AmazonCAUTION
🛍📦 Unofficial Amazon Model Context Protocol Server (MCP) - Search products and purchase directly from Claude AI! ✨
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
This server allows you to interact with Amazon's services using the MCP (Model Context Protocol) framework. This lets you use your Amazon account through ChatGPT or Claude AI interfaces.
Features
- Product search: Search for products on Amazon
- Product details: Retrieve detailed information about a specific product on Amazon
- Cart management: Add items or clear your Amazon cart
- Ordering: Place orders (fake for demonstration purposes)
- Orders history: Retrieve your recent Amazon orders details
Demo
Simple demo, showcasing a quick product search and purchase.
Full Demo
Another more complex demo with products search, leveraging Claude AI recommendations to compare and make a decision, then purchase.
It showcases how natural and powerful the Amazon MCP integration could be inside a conversation
Video: https://www.youtube.com/watch?v=xas2CLkJDYg
Install
Install dependencies
npm install -D
Build the project
npm run build
Claude Desktop Integration
Create or update ~/Library/Application Support/Claude/claude_desktop_config.json with the path to the MCP server.
{
"mcpServers": {
"amazon": {
"command": "node",
"args": ["/Users/admin/dev/mcp-server-amazon/build/index.js"]
}
}
}Restart the Claude Desktop app to apply the changes. You should now see the Amazon MCP server listed in the Claude Desktop app.
Troubleshooting
The MCP server logs its output to a file. If you encounter any issues, you can check the log file for more information.
See ~/Library/Logs/Claude/mcp-server-amazon.log
License
The MIT license
7c3f420b3ffbOBSERVED · 2026-10-08Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add mcp-server-amazon --env AMAZON_PASSWORD=${AMAZON_PASSWORD} -- npx -y [email protected]{
"mcpServers": {
"mcp-server-amazon": {
"command": "npx",
"args": [
"-y",
"[email protected]"
],
"env": {
"AMAZON_PASSWORD": "${AMAZON_PASSWORD}"
}
}
}
}Exposed tools (8)
5 read · 1 write · 2 destructive. Blast radius: 2 tools can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
add-to-cart | write | Add a product to the Amazon cart using ASIN - You should always ask for confirmation to the user before running this tool |
clear-cart | destructive | Clear all items from the Amazon cart |
get-cart-content | read | Get the current cart content for a user - Always provide the product link when you mention a product in the response |
get-orders-history | read | Get orders history for a user |
get-product-details | read | Get detailed information about a product using its ASIN - Always provide the product link when you mention a product in the response |
perform-purchase | read | Checkout with the current cart and complete the purchase - |
remove-from-cart | destructive | Remove a single item from the Amazon cart by its ASIN, leaving all other items untouched - |
search-products | read | Search for products on Amazon using a search term - Returns a list of products matching the search term - Always provide the product link when you mention a product in the response |
Trust audit
CAUTIONgrade B · trust 89/100 Install with care. The audit found things worth knowing before you trust its output.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | WARN |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (1 observation(s))
- Network
- declared (1 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (8)
claude_desktop_config.json
mcp-server-amazon.log
clear-cart, remove-from-cart
/9j/4AAQSkZJRgABAgEASABIAAD/2wBDAAEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQH/2wBDAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEBAQEB
@modelcontextprotocol/sdk, cheerio, puppeteer, zod, @types/node, typescript
demo.gif
mocks/getProductDetails.html
mocks/searchProducts.html
Gates applied: no_behavioural_pass.
7c3f420b3ffbfull audit observations/trust-audit/mcp-server/rigwild__amazon.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | 7c3f420b3ffb | CAUTION | B | 89 | first audit |
Questions
What is the Amazon MCP server?
🛍📦 Unofficial Amazon Model Context Protocol Server (MCP) - Search products and purchase directly from Claude AI! ✨
What tools does Amazon expose?
8 in total: 5 read-only, 1 that write, and 2 that can delete or overwrite (clear-cart, remove-from-cart). Every one is listed on this page with its risk.
Is Amazon safe to connect to an agent?
With care. The audit graded it B (89/100) and found 8 things worth knowing before you trust this server, listed below with the exact line each was found on. Separately from the audit: 2 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does Amazon need?
It reads AMAZON_PASSWORD from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does Amazon run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as mcp-server-amazon at 1.0.0.
How current is this page?
The grade is for one exact copy of the source (7c3f420b3ffb), read on 2026-10-08. The repository is watched and re-audited when it changes.