Skip to main content
Glama
Amazon Seller MCP Server - DataDoe

Amazon Seller MCP Server - DataDoe

Amazon Seller MCP Server by DataDoe

Hosted Amazon Seller Central, Vendor Central and Amazon Ads MCP server with read and write access. Connect Claude, ChatGPT, Cursor, Codex, Gemini and other AI tools to your live Amazon data. Then let your AI agent act on it: update listings, manage orders and optimize Amazon Ads campaigns. DataDoe handles Amazon API approval, sign-in and rate limits for you.

Book a demo

Amazon Seller & Vendor SP-API Selling Partner API Amazon Ads API MCP Server Read and Write AI clients supported AI skills License: MIT

๐Ÿ“… Want to see DataDoe in action? Book a demo call and we will show you how DataDoe works with your AI tools and your Amazon accounts.

๐Ÿ”— Start a 14-day free trial ยท ๐Ÿ“˜ Documentation ยท ๐Ÿ“Š Data scheme ยท โšก Actions ยท ๐Ÿ”” Webhooks ยท ๐Ÿง  AI skills ยท ๐ŸŽฅ Video demo


What is DataDoe MCP?

DataDoe connects your Amazon Seller Central, Vendor Central and Amazon Ads accounts and keeps your data up to date in ready-to-use tables. DataDoe MCP gives AI tools access to that data through the Model Context Protocol (MCP), the standard way for AI assistants to connect to other services.

Once connected, your AI assistant can:

  • answer questions about sales, profit, ads, inventory, traffic, returns, settlements and more, for each SKU and marketplace

  • build reports and dashboards, such as weekly briefings, profit and loss (P&L) statements, reconciliations and ad reviews

  • make changes on Amazon with Actions, such as updating listings, managing orders and shipments, and optimizing Amazon Ads

You can test every change before it runs. Each Action type stays off until you turn it on in DataDoe settings.

You do not need to install anything or write code. DataDoe runs the MCP server for you at https://mcp.datadoe.com/mcp/v1.

DataDoe MCP with Amazon Seller Central, Vendor Central and Amazon Ads data for AI agents through the SP-API

Related MCP server: mcp-amazon-sp-api

Get started in three steps

  1. Start a 14-day free trial of DataDoe and connect your Amazon accounts.

  2. Add DataDoe to your AI tool:

  3. Ask a question, for example "Show my top 10 ASINs by revenue last month across all Amazon marketplaces."

Who is DataDoe MCP for?

  • Amazon sellers get answers from their own data without opening Seller Central, whether they use Fulfillment by Amazon (FBA), fulfill orders themselves (FBM) or sell in many marketplaces.

  • Amazon vendors read Vendor Central data with the same tools as Seller Central data.

  • Agencies work with all the client accounts they connected in one place.

  • Operations and finance teams automate recurring reports and alerts.

  • AI builders and developers build agents, apps and dashboards on Amazon data without their own Amazon API integration.

What can you ask?

  • "What were my top 20 ASINs by revenue last month across all Amazon marketplaces?"

  • "Reconcile my Amazon settlements against orders for Q1 and show any differences."

  • "Which Amazon Ads campaigns had ACoS over 40% last week, and what did they spend?"

  • "Which products will run out of stock in the next 14 days?"

  • "Which search terms got the most clicks but no sales?"

  • "What is my real profit per SKU after fees, ads and product costs?"

With Actions turned on, your AI agent can also act on what it finds:

  • "Pause every campaign with ACoS over 50% last week."

  • "Add the search terms that wasted the most money as negative keywords."

  • "Update the bullet points of SKU ABC-123."

  • "Confirm shipment for order 123-4567890-1234567 with UPS tracking 1Z999..."

Ready-made AI skills

Skip the prompt writing. The DataDoe AI Skills repository has ready-made skills for common jobs, such as weekly sales briefings, profit reports, wasted ad spend checks, restock alerts, listing audits and Buy Box checks. Some skills can also apply changes. They always run a test first and wait for your approval.

Install a skill with:

npx skills@latest add Deltologic/datadoe-ai-skills --skill SKILL_NAME

Browse all skills and agents on the AI agents and skills page in the DataDoe Hub.

What Amazon data is available?

