get_reviews in Claude

Claude pulls store reviews through Trends MCP get_reviews after the connector is added. claude.ai uses OAuth at https://www.trendsmcp.ai/mcp with no API key in the connector. Desktop uses npx mcp-remote against https://api.trendsmcp.ai/mcp with a Bearer key. The homepage fetched 2026-11-10 still tells Claude Code to add the www host and sign in with /mcp. llms.txt and Trends MCP for Claude Code from the same morning still add the api host with a Bearer header. Ask, using TrendsMCP, for source App Store, Google Play, or Amazon plus a brand, app, product, or place name. Omit limit and Google Play still prints count 10 over window_days 90. App Store can match the listing and still print items [] with error null. That empty array billed 1 request on 2026-11-10. The live MCP list in this session names get_reviews and does not name get_top_posts.

Three Claude hosts, then name get_reviews

Host setup for the browser and Desktop lives on Trends MCP for Claude. The terminal add lives on Trends MCP for Claude Code. Those pages prove the connect with get_top_trends. This page is the review-shaped call those guides never name.

claude.ai: Settings, Customize, Connectors, Add custom connector.

• Name: Trends MCP
• Server URL: https://www.trendsmcp.ai/mcp

That www URL is intentional. Claude handles OAuth. No API key belongs in the connector. Deep link: https://claude.ai/customize/connectors?modal=add-custom-connector&connectorName=Trends%20MCP&connectorUrl=https%3A%2F%2Fwww.trendsmcp.ai%2Fmcp. Start a new chat after consent. An old thread will not see a connector added mid-conversation.

Desktop talks to the api host. User, Settings, Developer, Edit Config, inside mcpServers:

"trends-mcp": {
  "command": "npx",
  "args": ["-y", "mcp-remote", "https://api.trendsmcp.ai/mcp", "--header", "Authorization:${AUTH_HEADER}"],
  "env": { "AUTH_HEADER": "Bearer YOUR_API_KEY" }
}

Mac path: ~/Library/Application Support/Claude/claude_desktop_config.json. Windows path: %APPDATA%\Claude\claude_desktop_config.json. Key from the account page, format tmcp_live_. Fully quit Desktop after saving. A window reload is not enough.

llms-full.txt fetched 2026-11-10 inlines the Bearer string in args and drops the env map. The dedicated Claude page keeps Authorization:${AUTH_HEADER} plus AUTH_HEADER. Both hit the same api host. Do not paste Cursor's url plus transport http object into this file. Cursor Agent uses that shape on get_reviews in Cursor. Pasting it into Desktop is the wrong shape.

Ask with "using TrendsMCP" so Claude routes to the server instead of a web search. The proving prompt is a reviews call, not a board call.

get_reviews(source="Google Play", keyword="notion", limit=5)
get_reviews(source="App Store", keyword="1232780281", limit=5)

get_top_trends with type App Store Top Free still ranks charts. That is a different tool. Do not send a board type string into get_reviews. Do not send source YouTube into get_reviews. MCP source YouTube on keyword notion returned 400 invalid_source, message Unknown source 'YouTube'. Valid: App Store, Google Play, Amazon.

Homepage still puts Claude Code on www

The homepage fetched 2026-11-10 tells Claude Code operators to run this add, then type /mcp to sign in:

claude mcp add --transport http --scope user trends-mcp https://www.trendsmcp.ai/mcp

The same homepage groups Claude Code with Claude and ChatGPT as sign-in clients on https://www.trendsmcp.ai/mcp. Cursor and VS Code stay on the api host with a Bearer key.

llms.txt last updated 2026-11-10 still publishes the other recipe:

claude mcp add --transport http trends-mcp https://api.trendsmcp.ai/mcp --header "Authorization: Bearer YOUR_API_KEY"

The dedicated Claude Code page repeats that api-host line and says not to paste https://www.trendsmcp.ai/mcp into the CLI, because www is OAuth for claude.ai and a www add 401s. This run did not execute claude mcp add. It recorded both published commands. Match the recipe to the binary that is actually open. A Desktop npx block does not install the CLI. A claude.ai connector does not install the CLI.

