Amazon Ads MCP Server

What is Amazon Ads MCP Server about?

The server bridges Amazon Advertising data with any MCP‑compatible client, exposing campaigns, ad groups, keywords, product ads, targeting rules, and downloadable reports via a simple JSON‑over‑stdio interface. It also surfaces MarketplaceAdPros recommendations, experiments, and subscription‑based insights, allowing users to query the data in plain English.

How to use Amazon Ads MCP Server?

1. Add to MCP client configuration – Insert the server definition into your Claude Desktop, LibreChat, or any MCP client config file. Use the provided npx command for the easiest setup.
2. Set the BEARER_TOKEN environment variable – Obtain a bearer token from MarketplaceAdPros and supply it in the env section.
3. Run the server – The client will launch the server via npx and communicate over stdio.
4. Query – Send plain‑English prompts or structured requests to retrieve campaigns, generate reports, or get recommendations.

{
  "mcpServers": {
    "amazon-ads": {
      "command": "npx",
      "args": ["-y", "@marketplaceadpros/amazon-ads-mcp-server"],
      "env": { "BEARER_TOKEN": "<YOUR_BEARER_TOKEN>" }
    }
  }
}

Alternatively, point the client to the hosted streamable‑HTTP endpoint https://app.marketplaceadpros.com/mcp and pass the token in the Authorization header.

Key features of Amazon Ads MCP Server

• Access to Sponsored Products, Brands, and Display resources (campaigns, ad groups, keywords, product ads, targeting).
• On‑demand performance reports with plain‑English querying.
• MarketplaceAdPros recommendations, experiments, and subscription‑only insights.
• Multiple deployment options: local npx server, built binary, or hosted streamable‑HTTP endpoint.
• Debugging support via the MCP Inspector tool.

Use cases of Amazon Ads MCP Server

• Automated advertising dashboards – Pull real‑time campaign metrics into custom BI tools.
• AI‑assisted campaign optimization – Let Claude generate suggestions or rewrite targeting rules based on performance data.
• Report generation scripts – Schedule daily/weekly report retrieval without manual UI interaction.
• Integration with third‑party workflows – Connect Amazon Ads data to CRMs, inventory systems, or pricing engines.
• Rapid prototyping – Experiment with plain‑English queries to explore advertising data structures.

FAQ from the Amazon Ads MCP Server

Q: Do I need an Amazon Advertising account? A: Yes, a valid MarketplaceAdPros subscription linked to your Amazon Advertising account is required.

Q: Which environment variable holds the token? A: BEARER_TOKEN – obtain it from MarketplaceAdPros dashboard.

Q: Can I run the server on Windows/macOS/Linux? A: The npx approach works on all platforms with Node.js installed.

Q: How do I debug communication issues? A: Use the provided npm run inspector script, which launches the MCP Inspector UI.

Q: Is there a hosted version? A: Yes, use the streamable‑HTTP endpoint at https://app.marketplaceadpros.com/mcp with the bearer token in the Authorization header.

amazon-ads-mcp-server

Connect to your Amazon Advertising Data by integrating your account with MarketplaceAdPros.

Provides access to:

• Advertising Resources in Sponsored Products, Sponsored Brands and Sponsored Display, like Campaigns, Ad Groups, Keywords, Product Ads, Targeting
• Reports and ability to query them with plain english.
• Marketplace Ad Pros Recommendations, Experiments and more with purchased subscription plan

Also available as a Streamable HTTP MCP Server by connecting to https://app.marketplaceadpros.com/mcp

Installation

To add the amazon-ads-mcp-server to your MCP client of choice, add the following to the server config:

On MacOS: ~/Library/Application Support/Claude/claude_desktop_config.json

On Windows: %APPDATA%/Claude/claude_desktop_config.json

Env Vars

• BEARER_TOKEN: The Bearer token you got from MarketplaceAdPros.com

Configuration

You can use it via npx in your Claude Desktop configuration like this:

{
  "mcpServers": {
    "marketplaceadpros": {
      "command": "npx",
      "args": [
        "@marketplaceadpros/amazon-ads-mcp-server"
      ],
      "env": {
        "BEARER_TOKEN": "abcdefghijklmnop"
      }
    }
  }
}

Or, if you clone the repo, you can build and use in your Claude Desktop configuration like this:

{
  "mcpServers": {
    "marketplaceadpros": {
      "command": "node",
      "args": [
        "/path/to/amazon-ads-mcp-server/build/index.js"
      ],
      "env": {
        "BEARER_TOKEN": "abcdefghijklmnop"
      }
    }
  }
}

Or, if your client supports the Streamable HTTP MCP Servers, you can just point to the MCP endpoint at https://app.marketplaceadpros.com/mcp.

{
  "mcpServers": {
    "marketplaceadpros": {
      "type": "streamable-http",
      "url": "https://app.marketplaceadpros.com/mcp"
    }
  }
}

Or, configure in LibreChat like:

  MAP:
    type: streamable-http
    url: https://app.marketplaceadpros.com/mcp
    headers:
      Authorization: "Bearer abcdefghijklmnop"

Development

Install dependencies:

npm install

Build the server:

npm run build

For development with auto-rebuild:

npm run watch

Debugging

Since MCP servers communicate over stdio, debugging can be challenging. We recommend using the MCP Inspector, which is available as a package script:

npm run inspector

The Inspector will provide a URL to access debugging tools in your browser.

Acknowledgements

• Obviously the modelcontextprotocol and Anthropic teams for the MCP Specification. https://modelcontextprotocol.io/introduction
• MarketplaceAdPros for enabling and sponsoring this project.