` for email, calendar, recording, or transcript collections.
+- `call_service_connection_action` — run one exact mutation only after the account action switch is enabled. GitHub, Gmail, Calendar, Slack, Zoom, Facebook, Google Analytics, X, and YouTube actions remain provider-policy and tenant gated.
+- `set_scheduled_action_connections` — bind exact tenant connections and exact tool grants to one scheduled action. Deterministic connection syncs are model-free and one schedule occurrence counts as one metered event even when provider pagination spans many pages.
+
+GitHub uses the same provider-neutral read/action bridges: 27 reads cover repository discovery, issues, pull requests, commits, releases, and workflows; 20 mutations remain gated. Private-repository and workflow access requires reconnecting the OAuth account after the expanded scope set is deployed.
+
 ### Browser-agent tools
 
 - `browser_open` — open a direct/no-proxy hosted browser session. Pass a saved profile name to load an authenticated session. Returns a human `watch_url`.

package/dist/bin/api-server.cjs

@@ -27642,7 +27642,7 @@ var PACKAGE_VERSION;
 var init_version = __esm({
   "src/version.ts"() {
     "use strict";
-    PACKAGE_VERSION = "0.13.0";
+    PACKAGE_VERSION = "0.14.0";
   }
 });
 
@@ -29677,7 +29677,7 @@ function registerPaaExtractorMcpTools(server, executor, options = {}) {
   }, async (input) => formatCreditsInfo(await executor.creditsInfo(input), input));
   server.registerTool("list_service_connections", {
     title: "List Connected Services",
-    description: "List every third-party service connection this MCP Scraper account has authorized, including Google Analytics, YouTube, Facebook Pages, LinkedIn, X, Meta Marketing, Slack, Gmail, Calendar, Drive, Zoom, Xero, and others. Returns the tenant-scoped connectionId plus exact readTools and actionTools. Get a connectionId and exact tool name here before calling read_service_connection or call_service_connection_action. For already-digested history, prefer the returned vaultName or tableName.",
+    description: "List every third-party service connection this MCP Scraper account has authorized, including GitHub, Google Analytics, YouTube, Facebook Pages, LinkedIn, X, Meta Marketing, Slack, Gmail, Calendar, Drive, Zoom, Xero, and others. Returns the tenant-scoped connectionId plus exact readTools and actionTools. Get a connectionId and exact tool name here before calling read_service_connection or call_service_connection_action. GitHub repository, issue, pull-request, release, and workflow operations use these same provider-neutral bridges; mutations still require the account action switch and an exact allowed action. For already-digested history, prefer the returned vaultName or tableName.",
     inputSchema: ListServiceConnectionsInputSchema,
     outputSchema: recordOutputSchema("list_service_connections", ListServiceConnectionsOutputSchema),
     annotations: { title: "List Connected Services", readOnlyHint: true, destructiveHint: false, idempotentHint: true, openWorldHint: false }
@@ -29712,7 +29712,7 @@ function registerPaaExtractorMcpTools(server, executor, options = {}) {
   }, async (input) => executor.zoomCreateMeeting(input));
   server.registerTool("read_service_connection", {
     title: "Read Connected Service",
-    description: "Call one small live, read-only operation on any connected service. Do not loop this tool to fetch a time range or collection: use export_connected_service_data for emails, calendar events, Zoom recordings, or transcripts. Requires a connectionId and an exact name from that connection's readTools in list_service_connections; an unlisted tool is rejected server-side.",
+    description: "Call one small live, read-only operation on any connected service, including GitHub repository, issue, pull-request, release, and workflow reads. Do not loop this tool to fetch a time range or collection: use export_connected_service_data for emails, calendar events, Zoom recordings, or transcripts. Requires a connectionId and an exact name from that connection's readTools in list_service_connections; an unlisted tool is rejected server-side.",
     inputSchema: ReadServiceConnectionInputSchema,
     outputSchema: recordOutputSchema("read_service_connection", ReadServiceConnectionOutputSchema),
     annotations: { title: "Read Connected Service", readOnlyHint: true, destructiveHint: false, idempotentHint: true, openWorldHint: true }
@@ -29733,7 +29733,7 @@ function registerPaaExtractorMcpTools(server, executor, options = {}) {
   }, async (input) => executor.renewConnectedDataDownload(input));
   server.registerTool("call_service_connection_action", {
     title: "Run Connected Service Action",
-    description: "Run one explicitly allowlisted write or mutation on a tenant-owned OAuth connection. First call list_service_connections, use a connection with actionsEnabled true, choose one exact actionTools entry, and supply that action's arguments. The server rejects arbitrary action names, inactive or foreign connections, disabled actions, and tools outside the provider allowlist. This may publish, update, send, subscribe, or delete provider data depending on the chosen tool.",
+    description: "Run one explicitly allowlisted write or mutation on a tenant-owned OAuth connection. First call list_service_connections, use a connection with actionsEnabled true, choose one exact actionTools entry, and supply that action's arguments. The server rejects arbitrary action names, inactive or foreign connections, disabled actions, and tools outside the provider allowlist. This includes GitHub issue, pull-request, repository-content, release, and workflow actions when the exact action is exposed; delete, merge, review-submission, workflow-execution, and content-changing operations are high impact. This may publish, update, send, subscribe, merge, run, or delete provider data depending on the chosen tool.",
     inputSchema: CallServiceConnectionActionInputSchema,
     outputSchema: recordOutputSchema("call_service_connection_action", CallServiceConnectionActionOutputSchema),
     annotations: { title: "Run Connected Service Action", readOnlyHint: false, destructiveHint: true, idempotentHint: false, openWorldHint: true }
@@ -39773,6 +39773,15 @@ var init_nango_control = __esm({
         "list-recordings",
         "get-recording",
         "get-meeting-transcript"
+      ],
+      "github-getting-started": [
+        "list-repositories",
+        "list-issues",
+        "list-pull-requests",
+        "list-commits",
+        "list-releases",
+        "list-workflows",
+        "list-workflow-runs"
       ]
     };
     NangoControlError = class extends Error {