DataDoe MCP can read every table in DataDoe. There are more than 200 tables across Seller Central, Vendor Central and Amazon Ads, including:

  • profit and finance data, such as profit by SKU and by day, settlements, fees, reimbursements, payouts and invoices

  • orders and sales data, such as order line items, sales and traffic, returns and refunds

  • Amazon Ads data for Sponsored Products, Brands, Display and Television and for Amazon DSP, such as search terms, targets, budgets, impression share and change history

  • inventory and fulfillment data, such as Fulfillment by Amazon (FBA) inventory, restock recommendations, stranded and aged stock, inbound shipments, Amazon Warehousing and Distribution (AWD) and storage fees

  • listings and catalog data, such as catalog attributes, Buy Box, competitive offers and listing issues

  • Brand Analytics data, such as Search Query Performance, search terms, market basket and repeat purchases

  • Vendor Central data, such as purchase orders, sales, traffic and inventory, forecasts, Direct Fulfillment and margins

  • near real-time events, such as order changes, offer changes, listing status and inventory events

See every table and column in the interactive data scheme in the DataDoe Hub or in the data scheme index in this repository, which links to one page per table.

Actions: act on your Amazon account

Actions let your AI agent read and change your Amazon account through DataDoe. With more than 200 Action types, your AI agent can:

  • create, update, check and delete listings, update prices and images, and manage A+ Content

  • find orders, cancel order items, confirm shipments, buy shipping labels and message buyers

  • manage Fulfillment by Amazon (FBA), Multi-Channel Fulfillment, FBA inbound and External Fulfillment shipments

  • acknowledge Vendor Central purchase orders, submit shipments, and get labels and packing slips

  • find, create, update and remove Amazon Ads campaigns, ad groups, targets and ads

  • manage Amazon Ads budget and bid rules, audiences, attribution and recommendations

  • check balances and payouts, and work with Amazon action items

DataDoe protects your account in these ways:

  • every Action type stays off until someone in your Organization turns it on in Actions settings

  • every Action supports a dry run (dryRun: true) that checks the request without sending anything to Amazon

  • listing, A+ Content and Multi-Channel Fulfillment changes also have previews that Amazon checks without applying them

  • changes need Read and write access for the account

  • DataDoe records every Action so you can review it later

See every Action type and its fields on the Actions page in the DataDoe Hub, in the Actions guide or in the Actions list in this repository. The Action scheme file has the full JSON Schema of each Action.

How an AI agent runs an Action:

  1. Call actions_details_schema_get to get the fields the Action needs.

  2. Call actions_start with dryRun: true to check the request.

  3. Call actions_start to run the Action.

  4. Call actions_get to check the result.

Webhooks

DataDoe can also send events to your own systems as they happen, for example order updates, offer changes, listing status changes and inventory events. Use webhooks to start automations or keep your enterprise resource planning (ERP) or warehouse system in sync.

Read the webhooks guide, see all webhook messages in the DataDoe Hub or open the webhook scheme in this repository.

Pricing

DataDoe MCP is part of every DataDoe subscription. It costs $97 per month, with a 14-day free trial for new Organizations. Exports, Actions, webhooks and Amazon Marketing Cloud (AMC) queries use AI Tokens from your monthly balance. See Subscription and pricing.

Step-by-step guides for your AI tool

Start with the DataDoe MCP overview.

Chat assistants

Office apps

Coding agents and code editors

Autonomous agents and automation

Agent development kits

Technical setup

This section is for developers and AI tools without a dedicated guide. DataDoe MCP is a remote MCP server with:

  • the URL https://mcp.datadoe.com/mcp/v1

  • the Streamable HTTP transport, which clients that support only stdio or Server-Sent Events (SSE) can reach through mcp-remote

  • OAuth sign-in where your client supports it, or the header datadoe-mcp-key: <YOUR_DATADOE_MCP_KEY> with a key from the MCP integrations page

{
  "mcpServers": {
    "datadoe": {
      "url": "https://mcp.datadoe.com/mcp/v1",
      "headers": {
        "datadoe-mcp-key": "<YOUR_DATADOE_MCP_KEY>"
      }
    }
  }
}

Claude Code:

claude mcp add datadoe "https://mcp.datadoe.com/mcp/v1" --transport http --header "datadoe-mcp-key: <YOUR_DATADOE_MCP_KEY>"

Put the server name and URL before --header. To sign in with OAuth instead, leave out --header, run /mcp in Claude Code and choose Authenticate.

