mcpscraper-cli 0.57.2 → 0.57.4

package/README.md

@@ -198,4 +198,4 @@ This release targets MCP Scraper 0.103.0 with 352 tools. Calls use protocol `202
 
 Publication management: list owned subdomains with `commons_list_publications`, select a name with `commons_get_publication`, and edit one article with `commons_update_editorial_article`. Publishing merges omitted articles by default; use `mode:replace` for a full replacement. Use the dedicated article or publication delete tool with its current revision and a stable retry key. See the [release notes](https://github.com/VilovietaSEO/mcpscraper-sdk/releases/tag/v0.59.0).
 
-Omit the Reddit trending window for an all-time search. Dated searches verify post or comment activity, including when comment text is omitted. See the [0.59.3 release notes](https://github.com/VilovietaSEO/mcpscraper-sdk/releases/tag/v0.59.3) for the compatible package versions.
+Pass `linkResolution: deferred` to Search or a People Also Ask harvest to receive results before wrapped Google links are resolved, then read the same job again for the destination URLs. See the [0.59.5 release notes](https://github.com/VilovietaSEO/mcpscraper-sdk/releases/tag/v0.59.5) for the compatible package versions.

package/dist/bin/cli.js

@@ -10,7 +10,7 @@ import { randomUUID } from "crypto";
 import { ScraperClient, ScraperApiError } from "mcpscraper-sdk";
 
 // src/version.ts
-var CLI_VERSION = "0.57.2";
+var CLI_VERSION = "0.57.4";
 