CLAUDE.md can name get_reviews, sources App Store, Google Play, and Amazon, and a cap of one store then stop. It should not name this morning's review ids. Store rows turn over. A frozen review in CLAUDE.md becomes a stale architecture note.

Nightly review tracking belongs on REST, not a CLI loop that only runs when someone is in the terminal. POST https://api.trendsmcp.ai/api with the same Bearer key and mode get_reviews. Wrapping claude in GitHub Actions just to pull reviews is a roundabout way to hit that URL.

App Store can match Notion and still print zero reviews

A 2026-11-10 MCP get_reviews call that sent source App Store and keyword notion with no limit, days, or country field printed window_days 90, count 0, and requests_charged 1. The listing matched. subject.name was Notion: Notes, Tasks, AI. subject.id was 1232780281. subject.rating was 4.8 from 90,653 ratings. Developer Notion Labs, Incorporated. Genre Productivity. Version 1.7.353. Storefront URL:

https://apps.apple.com/us/app/notion-notes-tasks-ai/id1232780281

error was null. notes was []. summary.average_rating was null. items was []. REST with limit 1 printed the same empty array. MCP days 365 with limit 10 still printed window_days 365 and count 0. Keyword 1232780281 pinned that same id and left alternatives empty. Keyword notion filled three alternatives: Notion Calendar id 1607562761 (7,814 ratings), To-Do List & Tasks for Notion id 1594021892 (761), Widget for Notion id 6751851452 (41).

ChatGPT on App Store matched too. Keyword chatgpt with limit 1 returned subject.id 6448311069, rating 4.8 from 12,078,313 ratings, developer OpenAI OpCo, LLC, version 1.2026.300, and the same empty items array. Alternatives included Google Gemini id 6477489729 (2,413,626 ratings), Grok id 6670324846 (1,592,842), and Claude by Anthropic id 6473753684 (299,659). Claude itself showed up as a near-miss listing on a ChatGPT lookup. That is not a reviews payload for Claude.

The get_reviews page from 2026-11-08 printed 10 App Store rows for notion, rank 1 review 14752198455. That window is gone from this morning's payload. The listing is not. An empty items array is still an answer. get_usage cost_rule during this run: 1 request per data type in a call; get_top_trends: 1 request per 10 rows per board (limit defaults to 100). Only data types that answer are charged. Empty App Store items still charged 1.

Do not ask Claude to retry App Store in a loop. One call already billed. Switch source, or read subject.rating as the all-time store score and stop.

Google Play omit-limit still returns 10

MCP get_reviews with source Google Play and keyword notion, limit omitted, printed window_days 90, count 10, and requests_charged 1 on 2026-11-10. Subject was Notion: AI Workspace, package notion.id, rating 4.6 from 405,740 ratings. Alternatives included Notion Calendar (com.cron.calendar), Notion: Music Notation & Tab (com.presonus.notionmobile), and Microsoft OneNote (com.microsoft.office.onenote). The music app is a different product that shares the word. Read subject.id before averaging stars.

The 10-row window had summary.average_rating 3.6: three 1-star rows, zero 2-star, zero 3-star, two 4-star, five 5-star. replied_share 0.0. negative_share 0.3. Rank 1 was review f2d6bc78-34f9-40f9-86af-dd93c5058166, author Ridwaan Docrat, 5 stars, published 2026-11-09T04:58:22Z, title null, helpful 0, extra app_version 0.6.4288. Rank 6, Sally Geisler, 1 star, carried helpful 2. Rank 7 and rank 8 carried helpful 1. The other seven rows in that window carried 0. Play rows in this pull left title empty. reply was null on every Play row here.

A comma list billed both stores. source App Store, Google Play with limit 1 printed requests_charged 2, App Store count 0, Google Play count 1, and that same Play review id. Top-level count is the sum of items, not the number of sources.

