# Pangolinfo: Turn a Competitor's Amazon Reviews into a Product Brief in Minutes

> Connect Pangolinfo's Amazon Insight MCP to Claude Code and turn a competitor's Amazon reviews into a product brief from live data. Free starter lesson.

Source: https://learn.ynteractive.com/content/pangolinfo-amazon-review-analysis · Updated 2026-11-13 · Free from Stepthrough (https://learn.ynteractive.com)

## Practice it step by step

[Start the free practice](https://learn.ynteractive.com/modules/pangolinfo-amazon-review-analysis)

You do every step yourself in a practice copy of Pangolinfo. It checks your work as you go, and “Do it for me” finishes any step you get stuck on. Free, about 7 minutes, and nothing touches your real accounts. [Open full screen](https://learn.ynteractive.com/modules/pangolinfo-amazon-review-analysis)

Free · Live now · [AI Automation for Marketing](https://learn.ynteractive.com/courses/ai-for-marketing) · Pangolinfo · Claude Code · ~7 min · 8 steps · Beginner

Updated November 13, 2026. Built from Pangolinfo’s official docs and help center. Screens may differ slightly in your account. [How we check this lesson](https://learn.ynteractive.com/content/about#how-we-check)

Product teams at consumer brands lose a day every time they read competitor reviews by hand before a product decision. Pangolinfo is a data service for Amazon, e-commerce, social media and search data, and its Amazon Insight MCP lets an [AI agent](https://learn.ynteractive.com/content/n8n-ai-agent-tutorial-google-sheets) pull that data in a conversation. In this free starter lesson, built from Pangolinfo's documentation, you get a Permanent Key from the Pangolinfo Console, add the Amazon Insight MCP to [Claude](https://learn.ynteractive.com/content/tool/claude) Code for Brightfern Home, a made-up kitchen-storage brand, restart and check the tool list, then ask for a review analysis of a competitor's ASIN and turn it into a product brief. 8 steps, about seven minutes, no account needed to practice.

## At a glance

| Question | Answer |
| --- | --- |
| What you build | Reading hundreds of competitor reviews by hand before every product decision takes a day. Connect Pangolinfo's Amazon Insight MCP to Claude Code with one config file, then ask for a review analysis of any ASIN and get the praise, complaints and repeated themes from live review data, all from Pangolinfo's own docs. |
| Tools | Pangolinfo |
| Time | About 7 minutes, 8 steps in 3 parts |
| Level | Beginner, no coding needed |
| Cost | Free |
| Ways to learn it | [Practice it step by step](https://learn.ynteractive.com/modules/pangolinfo-amazon-review-analysis) |

### The 3 parts, in short

1. Connect the MCP: Get a Permanent Key, Add the server to .mcp.json, Restart and check the tool list.
2. Ask for the review analysis: Open a new session, Ask the business question, Send it and read the themes.
3. Turn it into a brief: Ask a follow-up on the same data, Get the brief, then try it in your own account.

## The business case: a day of reading reviews per product brief

Picture Priya, product manager at Brightfern Home, a made-up brand that sells kitchen storage on Amazon. Before every new product spec she reads the reviews of three or four competing listings, copies the complaints into a spreadsheet and counts them by hand. With example numbers you should replace with your own, that is about a day per brief, several briefs a quarter, and a spreadsheet that is out of date a month later.

Pangolinfo's docs describe a shorter path. The Amazon Insight MCP gives an AI client such as [Claude](https://learn.ynteractive.com/content/glossary#claude) Code tools that [fetch](https://learn.ynteractive.com/content/glossary#fetch-api) Amazon product details and reviews as structured data. Priya asks one business question with the competitor's ASIN, the agent calls get_amazon_product and get_amazon_reviews, and she gets the praise, complaints and repeated themes with counts. A follow-up on the same data turns the complaints into requirements for her next product.

## Who this is for, and what you need

Product managers, category managers, market researchers and e-commerce marketers who make decisions from Amazon reviews and listings. The practice runs in your browser with no account. To do it for real you need a Pangolinfo account with a plan or quota (the docs say data calls check your available quota), a Permanent Key from Profile › API Key in the Pangolinfo Console, and an MCP client that supports remote Streamable HTTP, such as Claude Code or Cursor.

## What the docs say about setup and the data

The Client Setup Guide puts the Claude Code config in .mcp.json at the project root or in the user-level .claude.json, and says the "type": "http" field is required, because without it Claude Code tries to parse the server as stdio and fails. To pick up the change, type /exit and run claude again. If the tool list is empty, the guide says to check the URL spelling, confirm you used a Permanent Key and fully restart the client; a 401 means the key is invalid and a timeout means the client cannot reach mcp.pangolinfo.com. Header Auth is the recommended way to send the key; URL Query Auth is only for clients that cannot set custom headers.

The tool reference lists get_amazon_reviews with an asin, a site such as amz_us, amz_uk or amz_de, and a pageCount. Each review returns fields such as title, content, star, date, country and purchased (Verified Purchase), alongside totalReviews. The workflow page advises fetching factual data first and drawing conclusions second, relying on real-time tool results for prices, rankings and reviews rather than model memory, and naming the dimensions you want compared.