 // src/generated-tools.ts
 var MCP_TOOL_CATALOG = [
@@ -21385,6 +21385,15 @@ var MCP_TOOL_CATALOG = [
           "default": false,
           "description": "Include the What People Are Saying social surface when present.",
           "type": "boolean"
+        },
+        "linkResolution": {
+          "default": "complete",
+          "description": "complete waits until source and result links are real destination URLs, which adds about a minute after the questions are captured. deferred finishes the job as soon as questions, answers and sources are captured: wrapped links stay as working Google redirect links marked pending, and a later harvest_paa_status read of the same jobId supplies the destination URLs at no charge.",
+          "type": "string",
+          "enum": [
+            "complete",
+            "deferred"
+          ]
         }
       },
       "required": [
@@ -21405,7 +21414,7 @@ var MCP_TOOL_CATALOG = [
     "name": "harvest_paa_status",
     "category": "other",
     "title": "Check Durable Google PAA Harvest",
-    "description": "Poll an owner-scoped harvest_paa_start job for saved answers and source descriptions, progress, recovery count, target/discovery status, material completeness, pagination, interaction outcomes, attempt outcomes, terminal rows, and billing. Wait at least 20 seconds before the first check and each later check. Unfinished answer expansion is partial. Page two adds organic results; PAA comes from page one. frontier_exhausted requires every observed eligible PAA control processed plus three healthy no-growth confirmations; interruption never means exhaustion. Polling never starts or bills another run.",
+    "description": "Poll an owner-scoped harvest_paa_start job for saved answers and source descriptions, progress, recovery count, target/discovery status, material completeness, pagination, interaction outcomes, attempt outcomes, terminal rows, and billing. Wait at least 20 seconds before the first check and each later check. Unfinished answer expansion is partial. Page two adds organic results; PAA comes from page one. frontier_exhausted requires every observed eligible PAA control processed plus three healthy no-growth confirmations; interruption never means exhaustion. For a job started with linkResolution:deferred, a later read supplies destination URLs for links that were pending. Polling never starts or bills another run.",
     "inputSchema": {
       "type": "object",
       "properties": {
@@ -25416,7 +25425,7 @@ var MCP_TOOL_CATALOG = [
     "name": "reddit_trending",
     "category": "reddit",
     "title": "Reddit Trending",
-    "description": "Find Reddit discussions through Google with DuckDuckGo fallback, then rank verified threads by observed engagement. Use reddit_thread for a known URL. Omit the window for a topic search across all time. For a dated search, a recent post or a recent captured comment can qualify a thread. Returns captured comments, observed dates and coverage. Discovery, scraped threads and returned comments are billed separately. Undated link-only results leave post dates unverified; inspect partial, resultQuality and comment coverage before treating a shortfall as complete.",
+    "description": "Find Reddit discussions or fetch them in a batch. Review mode lists URLs without opening threads. Show those URLs and wait for the user to choose, then call approve mode with the review ID and selected URLs. Reviews can be approved for one hour. If the user says auto-approve, use auto mode to search and fetch immediately. Discovery, fetched threads and returned comments are billed separately. Use reddit_thread for one known URL.",
     "inputSchema": {
       "type": "object",
       "properties": {
@@ -25427,9 +25436,35 @@ var MCP_TOOL_CATALOG = [
           "maxLength": 512
         },
         "topic": {
+          "description": "Topic to search. Required for discovery. Approve mode uses the original review topic and options.",
           "type": "string",
-          "minLength": 1,
-          "description": 'Topic to scan (e.g. "crm for small business"). Include quotation marks inside the value for an exact-phrase site search. Not a URL \u2014 pass a known thread URL to reddit_thread instead.'
+          "minLength": 1
+        },
+        "approvalMode": {
+          "description": "Use review to preview links, approve after the user selects links, or auto when the user wants immediate fetching.",
+          "type": "string",
+          "enum": [
+            "review",
+            "approve",
+            "auto"
+          ]
+        },
+        "reviewId": {
+          "description": "ID returned by review mode. Required in approve mode.",
+          "type": "string",
+          "format": "uuid",
+          "pattern": "^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$"
+        },
+        "approvedUrls": {
+          "description": "Only the URLs the user approved from this review. Required in approve mode.",
+          "minItems": 1,
+          "maxItems": 40,
+          "type": "array",
+          "items": {
+            "type": "string",
+            "minLength": 1,
+            "maxLength": 2048
+          }
         },
         "subreddit": {
           "description": 'Bare subreddit name to scope the scan to one community, e.g. "SEO" (no r/ prefix, no URL). Omit to scan all of Reddit.',
@@ -25438,7 +25473,7 @@ var MCP_TOOL_CATALOG = [
         },
         "window": {
           "default": "all",
-          "description": "Activity window. A post or captured comment must fall within the selected period. Omit for an undated topic search.",
+          "description": "Activity window. Review uses it as a search hint. Fetching verifies that a post or captured comment falls in the window.",
           "type": "string",
           "enum": [
             "week",
@@ -25450,14 +25485,14 @@ var MCP_TOOL_CATALOG = [
         },
         "maxThreads": {
           "default": 20,
-          "description": "Maximum number of threads returned. If a candidate cannot be delivered, other relevant candidates can take its place. Each delivered thread and its comments are billed separately.",
+          "description": "Maximum links shown in review mode, or threads returned when fetching.",
           "type": "integer",
           "minimum": 1,
           "maximum": 40
         },
         "includeComments": {
           "default": true,
-          "description": "Return captured comments. Dated searches open and bill for verified threads even when comments are omitted. Undated searches with comments omitted return search links without per-thread charges.",
+          "description": "Return captured comments when fetching. Review only lists links. In the default mode, undated searches with comments omitted return links without opening threads.",
           "type": "boolean"
         },
         "maxCommentsPerThread": {
@@ -25468,9 +25503,6 @@ var MCP_TOOL_CATALOG = [
           "maximum": 200
         }
       },
-      "required": [
-        "topic"
-      ],
       "$schema": "https://json-schema.org/draft/2020-12/schema"
     },
     "annotations": {
@@ -26448,6 +26480,15 @@ var MCP_TOOL_CATALOG = [
           "minimum": 1,
           "maximum": 2
         },
+        "linkResolution": {
+          "default": "complete",
+          "description": "complete waits until result links are real destination URLs, which can add 30 to 60 seconds when Google wraps them. deferred returns as soon as results are captured: wrapped links stay as working Google redirect links marked pending, and search_serp_status with the returned jobId supplies the destination URLs at no charge once they unlock, usually within a minute; the result states the wait.",
+          "type": "string",
+          "enum": [
+            "complete",
+            "deferred"
+          ]
+        },
         "recency": {
           "description": "Limit Google results to the past week or past month. Date filters require one page, use the managed browser, and return organic listings only at 35 Credits. Omit for unfiltered results.",
           "type": "string",
@@ -26474,7 +26515,7 @@ var MCP_TOOL_CATALOG = [
     "name": "search_serp_status",
     "category": "other",
     "title": "Check Durable Google SERP Lookup",
-    "description": "Read an existing durable SERP job created through the legacy HTTP job API. Ordinary search_serp calls now return directly and do not create a job to poll. Reading status never starts or bills another search.",
+    "description": "Read a saved search by jobId. For a search made with linkResolution:deferred, this supplies the destination URLs for links that were pending; wait the number of seconds the search result states, and call again if some are still pending. Also reads a durable SERP job created through the legacy HTTP job API. Reading never starts or bills another search.",
     "inputSchema": {
       "type": "object",
       "properties": {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "mcpscraper-cli",
-  "version": "0.57.2",
+  "version": "0.57.4",
   "description": "Command-line interface for the unified mcpscraper.dev MCP and REST APIs",
   "license": "MIT",
   "type": "module",
@@ -19,7 +19,7 @@
   },
   "dependencies": {
     "commander": "^12.0.0",
-    "mcpscraper-sdk": "^0.59.3"
+    "mcpscraper-sdk": "^0.59.5"
   },
   "devDependencies": {
     "@types/node": "^22.0.0"