You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: products/tools/mcp-server/mcp-concepts.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -7,7 +7,7 @@ nav:
7
7
8
8
# MCP Concepts
9
9
10
-
The Model Context Protocol (MCP) defines three building blocks that servers can expose: **tools**, **resources**, and **prompts**. Knowing what each one does and when to use it is the starting point for working with Shopware's MCP server or building your own extensions.
10
+
MCP defines three building blocks that servers can expose: **tools**, **resources**, and **prompts**. Knowing what each one does and when to use it is the starting point for working with Shopware's MCP server or building your own extensions.
11
11
12
12
The full specification is available at [modelcontextprotocol.io](https://modelcontextprotocol.io/specification/).
# Build an AI Merchant Assistant using the Shopware Admin MCP Server
8
+
# Build an AI Catalog Quality Assistant using the Shopware Admin MCP Server
10
9
11
10
## Overview
12
11
13
-
The Shopware Admin MCP Server enables AI agents to interact with Shopware MCP. Instead of integrating directly with the Shopware Administration API, AI clients discover and execute Shopware capabilities exposed by the MCP Server.
14
-
15
-
In this tutorial, you'll build an AI-powered merchant assistant that can perform common administrative tasks through natural language while abiding by Shopware's authentication and authorization model.
12
+
In this tutorial, you'll build an AI-powered catalog quality assistant that helps merchants identify and improve incomplete or inconsistent product data using the Shopware Admin MCP Server.
16
13
17
-
At the end of this tutorial, your assistant will be able to:
14
+
Instead of implementing custom integrations against the Shopware Administration API, you'll use the Shopware Admin MCP Server to expose Shopware capabilities to any MCP-compatible AI client.
18
15
19
-
- Connect to a Shopware instance
20
-
- Discover available MCP capabilities
21
-
- Retrieve Shopware entities
22
-
- Execute merchant operations
23
-
- Abide Shopware ACL permissions
24
-
- Execute write operations safely
16
+
The assistant will help merchants maintain a high-quality product catalog by identifying missing or incomplete product information and enriching it using AI.
25
17
26
18
## Use case
27
19
28
-
You're developing an AI-powered administration assistant for a Shopware project.
20
+
Catalog quality has a direct impact on customer experience, search relevance, and conversion rates. As product catalogs grow, manually identifying and correcting incomplete product data becomes increasingly time-consuming.
29
21
30
-
Instead of implementing custom integrations against the Shopware Administration API, you'll use the `shopware-admin-mcp` server to expose Shopware capabilities to any MCP-compatible AI client.
22
+
In this tutorial, you'll build an AI assistant that can help merchants:
31
23
32
-
The resulting assistant can help merchants perform administrative tasks such as:
24
+
- Identify products without descriptions.
25
+
- Find products missing images.
26
+
- Detect missing SEO metadata.
27
+
- Identify incomplete product attributes.
28
+
- Review inconsistent catalog data.
29
+
- Generate or improve product descriptions.
30
+
- Enrich missing catalog information.
31
+
- Safely apply bulk updates after validation.
33
32
34
-
- Searching products
35
-
- Retrieving order information
36
-
- Updating product data
37
-
- Reviewing inventory
38
-
- Managing media
39
-
- Generating reports
40
-
41
-
Because the assistant communicates through the Shopware MCP Server, all requests follow Shopware's authentication, authorization, and extension mechanisms.
33
+
The assistant communicates with Shopware through the Shopware Admin MCP Server and performs all operations using the authenticated user's permissions.
42
34
43
35
# What you'll build
44
36
45
-
By the end of this tutorial, your architecture will look like this:
37
+
By the end of this tutorial, you'll have an AI assistant capable of:
38
+
39
+
- Connecting to the Shopware Admin MCP Server.
40
+
- Discovering available MCP capabilities.
41
+
- Searching and retrieving Shopware entities.
42
+
- Identifying catalog quality issues.
43
+
- Enriching catalog data.
44
+
- Previewing changes using Dry Run.
45
+
- Applying validated updates.
46
+
47
+
# Prerequisites
48
+
49
+
Before you begin, ensure you have:
50
+
51
+
- Shopware **6.7.11.0** or later
52
+
- Shopware Admin MCP Server configured
53
+
- API credentials with appropriate permissions
54
+
- An MCP-compatible AI client (Claude Desktop, Cursor, Codex, ChatGPT, or another compatible client)
55
+
56
+
If you haven't configured the Shopware Admin MCP Server yet, complete the following guides first:
57
+
58
+
- Install the Shopware Admin MCP Server
59
+
- Configure Authentication
60
+
61
+
# Architecture
46
62
47
63
```text
48
64
Merchant
49
65
50
-
↓
66
+
│
67
+
68
+
▼
51
69
52
-
Claude Desktop / Cursor / Codex
70
+
AI Client
71
+
(Claude, Cursor, Codex, ...)
53
72
54
-
↓
73
+
│
74
+
75
+
▼
55
76
56
77
Shopware Admin MCP Server
57
78
58
-
↓
79
+
│
80
+
81
+
▼
59
82
60
83
Shopware Administration API
61
84
62
-
↓
85
+
│
86
+
87
+
▼
63
88
64
89
Shopware Core
65
90
```
66
91
67
-
The assistant communicates with Shopware through MCP instead of calling the Administration API directly.
92
+
The Shopware Admin MCP Server exposes Shopware capabilities as MCP Tools, Resources, and Prompts that AI clients can discover automatically.
68
93
69
-
# Prerequisites
94
+
# Step 1 — Connect your AI client
70
95
71
-
Before you begin, ensure you have:
96
+
Connect your preferred MCP-compatible AI client to the Shopware Admin MCP Server.
0 commit comments