GitHub Copilot in VS Code (.vscode/mcp.json):

{
  "servers": {
    "datadoe": {
      "type": "http",
      "url": "https://mcp.datadoe.com/mcp/v1",
      "headers": {
        "datadoe-mcp-key": "<YOUR_DATADOE_MCP_KEY>"
      }
    }
  }
}

Access and security

Personal MCP keys and OAuth connections follow their owner's roles and access to accounts, Actions and data tables. Service account keys use their own permissions. Read more about keys and service accounts and roles and permissions.

To read personal data, also called personally identifiable information (PII), a user needs the Access PII data permission, and the Organization needs Amazon PII access.

If your Organization uses its own domain prefix, copy the full server URL from the MCP integrations page.

Available MCP tools

Tool

Area

What it does

datadoe_user_docs_table_of_contents_get

Docs

Lists the pages of the DataDoe documentation.

datadoe_user_docs_page_get

Docs

Returns one DataDoe documentation page.

organization_and_subscription_details_get

Account

Returns your Organization, plan, billing health and AI Token balances.

sellers_and_vendors_list

Accounts

Lists the Amazon Sellers and Vendors connected to DataDoe.

sellers_and_vendors_get

Accounts

Returns one Seller or Vendor, including whether you need to reconnect it.

sellers_and_vendors_rename

Accounts

Renames a Seller or Vendor.

sellers_and_vendors_invitation_create

Accounts

Creates a link to connect a new Amazon account or reconnect one.

sellers_and_vendors_invitations_get

Accounts

Lists pending invitations or returns one invitation.

sellers_and_vendors_invitation_revoke

Accounts

Revokes a pending invitation.

exports_sources_get

Data

Searches DataDoe tables by keyword.

exports_source_get

Data

Returns the columns and details of one table.

exports_create

Data

Queries a table with filters, grouping, totals and date intervals, and returns CSV or JSON.

exports_get

Data

Returns the status of an export.

exports_list

Data

Lists exports.

exports_raw_url_get

Data

Returns a download link for a completed export.

exports_raw_download

Data

Returns the content of a completed export.

exports_delete

Data

Deletes an export.

files_create

Files

Uploads a file used by Actions, such as a listing image, A+ image or video ad asset.

files_list

Files

Lists uploaded files.

files_get

Files

Returns details of one file.

files_download_url_get

Files

Returns a download link for a file.

files_delete

Files

Deletes a file.

actions_details_schema_get

Actions

Returns the fields an Action type needs and whether it is enabled.

actions_start

Actions

Starts an Action, or tests it with dryRun: true.

actions_acknowledge

Actions

Releases an Action whose submission to Amazon was uncertain, after you have checked it on Amazon.

actions_get

Actions

Returns the status and result of an Action.

actions_list

Actions

Lists past Actions.

cogs_upsert

Product costs

Adds or updates product costs (COGS) for a Seller Central account.

cogs_delete

Product costs

Deletes product costs.

cogs_missing_list

Product costs

Lists SKUs sold in the last 30 days without a product cost.

vendor_code_upsert

Vendor codes

Adds or updates vendor codes for a Vendor Central account.

vendor_code_delete

Vendor codes

Deletes vendor codes.

vendor_codes_missing_list

Vendor codes

Lists ASINs sold in the last 30 days without a vendor code.

kdp_royalty_parameters_upsert

KDP

Adds or updates royalty parameters for Kindle Direct Publishing books.

kdp_royalty_parameters_delete

KDP

Deletes KDP royalty parameters.

sqp_asins_get

Search Query Performance

Returns the ASINs tracked in Search Query Performance.

sqp_asins_add

Search Query Performance

Adds ASINs to Search Query Performance.

sqp_asins_remove

Search Query Performance

Removes ASINs from Search Query Performance.

plugins_get

Plugins

Returns saved Memories, Skills and Files.

plugins_memories_create

Plugins

Saves a Memory.

plugins_memories_edit

Plugins

Edits a Memory.

plugins_memories_delete

Plugins

Deletes a Memory.

plugins_skills_get

Plugins

Returns a Skill file.

plugins_files_get

Plugins

Returns the content of a File.

amc_workflows_find

Amazon Marketing Cloud

Lists AMC workflows.

amc_workflows_create

Amazon Marketing Cloud

Creates an AMC workflow.

amc_workflows_update

Amazon Marketing Cloud

Updates an AMC workflow.

amc_workflows_delete

Amazon Marketing Cloud

Deletes an AMC workflow.

amc_query_validate

Amazon Marketing Cloud

Checks an AMC query without running it.

amc_query_start

Amazon Marketing Cloud

Runs an AMC query.

