# Amazon Trends MCP [Health: Active]

**Category:** 💻 Developer Tools  
**Repository:** https://github.com/trendsmcp-ai/amazon-trends-mcp  
**GitHub Stars:** 0  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/amazon-trends-mcp-3

## Description
Amazon product search demand over time, with growth for any keyword. Free key at trendsmcp.ai

## Tools
Capabilities this server exposes over MCP:

- **get_time_series** — Weekly (or daily) history for **one** `source` + `keyword`. Same name on MCP and REST (`mode: "get_time_series"`). REST also accepts `get_trends` as an alias.
- **get_growth** — Point-to-point percent change. Several windows in one call still count as **one** request for that source + keyword.
- **get_top_trends** — Live ranked list. **No keyword.** On MCP, `type` is required and must match the feed name **exactly** (including capitals). On REST, omit `type` only if you intend to pull every feed (billed per feed).

## Claude Desktop Quick Installation
Remote MCP endpoint (confidence: high). Install path detected from listing signals. Add as a URL/SSE server in your client:

```json
"mcpServers": {
  "amazon-trends-mcp": {
    "url": "https://api.trendsmcp.ai/mcp"
  }
}
```

## Documentation

## What Amazon Trends MCP MCP server does

Amazon Trends MCP MCP server exposes Amazon demand data to AI agents through the Model Context Protocol. It provides three operations: historical values for a keyword, percentage growth between time periods, and ranked live-trend feeds. The hosted MCP endpoint is `https://api.trendsmcp.ai/mcp`, and the repository also includes a stdio adapter for local MCP hosts and Glama.

The service can be used for Amazon-specific analysis and, through the broader TrendsMCP API, supports other sources such as Google, TikTok, YouTube, and Reddit. The tools in this listing focus on selecting a source and querying it with the appropriate keyword or feed parameters.

## How it works

The MCP connection uses a TrendsMCP API key for hosted requests. The same bearer key is used with the REST endpoint at `https://api.trendsmcp.ai/api`. A hosted connection is the default approach; the local adapter starts with Python and reads `TRENDSMCP_API_KEY` when paid calls are made.

`get_time_series` accepts one source and one keyword per request. It returns weekly data by default, with daily data available through the REST API, and may include both a normalized 0–100 index and an absolute volume series when the source provides one. The index is relative to the returned time window, where 100 represents its peak.

`get_growth` calculates point-to-point percentage change. Multiple comparison windows can be included in one call, and they still count as one request for the selected source and keyword. The source can be one source or a comma-separated list, including combinations such as Google Search, TikTok, and Amazon.

## Setup and configuration

Create an API key through the TrendsMCP account page before making hosted requests. Cursor, VS Code, Windsurf, Cline, Claude Code, and other HTTP-capable MCP clients can be configured with the hosted URL and an authorization header. Cline requires the transport type `streamableHttp`.

Claude.ai can add the hosted URL as a custom connector and uses OAuth on the `www.trendsmcp.ai` path instead of a bearer key. Claude Desktop does not provide native HTTP support in the documented setup; it can connect through `mcp-remote` and an `AUTH_HEADER` environment value.

For local stdio use, install the repository package and run `python -m trends_mcp_server`. The process can list tools without a key, but paid calls require `TRENDSMCP_API_KEY` and use the same quota.

## Tools and capabilities

- `get_time_series`: Retrieves weekly or daily history for one keyword on one source. The REST API uses `mode: "get_time_series"` and also accepts `get_trends` as an alias.
- `get_growth`: Returns percentage change for presets ranging from 7 days to 5 years, plus MTD, QTD, YTD, and custom date objects. Its default window is 12 months.
- `get_top_trends`: Returns a live ranked list without a keyword. It supports pagination, result limits up to 200, ranking or rank-change sorting, and optional category and time-window filters.

## Limitations and notes

The Amazon Trends MCP MCP server requires a source and keyword for the first two tools, and each time-series request handles only one source. For `get_top_trends`, MCP requires `type`, and its value must exactly match the feed name, including capitalization. REST can omit `type` to request every feed, but billing applies per feed.

The documented free tier provides 100 requests per month. Current parameter lists and feed names are maintained in the linked service documentation, so clients should not assume that a feed name or source format is interchangeable across platforms.

_Full upstream README: https://allmcps.com/mcp/amazon-trends-mcp-3/readme_