## What this saves you

- The day of reading and copying reviews before each product brief.
- A brief you can rerun with fresh data whenever the market moves, with each point tied back to real reviews.
- What it does not do: Pangolinfo's docs point batch jobs and system integration to its REST API rather than MCP, and say patent and litigation results are risk pre-checks, not legal advice.

## How this lesson was built

This is a starter lesson based on Pangolinfo's documentation (the five docs pages linked above), read on 13 November 2026. The practice screen is a plain frame that uses the labels from those docs; it is not a copy of Pangolinfo's or Claude Code's own screens, and the lesson was not run in a Pangolinfo account. The ASIN and the review themes are made up. A full lesson on real screens is planned.

## Links

- [Pangolinfo docs: MCP for Agent: Overview](https://docs.pangolinfo.com/en-help-center/mcp/overview): the Amazon Insight MCP and the VOC Insight MCP, the supported clients, the three setup steps and the full list of Amazon Insight tool names.
- [Pangolinfo docs: Pangolinfo MCP - Client Setup Guide](https://docs.pangolinfo.com/en-help-center/mcp/agents): the Permanent Key from Profile › API Key, the .mcp.json config for Claude Code with "type": "http", the Cursor config, restarting, and what an empty tool list, a 401 or a timeout means.
- [Pangolinfo docs: Amazon Insight MCP: Connection & Authentication](https://docs.pangolinfo.com/en-help-center/mcp/amazon-insight/connect): the MCP URL, Header Auth versus URL Query Auth, what you need before you start, key handling and the first test prompt.
- [Pangolinfo docs: Pangolinfo MCP - Full Reference of 21 Tools](https://docs.pangolinfo.com/en-help-center/mcp/tools): the parameters and return fields of get_amazon_reviews, get_amazon_product, search_amazon and the other tools.
- [Pangolinfo docs: Amazon Insight Workflow](https://docs.pangolinfo.com/en-help-center/mcp/amazon-insight/workflow): the five-step workflow, example chains such as an ASIN audit or a keyword market scan, and prompt templates.

## Try it yourself, free

Reading the steps is a start. Doing them is how it sticks. The interactive module walks you through every click in a practice copy of Pangolinfo, checks each step, and never touches your real accounts.

[Start “Pangolinfo: Turn a Competitor's Amazon Reviews into a Product Brief in Minutes”](https://learn.ynteractive.com/modules/pangolinfo-amazon-review-analysis)

## Step by step: Pangolinfo: Turn a Competitor's Amazon Reviews into a Product Brief in Minutes

All 8 steps of the interactive module, in order. In the module you watch each one, then do it yourself in a practice copy that checks your work.

### 1. Connect the MCP

1. **Get a Permanent Key.** You are Priya, product manager at **Brightfern Home**, a made-up kitchen-storage brand on Amazon. Reading competitor reviews by hand takes a day per product brief. In the Pangolinfo Console, open **Profile › API Key** and click **Permanent Key**; the Client Setup Guide says MCP needs a Permanent Key.
   You are Priya, product manager at Brightfern Home. Reading competitor reviews by hand takes a day per product brief. In the Pangolinfo Console, open the API Key page and click Permanent Key.
2. **Add the server to .mcp.json.** In your project root, the **.mcp.json** file has a server named **pangolinfo** with **"type": "http"**, which the docs say is required in Claude Code. Type the **url**: **https://mcp.pangolinfo.com/mcp**. The **Authorization** header is **Bearer** followed by your key.
   In the project config file, the server is named pangolinfo and the type is http, which Claude Code requires. The authorization header carries your key.
3. **Restart and check the tool list.** Restart Claude Code: type **/exit**, then run **claude** again. The Amazon Insight tools now show in the tool list, including **get_amazon_product**, **get_amazon_reviews** and **pangolinfo_capabilities**, which the reference says to call first.
   Restart Claude Code. The Amazon Insight tools now show in the tool list.

### 2. Ask for the review analysis

4. **Open a new session.** Open a new Claude Code session. Pangolinfo's workflow page says to start with a specific business question and give an identifier: an ASIN, keyword, seller ID, category, URL or product name.
   The workflow page says to start with a specific business question and give the agent an identifier, such as an ASIN.
5. **Ask the business question.** Type the question for the competitor's lid-top container, ASIN **B0EXAMPLE0** (made up): **Run a review analysis on ASIN B0EXAMPLE0 on Amazon US. Pull the product details and 3 pages of reviews, then tell me what buyers praise, complain about and repeatedly mention, with a count for each theme.**
   Ask for a review analysis on the competitor ASIN on Amazon US, with the product details and three pages of reviews, and a count for each theme.
6. **Send it and read the themes.** Click **Send**. The agent calls **get_amazon_product** and **get_amazon_reviews** with **site: amz_us** and **pageCount: 3**. Each review comes back with its **title**, **content**, **star**, **date** and whether it was a **Verified Purchase**, and the agent groups them into themes.
   The agent calls the product and review tools for Amazon US, three pages deep. Each review comes back with its title, text, stars and date, and the agent groups them into themes.

