AMAZON PUBLIC FETCH / MCP

Your bot.
Your PC.
Amazon.

Give Grok Bot a route through your own computer. Fetch public Amazon product pages using your network or a proxy you configure locally.

Download source ↓Get connected ↗

Python 3.11+ · Mac / Windows / Linux · MIT license

LOCAL ROUTE / 01

01 / THE REQUEST

Grok Bot

Send a product URL or ASIN.

Your existing computer connection

02 / YOUR MACHINE

Your PC does the fetch.

Local network or configured proxy.

HTTPS / public product page

↗

03 / THE SOURCE

Amazon

Public /dp/ pages. Fresh session.

RETURN TO BOT

Title · ASIN · Price · Features

Connection diagram. Access depends on Amazon.

LOCAL STDIO MCPNO API KEYNO ACCOUNT LOGINNO PROJECT TELEMETRY

THE WHOLE IDEA

Put the request
on your network.

When a cloud bot cannot read a product page, let your connected computer make the request.

Install the connector on that computer. Your bot launches it through an existing computer connection, or a local MCP client runs it directly. Product details come back as structured data.

Have an HTTP(S) proxy running on your PC? Point the connector at it. Otherwise, it uses your normal internet connection. Amazon can still block either route.

SMALL TOOLBOX

Three tools. One job.

PUBLIC PRODUCT DATA

01amazon_product

Title, ASIN, visible price and up to eight feature bullets.

START HERE

02amazon_fetch

Fetch status, engine, titles and size. Raw HTML when requested.

03health

Installed engines, proxy configuration and request limits.

RUN IT ON YOUR MACHINE

Install. Connect. Fetch.

Full README ↗

Install Python 3.11 or newer. Download and unzip the source, then open a terminal in that folder.

01 / INSTALL

macOS & Linux

python3 -m venv .venv
.venv/bin/python -m pip install .
.venv/bin/python -m mcp_server

The final command starts the local stdio server. It waits for an MCP client; it is not an interactive prompt.

01 / INSTALL

Windows PowerShell

py -3 -m venv .venv
.\.venv\Scripts\python.exe -m pip install .
.\.venv\Scripts\python.exe -m mcp_server

No virtual-environment activation is needed. Use this Python executable in your client's configuration.

02 / CONNECT THE BOT

Use your existing
computer connection.

Tell Grok Bot to run this on the connected PC. If the client supports local stdio MCP, configure its executable and arguments as shown.

GENERIC MCP CONFIG · REPLACE THE LOCAL PATH

{
  "mcpServers": {
    "amazon-public-fetch": {
      "command": "/absolute/path/.venv/bin/python",
      "args": ["-m", "mcp_server"]
    }
  }
}

Windows: use the absolute path to .venv\Scripts\python.exe. Configuration format depends on your client.

03 / ASK FOR THE PRODUCT

“Use amazon_product on my connected computer to read this Amazon product URL. Return the title, price and features. Tell me if the request is blocked.”

ALREADY HAVE A LOCAL PROXY?

Bring your own route.

Set AMAZON_FETCH_PROXY in the server's environment. The address below is an example of a proxy already running on your computer.

Without a proxy variable, the connector uses your normal network. System and browser proxy settings are not detected automatically.

OPTIONAL / MACOS & LINUX EXAMPLE

Point at your local proxy.

AMAZON_FETCH_PROXY=http://127.0.0.1:8080 \
  .venv/bin/python -m mcp_server

For Windows and client env setup, see the proxy instructions →.

KNOW WHAT YOU'RE RUNNING

Local. Small. Specific.

Public product pages

Amazon US, UK, Canada, Germany and Japan. Full HTTPS product URLs or a 10-character ASIN. Tracking parameters are removed.

Your connection stays yours

No API key, project analytics or imported browser cookies. Amazon sees the egress IP. Your proxy and MCP client can retain their own logs.

Blocks are still blocks

No login, CAPTCHA solving or promise of access. Rate limits stop the fetch. This connector is not an anonymity tool.

GitHub hosts this page and the source. The connector runs on your PC. It does not create a remote computer connection or a public proxy service. Read the security model ↗

BUILT BY ABOMINATION81

Let your PC do the fetching.

Get it on GitHub ↗

Your bot.Your PC.Amazon.