MCP server
Amazon Selling Partner API MCP server
Amazon's SP-API MCP server helps sellers and developers explore the Selling Partner API, generate code and run API calls, with approval for writes.
- Official server
- Ecommerce
Why it matters
What it does and why it matters
Amazon's Selling Partner API is how sellers and their developers automate listings, orders, inventory and reports. It is large and complex. Amazon's MCP server for it lets an assistant look up API references, explore the catalog, run API calls, generate code samples, help migrate from older APIs and suggest optimizations.
For sellers with a developer or agency, it shortens the time to build integrations. For operations, it can answer questions using live API data.
The execute tool can call live endpoints. On the hosted connector, Amazon drafts write actions for human approval, which is a sensible default. We keep that approval step for anything that changes listings, prices or inventory, and use sandbox credentials while building.
Use cases
Best business use cases
Integration building
Catalog exploration
API migrations
Report questions
Optimization tips
Tools
Available tools
Tool names as published by the maintainer at review time. Versions change, so check the repository for the current list.
Read and write: Read and write: sp_api_execute can call live SP-API endpoints. On the remote connector, write actions are always drafted for human approval.
-
sp_api_referenceSearches official SP-API documentation in plain language.
-
sp_api_explore_catalogBrowses endpoints, parameters, schemas and examples.
-
sp_api_executeRuns live SP-API requests with auth and signing.
-
sp_api_generate_code_sampleGenerates code samples in Python, Java, JavaScript, C# or PHP.
-
sp_api_migration_assistantHelps migrate code between API versions.
-
sp_api_optimizeReviews code for SP-API best practices.
Workflows
Example workflows
Build an inventory sync
-
1
Describe the sync needed.
-
2
The assistant finds the right API operations.
-
3
It generates a code sample.
-
4
The developer tests with sandbox credentials.
-
5
The team reviews before going live.
Update a listing safely
-
1
Ask to change a listing attribute.
-
2
The assistant drafts the API call.
-
3
The connector holds it for approval.
-
4
A person approves.
-
5
The change is applied and confirmed.
Prompts
Example prompts
“Which SP-API operations do I need to update inventory quantities?”
“Generate a Python sample to fetch yesterday's orders.”
“We use a deprecated reports API. What should we migrate to?”
Setup
Installation and configuration
Examples for Claude Desktop and Claude Code. Replace the placeholder values with your own keys, and store secrets outside version control.
Claude Desktop: claude_desktop_config.json
{
"mcpServers": {
"sp-api-dev-assistant": {
"command": "npx",
"args": [
"-y",
"@amazon-sp-api-release/sp-api-dev-mcp",
"sp-api-dev-assistant-mcp-server"
],
"env": {
"SP_API_CLIENT_ID": "your-client-id",
"SP_API_CLIENT_SECRET": "your-client-secret",
"SP_API_REFRESH_TOKEN": "your-refresh-token"
}
},
"sp-api-workflow": {
"command": "npx",
"args": [
"-y",
"@amazon-sp-api-release/sp-api-dev-mcp",
"sp-api-workflow-mcp-server"
],
"env": {
"SP_API_CLIENT_ID": "your-client-id",
"SP_API_CLIENT_SECRET": "your-client-secret",
"SP_API_REFRESH_TOKEN": "your-refresh-token"
}
}
}
}Claude Code: terminal
claude mcp add sp-api-dev-assistant -e SP_API_CLIENT_ID=your-client-id -e SP_API_CLIENT_SECRET=your-client-secret -e SP_API_REFRESH_TOKEN=your-refresh-token -- npx -y @amazon-sp-api-release/sp-api-dev-mcp sp-api-dev-assistant-mcp-server- Hosted server: https://sellingpartner-ai.amazon.com/mcp (Amazon Selling Partner Connector, from github.com/amzn/selling-partner-agentic-toolkit).
- Local server: npx -y @amazon-sp-api-release/sp-api-dev-mcp sp-api-dev-assistant-mcp-server.
