# Amazon Trends API [Health: Active]

**Category:** 💻 Developer Tools  
**Repository:** https://github.com/trendsapi-ai/amazon-trends-api  
**GitHub Stars:** 0  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/amazon-trends-api-2

## Description
Amazon product search volume and best-seller trends. Free key at trendsapi.ai

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `uvx` (confidence: high):

```json
"mcpServers": {
  "amazon-trends-api": {
    "command": "uvx",
    "args": ["trendsapi-amazon"],
    "env": {
      "TRENDSAPI_KEY": ""
    }
  }
}
```

**Requires environment variables:** `TRENDSAPI_KEY` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What Amazon Trends API MCP server does

Amazon Trends API MCP server provides access to Amazon-focused demand data through the Trends API. It covers two related use cases: historical keyword interest and current bestseller feeds. Historical queries return dated values for a product phrase such as “standing desk”; live queries return ranked label rows from supported feeds.

The search-history response can include a normalized value from 0 to 100, the queried keyword, dates, and, when available, absolute volume. The index represents search interest for that series and should not be interpreted as the number of products sold. Bestseller feeds answer a different question by showing what is ranking at the time of the snapshot.

## How it works

Requests are sent to `https://api.trendsapi.ai/api` with `Authorization: Bearer $TRENDSAPI_KEY`. History requests use the `get_time_series` or `get_growth` modes with `source: amazon`. Live requests use the `get_top_trends` mode and can select a supported feed type.

The documented Python package returns dataclass-style objects, while the JavaScript package returns objects with the same fields. Both SDKs also provide asynchronous or promise-based calls. The HTTP API can be called directly with JSON, although raw responses may require decoding the JSON contained in the `body` field.

Growth requests support preset periods including 7D, 30D, 3M, 12M, YTD, and longer ranges, along with custom recent and baseline dates. Multiple periods can be requested in one API request. Live results include snapshot time, feed type, pagination details, and `[rank, label]` rows.

## Setup and configuration

Create a key at trendsapi.ai, then install either the Python package with `pip install trendsapi-amazon` or the JavaScript package with `npm install trendsapi-amazon`. Python requires version 3.9 or newer; the JavaScript package supports Node 18+, Deno, Bun, and Workers.

Set `TRENDSAPI_KEY` in the process environment, or pass the key directly to the SDK constructor. The same key is used by the HTTP API and the language clients. Python also includes `AsyncTrendsAPI` for asynchronous calls.

The provided material documents SDK and HTTP usage rather than a separate MCP transport, client registration file, or MCP-specific installation command. Confirm the intended adapter or host configuration before treating this package as a native MCP deployment.

## Tools and capabilities

The documented operations are:

- `get_time_series` / `getTimeSeries`: retrieve dated search-interest points for a keyword.
- `get_growth` / `getGrowth`: compare search interest across preset or custom periods.
- `get_live` / `getLive`: retrieve a paginated live trend feed.
- `get_top_trends` / `getTopTrends`: request a live feed by its exact type.

Keyword helpers default to the Amazon source, while the client can accept another source where supported. The live API accepts `limit`, `offset`, and, for applicable feeds, `category`. Pandas users can convert Python time-series dataclasses into a DataFrame for date conversion and resampling.

## Limitations and notes

Amazon Trends API MCP server does not provide product-sales units through the search-interest value. A missing keyword series produces a 404 and should not be retried. Invalid source or feed-type combinations produce a 400; an invalid key produces a 401, and quota exhaustion produces a 429. Server errors are retried by the SDK before being raised.

The source value must use the documented lowercase spelling `amazon`. Feed types must use their exact names, such as `Amazon Best Sellers Top Rated` or `Amazon Best Sellers by Category`; mixing a source value with a feed type incorrectly can result in a 400. Each successful asynchronous call counts as one billed request.

_Full upstream README: https://allmcps.com/mcp/amazon-trends-api-2/readme_