Koongo MCP Is Live: Manage Your Ecommerce Operations with AI Assistants

Managing a multichannel ecommerce business usually means switching between dashboards, configuring integrations, mapping product attributes, creating feeds, checking product data, and managing marketplace orders.

Koongo MCP introduces a different way to handle these tasks through a conversation with your AI assistant.

Koongo MCP is now live and connects MCP-compatible AI assistants such as Claude, ChatGPT, Gemini, and Cursor to your Koongo account. Instead of navigating every setting manually, you can describe what you want to accomplish and let the AI assistant work with supported Koongo tools and data on your behalf.

You can use it to create marketplace integrations, configure product feeds, map attributes and categories, enrich missing product information, inspect your setup, and work with marketplace orders.

Koongo MCP is available to Koongo customers on the new pricing model, including the Free plan, with no separate MCP fee currently planned.

TL;DR: Key Takeaways

• Koongo MCP connects compatible AI assistants directly to Koongo, allowing you to work with your ecommerce setup through natural-language instructions.
• You can use it with MCP-compatible clients including Claude, ChatGPT, Gemini, and Cursor.
• Supported tasks include marketplace and advertising integrations, product feed creation, attribute mapping, category mapping, product enrichment, feed exports, and order management.
• Each MCP connection can use either Read-only or Full access, so you decide whether an AI assistant can only inspect your account or also make supported changes.
• When testing marketplace integrations, MANUAL product mode can be used to avoid publishing an entire matching catalog at once.
• Koongo MCP is included within Koongo’s current pricing model, including the Free plan, with no separate MCP fee currently planned.

Table of Contents

1. What Is Koongo MCP?
2. How Does Koongo MCP Work?
3. What Can You Do with Koongo MCP?
4. Read-only vs Full Access
5. Which AI Assistants Can You Connect to Koongo?
6. How to Connect an AI Assistant to Koongo
7. Practical Koongo MCP Use Cases
8. Why MCP Changes Multichannel Ecommerce Management
9. Frequently Asked Questions

What Is Koongo MCP?

Koongo MCP is a Model Context Protocol integration that allows compatible AI assistants to connect to a user’s Koongo account and interact with supported Koongo tools and data.

MCP, or Model Context Protocol, is an open standard designed to connect AI applications with external systems, tools, and data sources. You can learn more in the official Model Context Protocol documentation.

For Koongo merchants, the important part is what this enables in practice.

Traditionally, if you wanted to create a new product feed, configure a marketplace integration, map required product attributes, or inspect your setup, you would open Koongo and work through the relevant interface.

With Koongo MCP, your AI assistant becomes another way to interact with these capabilities.

Instead of searching for the right setting, you can make a request such as:

Create a Google Shopping feed for my store and help me map the required attributes.

Or:

I want to add an Amazon marketplace integration. Walk me through it one step at a time.

The assistant can then use the Koongo MCP tools available to your connection to work through the task.

How Does Koongo MCP Work?

Koongo MCP sits between your AI client and your Koongo account.

Once a compatible AI assistant is connected, it can access the Koongo capabilities permitted by that connection.

You can then describe a task in normal language. If the request is ambiguous, for example because you have multiple projects or several possible channels, the assistant can ask for clarification before continuing.

This changes the interface, but not the underlying ecommerce workflow.

For example, a marketplace integration may still require you to configure identifiers, product filters, categories, attributes, and the marketplace connection itself. The difference is that the AI assistant can guide you through those steps conversationally.

Marketplace login or authorization steps that require channel credentials are completed by the user in the browser. The AI assistant does not enter your marketplace credentials for you.

MCP therefore does not replace Koongo’s underlying marketplace and product feed management infrastructure. It provides a conversational way to work with supported parts of it.

If you are new to product feeds, read the Complete Guide to Product Feed Management to understand how product data is transformed and distributed across ecommerce channels.

What Can You Do with Koongo MCP?

Koongo MCP currently covers several core areas of multichannel ecommerce management.

1. Create marketplace and advertising integrations

You can ask the assistant to help create integrations for supported marketplaces and advertising channels.

For example:

Add an Amazon marketplace integration and walk me through the setup.

The assistant can create the integration and guide you through the available wizard steps, including identifiers, product filters, category mapping, attribute mapping, and supported order settings.

The actual marketplace authorization step remains in your browser.

Koongo’s existing Marketplace Integration Wizard supports integrations with marketplaces including Amazon, eBay, bol.com, Kaufland, and many others.

