This is the module that unlocks the most powerful Level 3 and 4 workflows for Amazon sellers. The Seller Labs Amazon MCP connects Claude directly to your Amazon data — no CSV exports, no manual file prep, no snapshots. Live data, on demand or on a schedule.

What the Seller Labs Amazon MCP Is

Seller Labs built the Amazon MCP (Model Context Protocol server) to give AI assistants — including Claude — direct query access to a seller’s Amazon Seller Central data.

It works through the Seller Labs Data Hub: a dedicated database instance that syncs your Amazon data (orders, inventory, advertising, profit, etc.) in near real-time. The MCP server sits on top of that database and exposes it to Claude as a queryable data source.

From Claude’s perspective, it is like having a direct connection to your Amazon business data. You can ask questions in plain English and get answers drawn from your real numbers.

What Data You Can Query

With the Seller Labs Amazon MCP connected, Claude can access:

• Orders: units sold, revenue, returns, order velocity by ASIN and time period
• Inventory: current FBA stock levels, days of cover, reorder signals
• Advertising: campaign performance, keyword-level ACoS, spend, impressions, clicks
• Profit: revenue, COGS, FBA fees, ad spend, estimated profit per unit and per ASIN
• Product performance: BSR trends, review counts, listing health indicators

What This Enables (That Was Not Possible Before)

With the Amazon MCP connected, you can ask Claude questions like:

• “What are my five worst-performing keywords by ACoS this week?”
• “Which ASINs are at risk of running out of stock in the next 30 days?”
• “How did my total ad spend change week over week, and which campaigns drove the change?”
• “What was my estimated profit margin on Product X last month after fees and ad spend?”
• “Show me any SKU where sales velocity dropped more than 20% compared to last month.”

These questions get answered from your live data — not from a CSV you exported three days ago.

And because these are just Claude prompts, you can schedule them. Your weekly PPC digest becomes a query against live data, not against an export you had to remember to pull.

How to Get Access

The Seller Labs Amazon MCP is available to Seller Labs customers. To get access:

How to Set It Up in VS Code

Once you have access, the setup process takes about 10-15 minutes. Watch the full walkthrough here:

The high-level steps are:

1. Get your API credentials from the Seller Labs dashboard
2. Add the MCP configuration to your VS Code Claude Code settings (a JSON snippet — copy-paste, no coding)
3. Authenticate with your Seller Labs account
4. Test the connection by asking Claude a question about your Amazon data

Example Queries to Test After Setup

After the MCP is connected, run these to confirm it is working:

"What are my top 5 ASINs by revenue over the past 30 days?"
        

"Show me my total ad spend and average ACoS for the past 7 days."
        

"Which of my FBA ASINs has the lowest days of cover right now?"
        

If Claude responds with actual numbers from your account, the MCP is connected and working.

What to Do If You Don’t Have the MCP Yet

If you are working through this course before getting MCP access, you can still practice the scheduled workflows from Modules 2 and 3 using CSV exports in your /reports folder. The workflows are structurally identical — the only difference is whether Claude queries live data or reads from a saved file.

When you get the MCP connected, updating your scheduled tasks to use live data instead of files takes just a few minutes per task.