### 3. Turn it into a brief

7. **Ask a follow-up on the same data.** The docs say to ask follow-up questions that draw on the same source data. Type: **Turn the complaints into three requirements for our next container, most common first, and cite the review IDs behind each one.**
   The docs say to ask follow-ups on the same data. Ask the agent to turn the complaints into three requirements for your next product, and to cite the reviews behind each one.
8. **Get the brief, then try it in your own account.** Click **Send**. The brief lists three requirements with the reviews behind them. This lesson was built from Pangolinfo's docs, not run in an account: in yours, open **Profile › API Key** in the Pangolinfo Console, follow the **Client Setup Guide**, and ask about an ASIN you compete with.
   The brief lists three requirements with the reviews behind them. In your own account, copy a Permanent Key from the console, follow the client setup guide, and ask about a product you compete with.

## Frequently asked questions

**How do I connect Pangolinfo's Amazon Insight MCP to Claude Code?**

Pangolinfo's Client Setup Guide says to add a server to .mcp.json in your project root (or to the user-level .claude.json) with "type": "http", the URL https://mcp.pangolinfo.com/mcp and a header Authorization: Bearer followed by your API key. Then type /exit and run claude again.

**Where do I get the Pangolinfo API key for MCP?**

The Client Setup Guide says to get a Permanent Key from the Pangolinfo Console under Profile › API Key. The docs say to treat it like a long-term password and to revoke and regenerate it if it may have leaked.

**What review data does get_amazon_reviews return?**

According to the tool reference, it takes an asin, a site such as amz_us and a pageCount, and returns totalReviews plus a list of reviews with fields such as title, content, star, date, country and purchased (Verified Purchase).

**Was this lesson tested in a real Pangolinfo account?**

No. It is a free starter lesson built from Pangolinfo's public documentation. The practice frame is generic and the labels come from the docs; check them against your own account.

[Start this module free](https://learn.ynteractive.com/modules/pangolinfo-amazon-review-analysis)

- Next in AI Automation for Marketing: [How to Use n8n in Marketing: An AI Agent That Plans Your Content Calendar](https://learn.ynteractive.com/content/n8n-ai-agent-content-calendar)

## More in AI Automation for Marketing

- [Build a Lead Magnet Signup Form with an AI Welcome Email](https://learn.ynteractive.com/content/lead-magnet-signup-form-ai-welcome-email): Build a "Get the free guide" signup form in Lovable for Harbor & Pine Coffee Roasters and connect it to n8n — the AI agent tags each lead by interest, saves it to…
- [Get Your Business Recommended by ChatGPT](https://learn.ynteractive.com/content/get-recommended-by-chatgpt): Ask ChatGPT who it recommends in your town, see which pages it trusts, then use it to write an FAQ page and Business Profile copy that answer real customer questions.
- [iMini AI Agent: Turn a Rough Idea into a Launch Kit](https://learn.ynteractive.com/content/imini-ai-agent-marketing-launch-kit): Brief the iMini Creative Agent once and get a launch plan, a LinkedIn visual, a post and a 5-slide deck, then fix the visual and humanize the post with follow-ups.
- [Post to Every Channel Daily from One Sheet using n8n + Blotato](https://learn.ynteractive.com/content/post-everywhere-from-google-sheets-n8n-blotato): Copying the same post into every social app can take half an hour a day, and on busy days nothing goes out. Plan a week of posts in one Google Sheet; every morning n8n…
- [VisualGPT: Product Photos, Listing Images and Ads from One Phone Photo](https://learn.ynteractive.com/content/visualgpt-product-photos-listing-images-ads): Turn one phone photo into a lifestyle product shot, a five-image Amazon listing set and three launch ads in VisualGPT's Commercial Studio, checking the plan and every…
- [Flova: Plan a Promo Video from a Script Before You Spend a Credit](https://learn.ynteractive.com/content/flova-promo-video-from-a-script): Give Flova's AI video agent a ten-second script for a small business, check its storyboard and credit estimate first, then make and fix one reference image.

[All 26 modules in AI Automation for Marketing](https://learn.ynteractive.com/courses/ai-for-marketing)

## Terms in this lesson

Plain-English definitions, each with an example from a free lesson.

[AI agent](https://learn.ynteractive.com/content/glossary/ai-agent) · [Claude](https://learn.ynteractive.com/content/glossary#claude) · [fetch (JavaScript)](https://learn.ynteractive.com/content/glossary#fetch-api) · Unstructured data · [API](https://learn.ynteractive.com/content/glossary#api) · [API key](https://learn.ynteractive.com/content/glossary#api-key)

[The full AI glossary](https://learn.ynteractive.com/content/glossary)

## Browse related tutorials

[AI for Marketing (26)](https://learn.ynteractive.com/courses/ai-for-marketing)

**Get the badge / embed this lesson**
