Reviews MCP Server — Trustpilot, App Store & Google Play for Claude, Cursor & ChatGPT

Give your AI assistant live voice-of-customer data. This Actor runs as a hosted MCP server (Model Context Protocol): connect Claude, Cursor, ChatGPT, VS Code or any MCP client and it can find companies on Trustpilot, read real review texts, and pull a multi-platform rating snapshot across Trustpilot, the App Store and Google Play — with no API keys, no OAuth app, and no scraping code to maintain.

The differentiator: actual review bodies, not just star averages. Your agent reads what customers wrote — the complaint, the praise, the verified flag, the company's reply — so it can cluster pain points, mine feature requests, or brief you on a brand before a call.

Apify hosts it, authenticates it with your Apify token, and bills it per tool call — you don't run or host anything.

---

🧰 Tools

Three tools, all returning clean, typed JSON — ready for an LLM to reason over.

Tool	What it does	Key arguments
find_company	Search Trustpilot for a company by name or domain. Returns matching businesses with TrustScore, star rating, review count, domain and country, plus the id you feed into get_reviews.	query (required), limit (1–20)
get_reviews	Get actual Trustpilot review texts for a company — full body, rating, title, author, dates, verified flag, likes and the company's reply. Pass a name, domain, or the id from find_company. Filter by star rating or language. This is the voice-of-customer differentiator.	company (required), limit (1–100), stars (1–5), sort (relevancy/recency), language
get_ratings	One-call multi-platform reputation snapshot: Trustpilot TrustScore + App Store & Google Play app score, review count and profile link — side by side.	brand (required), domain

Why the split? get_reviews returns real review text (currently sourced from Trustpilot, the platform with fast full-text access). get_ratings returns cross-platform scores + counts + links so your agent can spot when a brand's story differs by channel.

---

🔌 Connect it

This Actor runs in Standby mode and exposes an MCP endpoint at /mcp over the Streamable HTTP transport. Point your MCP client at the Actor's Standby URL and authenticate with your Apify API token as a bearer token.

Add this to your MCP client config (Claude Desktop, Cursor, VS Code, Windsurf, …):

{
  "mcpServers": {
    "reviews": {
      "url": "https://<YOUR-STANDBY-URL>/mcp",
      "headers": { "Authorization": "Bearer <YOUR_APIFY_TOKEN>" }
    }
  }
}

• Your Standby URL is shown on the Actor's page once you enable Standby (looks like https://<user>--reviews-mcp-server.apify.actor).
• Your Apify API token is under Settings → Integrations in the Apify Console.

That's it — your assistant discovers the three tools automatically and picks the right one.

---

💬 Example prompts

Once connected, just ask your assistant naturally:

• “What's Notion's TrustScore on Trustpilot, and how many reviews?”
• “Pull the latest 1-star Duolingo reviews and summarise the top complaints.”
• “Give me a reputation snapshot for Duolingo across Trustpilot, the App Store and Google Play.”
• “Read Trustpilot reviews for nike.com and cluster them into recurring themes.”
• “Is this brand loved on the app stores but hated on Trustpilot? Show me the gap.”

---

📦 Example tool calls & output

find_company

{ "query": "Notion" }

{
  "query": "Notion",
  "count": 3,
  "companies": [
    {
      "id": "5a0b1c2d3e4f5a6b7c8d9e0f",
      "name": "notion.so",
      "displayName": "Notion",
      "domain": "notion.so",
      "website": "https://www.notion.so",
      "trustScore": 2.3,
      "stars": 2.5,
      "reviewCount": 417,
      "country": "US",
      "trustpilotUrl": "https://www.trustpilot.com/review/notion.so"
    }
  ]
}

get_reviews

{ "company": "duolingo.com", "stars": 1, "limit": 3, "sort": "recency" }

{
  "company": "duolingo.com",
  "businessUnitId": "4d9d3b1e00006400050e3f2a",
  "count": 3,
  "reviews": [
    {
      "rating": 1,
      "title": "DO NOT PAY FOR DUOLINGO!",
      "text": "I paid for a full year of Super and the app locked me out of my streak after an update. Support just sends canned replies and won't refund. Save your money.",
      "author": "Marta K.",
      "country": "GB",
      "publishedDate": "2026-06-28T08:14:02.000Z",
      "experiencedDate": "2026-06-25T00:00:00.000Z",
      "isVerified": true,
      "likes": 34,
      "companyReply": null,
      "url": "https://www.trustpilot.com/reviews/6a1b2c3d4e5f"
    },
    {
      "rating": 1,
      "title": "Impossible to cancel",
      "text": "They auto-renewed my subscription and there is no clear way to cancel from the app. Had to dispute the charge with my bank.",
      "author": "David R.",
      "country": "US",
      "publishedDate": "2026-06-27T19:41:55.000Z",
      "experiencedDate": "2026-06-27T00:00:00.000Z",
      "isVerified": true,
      "likes": 12,
      "companyReply": "We're sorry to hear this — you can manage your subscription in your store account settings…"
    }
  ]
}

