AI Dev MCP Server EARLY ACCESS

Last updated:
Edit on GitHub

The AiDev module ships an MCP server that gives AI assistants live access to your running Spryker application: transfer objects, module interfaces, order management system (OMS) state machines, CSV data, and read-only database queries. This page describes what the server exposes, how to extend it, and how to debug it.

Install first

The ai-dev:mcp-server command exists only after the spryker-sdk/ai-dev module is installed in your project and its console commands are registered. To install the module and register the server in your assistant, see Installation.

About Model Context Protocol (MCP)

The Model Context Protocol (MCP) is an open-source standard for connecting AI applications to external systems. For Spryker developers, MCP allows AI assistants like Claude or Copilot to understand your project, improving the quality of AI-generated code and recommendations.

Learn more about MCP at modelcontextprotocol.io.

How the server runs

The module provides the server through the ai-dev:mcp-server console command, which runs inside your project’s Docker container over the MCP stdio transport:

docker/sdk console ai-dev:mcp-server -q

The command starts the server, registers all configured MCP tool plugins, and listens for requests from AI assistants. You do not usually run it by hand — your assistant starts it. For the registration steps per assistant, see Register the MCP server.

Available MCP tools

The AiDev module provides the following built-in tools that AI assistants can use:

Tool name Description
getTransferStructureByName Retrieves the structure of a Spryker transfer object by its name. Returns all properties with their types and metadata.
getTransferStructureByNamespace Retrieves the structure of a Spryker transfer object by its fully qualified namespace.
getInterfaceMethodsByNamespace Retrieves all method signatures, parameters, return types, and PHPDoc for a given interface FQN (Fully Qualified Name).
getOmsTransitionsByState Retrieves OMS state machine transitions for a specific state. Returns all transitions that start from the given state, optionally filtered by process name.
getOrderOmsTransitions Retrieves OMS state machine transitions for a specified order from the order’s current state. Helps identify the current state and possible transitions.
executeQuery Executes read-only database queries (SELECT, SHOW, DESCRIBE, EXPLAIN) for accessing project data without modification capabilities.
getSprykerModules Lists all available Spryker modules from project and vendor directories.
getSprykerModuleMap Retrieves detailed module metadata including class paths, method signatures, and extension points.
searchAlgoliaDocumentation Enables keyword-based Spryker documentation search through Algolia integration.
analyzeCsvFile Analyzes CSV file structure without loading full content. Returns headers, row count, and sample rows. Supports optional column analysis with unique values and null counts.
transformCsv Transforms and modifies CSV files with three operation modes: APPEND (add new rows), REPLACE (overwrite target), and UPDATE (modify existing rows in-place). Supports column mappings, row filters, value transformations, default values, and automatic backup creation.
deleteCsvRows Deletes rows from CSV files based on filter criteria with multiple operators (equals, not_equals, in, not_in, contains, not_contains, starts_with, ends_with, empty, not_empty). Includes safety checks and automatic backup creation.
splitOdsToCsv Splits ODS (OpenDocument Spreadsheet) files into separate CSV files per sheet. Skips empty sheets and returns details about created files. Useful for converting Google Sheets exports to Spryker-compatible CSVs.

AI assistants can automatically discover and use these tools when connected to the MCP server.

Extension points

The AiDev module provides plugin interfaces for extending the MCP server with custom functionality.

AiDevMcpToolPluginInterface

Implement this interface to add custom MCP tools that AI assistants can use to query or interact with your application.

Interface location: SprykerSdk\Zed\AiDev\Dependency\AiDevMcpToolPluginInterface

Integration: Register your tool plugins in AiDevDependencyProvider::getMcpToolPlugins():

<?php

namespace Pyz\Zed\AiDev;

use SprykerSdk\Zed\AiDev\AiDevDependencyProvider as SprykerAiDevDependencyProvider;
use Pyz\Zed\AiDev\Communication\Plugins\CustomAiDevMcpToolPlugin;

class AiDevDependencyProvider extends SprykerAiDevDependencyProvider
{
    /**
     * @return array<\SprykerSdk\Zed\AiDev\Dependency\AiDevMcpToolPluginInterface>
     */
    protected function getMcpToolPlugins(): array
    {
        return array_merge(parent::getMcpToolPlugins(), [
            new CustomAiDevMcpToolPlugin(),
        ]);
    }
}

Configuration

You can configure the AiDev module through the AiDevConfig class. Refer to the module’s configuration class for the available options and their default values.

Debug the MCP server

Before you connect the MCP server to an AI assistant, you can test and debug it with the MCP Inspector. The inspector provides a web interface to interact with your MCP server, test tools, and verify that everything works correctly.

Go to your Spryker project directory and run:

npx @modelcontextprotocol/inspector docker/sdk console ai-dev:mcp-server -q

This command:

  • Starts the MCP Inspector in your browser
  • Connects to your local MCP server
  • Displays all available tools
  • Lets you test tool calls interactively

To run the inspector with Xdebug:

npx @modelcontextprotocol/inspector docker/sdk cli -x console ai-dev:mcp-server
Node.js required

The MCP Inspector requires Node.js on your system. The npx command automatically downloads and runs the inspector without a global installation.

MCP Inspector listing the Spryker MCP tools