- Needs: Local: SP-API LWA app credentials (SP_API_CLIENT_ID, SP_API_CLIENT_SECRET, SP_API_REFRESH_TOKEN). Remote connector: Seller Central OAuth consent, limited to the user's Seller Central roles..
Security
Security considerations
-
Restricted keys
Use restricted keys or scoped sign-in that allow only the actions a workflow needs.
-
Test mode first
Build and test in sandbox or test mode before touching live money.
-
Approve money movement
Refunds, payouts, payments and deletions need a person to approve them.
-
Audit trail
Keep a record of API calls, especially writes to listings and inventory.
Related servers
Servers that pair well with it
Shopify Dev
Shopify's Dev MCP server helps developers build for Shopify: it searches Shopify docs and API schemas and validates GraphQL, components and themes.
WooCommerce
WooCommerce's official MCP support, in preview, exposes store abilities for products and orders through the WordPress MCP Adapter.
QuickBooks Online
Intuit's official QuickBooks Online MCP server lets AI search and create customers, invoices, bills and payments and pull financial reports.
Keep exploring
Solutions and services that use it
Related services
View all related services- MCP integration Connect AI assistants to your CRM, files, databases and tools through Model Context Protocol servers, with safe permissions.
- Ecommerce websites Online stores with clean product catalogs, simple checkout and the integrations that keep orders moving.
- Automation and Integrations CRM, payment and API integrations, MCP connections and workflows that stop people copying data by hand.
- Subscription platforms Products people pay for monthly or yearly, with plans, trials, upgrades and billing that just works.
- Billing and payments Checkout, invoices, subscriptions, refunds and the accounting exports your finance team needs.
- API development Well-documented APIs that let your product talk to apps, partners and AI tools reliably.
Solutions
View all solutions- AI support agent An assistant that answers routine questions from your own content, checks orders and hands anything else to a person.
- AI-assisted content workflows Briefs, first drafts, edits and repurposing in a workflow where people set the angle and approve every word.
- Forecasting dashboards Forecast demand from your sales history and seasonality, and flag items to reorder before they run out.
Industries
View all industries- Ecommerce Online stores, product data cleanup, support assistants, price monitoring and order automation for online retailers.
- Vacation rentals Booking websites, owner and guest portals, turnover scheduling and guest messaging for rental operators.
- Retail Stock forecasting, sales dashboards, feedback analysis and store tools for retailers with shops and online sales.
- Education and elearning Course platforms, student portals, content workflows and AI tutoring assistants for schools and training companies.
- Logistics and transportation Driver apps, order and shipment tracking, document processing and integrations for logistics and delivery companies.
- SaaS and startups MVPs, subscription billing, AI features, multi-tenant platforms and scaling support for founders and product teams.
Guides and articles
View all guides and articlesAI models
View all ai models- General purpose All-round language models for writing, summarizing, support, extraction and most everyday business tasks.
- ElevenLabs Eleven v3 ElevenLabs text-to-speech: the expressive Eleven v3 in 70+ languages, a real-time conversational version, and the low-latency Flash models, with voice cloning.
Glossary terms
View all glossary terms- API integration An API integration connects two or more systems through their APIs so data and actions flow between them automatically.
- Webhook A webhook is an automatic message one system sends to another's web address when something happens, such as a payment or a new order.
- API An API, or application programming interface, is a defined way for one piece of software to request data or actions from another.
- Large language model A large language model, or LLM, is an AI model trained on vast amounts of text that can understand and generate language, and often images and code.
- Model Context Protocol The Model Context Protocol, or MCP, is an open standard for connecting AI assistants to external tools and data in a consistent way.
- Prompt A prompt is the instruction and context you give an AI model to get the output you want.
FAQ
Questions people ask us
Have a question that is not here? Ask us directly.
Yes. It is provided by Amazon.
It can call live endpoints. On the hosted connector, writes are drafted for approval.
Sellers and developers building or running SP-API integrations.