Amazon is the third review source. The first MCP call in this run that sent keyword B0CJZMP7L1 returned envelope statusCode 502 and source-level error source unavailable, with subject null and requests_charged 1. A retry matched STANLEY Quencher H2.0 Flow State Tumbler, 30 oz, Lilac, brand STANLEY, rating 4.7 from 207,409 ratings. summary was null. insights.star_split was 5: 87, 4: 7, 3: 2, 2: 1, 1: 3. The one item had kind excerpt, id B0CJZMP7L1:Quality:0, rating null, published null. The source note was Amazon shows excerpts and topic counts without sign-in, not full dated reviews; days does not filter them. Keyword stanley quencher matched a different ASIN, B0CP9YB3Q4 (Rose Quartz 2.0), with the same 4.7 from 207,409 ratings. Pin the ASIN when the SKU matters. That listing is not the Amazon search index used on get_growth. See Amazon Trends API for the demand series.

get_context takes conversation sources such as YouTube, News, or Reddit. Those names are invalid_source here. Use get_reviews when the job is what customers wrote on a store page.

Claude's MCP list dropped get_top_posts

The MCP tool list in this session, fetched 2026-11-10, advertised get_time_series, get_growth, get_top_trends, get_context, get_reviews, and get_usage. Six tools. Five of them bill. get_usage is free. get_top_posts is absent.

REST still serves the old posts mode. POST https://api.trendsmcp.ai/api with mode get_top_posts, source tiktok, keyword fitness, and limit 1 returned HTTP 200, count 1, and https://www.tiktok.com/@acavallaro7/video/7704795647735614750, posted 2026-11-05, 1,400,000 plays, 151,400 likes. Claude that reads the live MCP list will not offer that call. The existing get-top-posts-in-claude page still tells Claude to call it.

A garbage REST mode getreviews returned envelope statusCode 400, error invalid_mode, and valid_modes get_context, get_growth, get_reviews, get_time_series, get_top_trends. The aliases object named reviews for get_reviews. It named aliases for get_time_series, get_growth, and get_top_trends. It omitted get_context. REST mode reviews with Google Play, keyword chatgpt, and limit 1 returned ChatGPT package com.openai.chatgpt, rating 4.8 from 67,845,290 ratings, and review 5b611fed-c68e-4d39-a441-5565bc7c5d20, 5 stars, published 2026-11-09T06:06:24Z.

Omit REST source and the message is source is required: one or more of App Store, Google Play, Amazon (comma-separated). Omit keyword and the message is keyword is required: a brand, app, product or place name. Failed 400s are free. A successful multi-source call is the one that bills.

OpenAPI 3.1 at https://www.trendsmcp.ai/openapi.json, fetched 2026-11-10, describes five tools in the path text, including get_reviews at 1 request per source. Request examples include a Get Reviews body. components.schemas still has no GetReviewsRequest and no reviews response type. The mode discriminator mapping lists get_context, get_growth, get_time_series, and get_top_trends only. HTML /docs grepped 9 hits for get_reviews and 0 for GetReviewsRequest. llms.txt, docs.md, pricing.md, and data-catalog.md last updated 2026-11-10 all name the tool.

A generated client that only emits the OpenAPI oneOf will never send get_reviews. Claude that reads the live tool list will. Wire the tool the host advertised in MCP, not the discriminator map.

Free plan 100 requests a month. Starter $19 for 1,000. Pro $49 for 5,000. Business $199 for 25,000. One Play call plus one App Store call is two requests even when App Store prints zero rows. Looping "until iOS reviews appear" finishes the free plan before the Project instruction is written. get_usage at the start of this run printed Business 3,287 of 25,000 used, then 3,299, reset 2026-12-01. buy_credits_url on that payload still uses the apex account billing tab:

https://trendsmcp.ai/account?tab=billing

Site links on this page stay on www.