amc_query_cancel

Amazon Marketing Cloud

Cancels an AMC query.

amc_query_results_find

Amazon Marketing Cloud

Lists AMC query results.

amc_query_result_get

Amazon Marketing Cloud

Returns one AMC query result with download links.

amc_schema_find

Amazon Marketing Cloud

Returns AMC tables and fields.

amc_operation_get

Amazon Marketing Cloud

Returns the status of an AMC operation.

amc_query_templates_find

Amazon Marketing Cloud

Lists ready-made AMC query templates.

amc_query_template_get

Amazon Marketing Cloud

Returns one AMC query template.

datadoe_support_ticket_create

Support

Sends a request to DataDoe support.

datadoe_support_ticket_check

Support

Reads a support request and its reply.

datadoe_support_ticket_update

Support

Adds information to a support request.

The tool definitions file has the input and output schema of every tool. Amazon Marketing Cloud (AMC) is in beta. Read the Amazon Marketing Cloud guide.


About this repository

This repository is the public schema facade of DataDoe MCP: a copy of everything DataDoe MCP offers, without access to any data. MCP registries, users and AI agents use it to see what DataDoe MCP can do before connecting. DataDoe hosts DataDoe MCP at mcp.datadoe.com. This repository does not connect to Amazon.

This repository contains:

Path

Contents

scheme/mcp-tools.json

Every tool of the hosted server, as MCP clients see it.

scheme/data-scheme.md, scheme/data-scheme/, scheme/data-scheme.json

Every DataDoe table with its description and columns.

scheme/actions.md, scheme/actions.json

Every Action type with its limits, result and details schema.

scheme/webhooks.json

Every webhook message type with its fields.

src/index.ts

The facade MCP server.

The facade server:

  • lists the same tools as the hosted server

  • answers exports_sources_get, exports_source_get and actions_details_schema_get from the data scheme and Action scheme in scheme/

  • offers the data scheme, Action scheme and webhook scheme as MCP resources (datadoe://scheme/...)

  • returns an error for every other tool, because it has no access to Amazon data

Run the facade locally

yarn install
yarn build
yarn start

The server uses stdio, so start it from an MCP client:

{
  "mcpServers": {
    "datadoe-facade": {
      "command": "node",
      "args": ["/absolute/path/to/datadoe-mcp/dist/index.js"]
    }
  }
}

Keep the facade up to date

Scripts generate the files in scheme/, so do not edit them by hand. To update them:

  • run yarn sync:scheme to download the data scheme, Actions and webhooks from the public DataDoe API specification

  • set DATADOE_SPEC_BASE_URL to read another DataDoe API, such as a local one, for example DATADOE_SPEC_BASE_URL=http://localhost:3001/api/v1/spec yarn sync:scheme

  • run yarn mcp:facade-snapshot <path to this repository>/scheme/mcp-tools.json from services in DataDoe core to update the tools

  • bump the version in src/index.ts and server.json when you publish changes


Documentation and resources

License

The schema facade in this repository is open source under the MIT license. DataDoe, Inc. provides the hosted MCP server, the DataDoe application and DataDoe infrastructure as proprietary software.


DataDoe is a product of DataDoe, Inc. ยท datadoe.com

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelogโ€ฆ

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      B
      quality
      C
      maintenance
      agentcentral is a hosted MCP server for Amazon sellers that connects Claude, ChatGPT, and other AI clients to Amazon Ads, Seller Central, inventory, orders, catalog, rankings, finance, and fulfillment data. It provides fast reads from pre-materialized Amazon data, scoped API keys, OAuth-based Amazon connections, encrypted tokens, per-tenant isolation, and guarded write tools with audit log
      100
      7
      MIT
    • A
      license
      C
      quality
      D
      maintenance
      MCP server that exposes the Amazon Selling Partner API to Claude Desktop and any other MCP client. It wraps the python-amazon-sp-api SDK and ships 55+ tools across 19 SP-API scopes with automatic pagination, throttle-aware retry and multi-marketplace support.
      53
      4
      MIT
    • A
      license
      Not graded
      quality
      D
      maintenance
      MCP server for Amazon Selling Partner API and Advertising API, enabling access to orders, inventory, pricing, ads, and reports via natural language.
      1
      MIT
    • F
      license
      Not graded
      quality
      D
      maintenance
      MCP servers for Amazon Seller Central and Amazon Ads API, enabling AI assistants to manage your Amazon account using natural language.
      3
      -