If you want to understand the marketplace workflow in more detail, see the Sell on Marketplaces with Koongo page.

2. Create and export product feeds

Koongo MCP can help create feeds for supported channels and work through the configuration process.

For example:

Create a Google Shopping feed for my store, help me map the required attributes, then generate the file and check it.

The assistant can help with tasks such as:

• choosing a feed template,
• creating the feed,
• configuring category mapping where required,
• mapping product attributes,
• setting supported filters and settings,
• generating the feed,
• verifying the feed,
• submitting it where the channel and integration method support submission.

This builds on Koongo’s existing Product Feed Management functionality for distributing and optimizing product data across ecommerce and advertising channels.

You can also read the Koongo Product Feed Management documentation for more detailed information about feed configuration.

3. Map product attributes

Different channels require different product fields and formats.

A marketplace or advertising platform may require attributes such as title, brand, price, GTIN, colour, material, condition, and other category-specific information.

With Koongo MCP, you can ask the assistant to identify missing required attributes and help map them to the appropriate fields in your store.

Show me which required attributes are still missing on my Google Shopping feed and help me map them.

This can be especially useful when expanding to a new sales channel whose product data requirements differ from those of your ecommerce platform.

4. Map product categories

Marketplaces and advertising channels often maintain their own category taxonomies.

Your store category might be called “Running Shoes”, while the destination channel expects products to be assigned to a specific branch of its own category tree.

Koongo MCP can inspect missing category mappings, search the available channel taxonomy, suggest suitable target categories, and create mappings where supported.

5. Enrich missing product attributes

Koongo MCP can also work with product enrichment.

The AI assistant can derive missing values such as colour, gender, or material from available product information and save the resulting values as additional attributes that can later be used for mapping or filtering.

This can help when a destination channel requires information that is not consistently structured in the original store catalog.

As with any AI-derived product data, merchants should review important values before relying on them across live sales channels.

6. Work with marketplace orders

For marketplaces that support order synchronization through Koongo, MCP can also help with order-related workflows.

Depending on the available connection, you can ask the assistant to view order information and manage supported order connections.

Fetch my latest orders.

Or:

Turn on order sync for my Amazon marketplace and help me configure the order settings.

The availability of specific order functionality depends on the marketplace and integration.

Read-only vs Full Access

Every Koongo MCP connection has its own access level.

Access mode	What the AI can do	Best suited for
Read-only	List, inspect, preview, verify and validate supported information without performing write actions.	Reporting, troubleshooting, account inspection and first-time testing.
Full Access	Read information and perform supported actions such as creating integrations, configuring feeds, mapping attributes and exporting feeds.	Merchants who want to actively manage their Koongo setup through an AI assistant.

Read-only access

A Read-only connection can inspect information but cannot use write tools.

If the AI attempts to perform a write action through a Read-only connection, Koongo refuses the operation.

This makes Read-only a useful starting point if you want to explore MCP for reporting, inspection, troubleshooting, or understanding your current configuration without allowing the assistant to modify it.

Full Access

A Full Access connection can both read information and perform supported changes.

This can include creating and configuring feeds, creating marketplace or advertising integrations, mapping attributes, mapping categories, exporting supported feeds, and other available write operations.

Access is configured per connection, not for the entire Koongo account.

For current information about permissions, see the Koongo MCP for AI Assistants documentation.

Which AI Assistants Can You Connect to Koongo?

Koongo MCP is designed to work with MCP-compatible AI clients.

Clients currently being used with Koongo MCP include:

• Claude
• ChatGPT
• Gemini
• Cursor

Connection options depend on the AI client and can include OAuth authentication or a static token.

Because MCP is an open protocol rather than a Koongo-specific AI interface, Koongo is not limited to a single AI provider. Other compatible MCP clients may also be connected where their MCP implementation supports the required connection method.

For broader technical information about MCP, see the official Model Context Protocol documentation.

For current Koongo setup instructions, refer to the Koongo MCP Installation and Connection documentation.

How to Connect an AI Assistant to Koongo

Before getting started, you need:

• a Koongo account,
• at least one Koongo project with your store connected and product data imported,
• an MCP-compatible AI client.

Koongo supports connection through OAuth or a static token, depending on the client.

OAuth

With OAuth, you authenticate through the browser and select the access level you want to grant to the connection.

You can choose between Full Access and Read-only.

Static token

A static token can be generated from the MCP section of your Koongo account.

You can also configure whether that token should use Read-only access before generating it.