Every review carries rating, title, text, author, publishedDate, experiencedDate, isVerified, likes, and companyReply — enough for an LLM to cluster complaints, score sentiment, or quote customers verbatim.

get_ratings

{ "brand": "Duolingo", "domain": "duolingo.com" }

{
  "brand": "Duolingo",
  "platforms": {
    "trustpilot": {
      "found": true,
      "trustScore": 1.6,
      "stars": 1.5,
      "reviewCount": 2841,
      "url": "https://www.trustpilot.com/review/duolingo.com"
    },
    "appStore": {
      "found": true,
      "score": 4.73,
      "reviewCount": 5300000,
      "url": "https://apps.apple.com/us/app/duolingo/id570060128",
      "app": "Duolingo - Language Lessons"
    },
    "googlePlay": {
      "found": true,
      "score": 4.7,
      "reviewCount": 47000000,
      "url": "https://play.google.com/store/apps/details?id=com.duolingo",
      "app": "Duolingo: language lessons"
    }
  }
}

The voice-of-customer contrast in one call. Duolingo scores 1.6 on Trustpilot but 4.73 on the App Store (5.3M reviews) and 4.7 on Google Play (47M reviews). Trustpilot skews toward billing and support complaints; the app stores capture the everyday product experience. Your agent sees both sides — and get_reviews tells it why.

---

✅ Why use it

• Real review text, not just averages. get_reviews returns full Trustpilot bodies — the actual words customers wrote, with verified flags and company replies.
• Multi-platform in one call. get_ratings lines up Trustpilot, App Store and Google Play so your agent spots channel gaps instantly.
• No keys, no OAuth. Reads public data — nothing to register, no rate-limit paperwork.
• Built for agents. Clean, typed tools with LLM-friendly descriptions, so your assistant picks the right one and gets structured JSON back.
• Pay per call. Billed per tool call — no subscription, no idle server cost, nothing to host.
• Works everywhere. Standard Streamable HTTP MCP transport — Claude, Cursor, ChatGPT, VS Code, Windsurf, or your own agent.

---

🎯 Great for

• Voice-of-customer & complaint analysis — cluster real review texts into recurring pain points.
• Competitive & reputation research — snapshot any brand's standing across three platforms before a pitch or a call.
• Product & feature discovery — mine what customers ask for and where a competitor is weak.
• Brand monitoring — track TrustScore and app ratings for your own or a rival's brand.
• Due diligence & sales prep — brief your agent on how a company is actually perceived.

---

⚙️ Notes & limitations

• Review text is currently sourced from Trustpilot — the platform with fast, reliable full-text access. App Store and Google Play contribute ratings + review counts via get_ratings (not per-review bodies).
• Read-only, public data only — no login, no posting, no private data.
• get_reviews returns up to 100 reviews per call; filter by stars and language to focus results.
• App-store matching is best-effort by brand/domain; pass domain to get_ratings to improve accuracy.

---

⚠️ Disclaimer

This Actor accesses only publicly available review and rating data and performs no login or private-account access. You are responsible for using the output in compliance with each platform's terms and applicable laws. Not affiliated with, endorsed by, or sponsored by Trustpilot, Apple, or Google. “Trustpilot”, “App Store” and “Google Play” are trademarks of their respective owners.

---

🔎 SEO keywords

Reviews MCP, Trustpilot MCP, Trustpilot MCP server, voice of customer MCP, review scraper MCP, App Store reviews MCP, Google Play reviews MCP, Trustpilot API alternative, TrustScore API, customer reviews for AI agents, review data for LLM, MCP server, Model Context Protocol, Trustpilot for Claude, reviews for Cursor, reviews for ChatGPT, multi-platform ratings, brand reputation API, no-auth reviews, hosted MCP server, Streamable HTTP MCP.