Once connected, a simple first test is:

List my Koongo projects.

If the assistant can return the project information available to the connection, you are ready to start working with Koongo through MCP.

For exact client-specific instructions, use the Koongo MCP Help Center, as configuration can differ between AI clients.

Practical Koongo MCP Use Cases

The easiest way to understand Koongo MCP is through the tasks it can help with.

Use case 1: Set up a new marketplace integration

I want to add an Amazon marketplace integration. Walk me through it one step at a time.

The assistant can create the integration and guide you through the setup sequence.

If you are new to marketplace integrations, the Koongo marketplace integration page explains how Koongo connects stores to ecommerce marketplaces.

Use case 2: Find missing feed attributes

Show me which required attributes are still missing on my Google Shopping feed, then help me map each one to the right field from my store.

This combines inspection with guided attribute mapping.

Use case 3: Fix category mapping

Which of my store categories aren’t mapped yet? Suggest the best matching target category for each.

The assistant can inspect mapping gaps and work with the destination channel’s taxonomy.

Use case 4: Build a new product feed

Create a Google Shopping feed, include only in-stock products, help me map the required attributes and generate the feed.

Instead of treating feed creation as several disconnected tasks, the conversation can carry the context from one step to the next.

Use case 5: Inspect your Koongo account without changing anything

List my integrations and tell me which ones need attention.

Using a Read-only connection allows the assistant to inspect supported information without granting permission to modify the setup.

Use case 6: Test a marketplace integration with MANUAL mode

One important distinction when working with marketplace integrations is MANUAL vs AUTO product mode.

When testing a new setup, MANUAL mode can be used to control which products are published rather than immediately applying the setup to the full matching catalog.

Set this marketplace to MANUAL mode and help me test the integration with one product first.

Why MCP Changes Multichannel Ecommerce Management

Multichannel ecommerce platforms have traditionally been built around graphical interfaces.

That means users need to understand where functions live and how different parts of the system connect.

MCP adds another interface: natural language.

The important change is not that AI replaces product feeds, marketplace APIs, mappings, rules, or synchronization. Those systems still perform the underlying ecommerce work.

What changes is how a merchant can interact with them.

Instead of searching for a category mapping screen, you can ask:

Which categories still need mapping?

Instead of manually locating product filtering rules, you can ask:

Only include products that are currently in stock.

A conversational interface can be particularly useful for marketplace setups that involve identifiers, category taxonomies, required attributes, product filters, authorization, listing configuration, and order settings.

It can also lower the learning curve for team members who understand what they want to accomplish but do not yet know every part of Koongo’s interface.

Experienced ecommerce managers can use MCP as another way to inspect configurations, work with mappings, create feeds, and perform supported actions using an AI client they already use.

Koongo MCP Pricing and Availability

Koongo MCP is available within Koongo’s current modular pricing model and there is currently no separate MCP fee planned.

Koongo also offers a permanent Free Plan, which can be used to test the platform before moving to a paid setup.

Koongo module	Starting price	What it is designed for
Free Plan	€0	Testing Koongo with limited resources.
Feeds Module	€24/month	Product feeds for channels such as Google, Meta, TikTok, comparison engines, and other feed-based destinations.
Ads Module	From €39/month	Connecting and synchronizing product data with advertising platforms.
Marketplace Module	From €69/month per marketplace	Marketplace API integration with product and order management.

The Feeds Module currently includes five feeds for €24/month, while the Marketplace Module starts at €69/month per marketplace on the Starter plan. Additional marketplaces and resources can change the final monthly cost.

For the latest prices, limits, update frequencies, and module details, always check the official Koongo pricing page.

Explore Koongo MCP: MCP for AI Assistants in the Koongo Help Center.

Frequently Asked Questions

Conclusion

Koongo MCP adds a conversational way to work with multichannel ecommerce operations.

Instead of relying only on dashboards, menus, and configuration screens, merchants can use compatible AI assistants to interact with supported Koongo capabilities using natural-language instructions.

The underlying ecommerce infrastructure remains the same. Product feeds still need accurate data, marketplace integrations still need correct configuration, attributes and categories still need mapping, and marketplace orders still rely on reliable synchronization.

What changes is how you can manage those processes.

You can inspect your existing setup with Read-only access, create product feeds, configure supported marketplace integrations, work with product mappings, enrich missing attributes, or manage supported order workflows through an MCP-compatible AI assistant.

Explore Koongo MCP documentation

View Koongo pricing and start free

Continue reading →