Ein auf dem Model Context Protocol (MCP) basierender Produktsuchserver, der es Claude ermöglicht, Produktkataloge anhand von Schlüsselwörtern zu durchsuchen und strukturierte Daten zurückzugeben.
Stellt drei Tools für Claude zur Verfügung:
Tool | Beschreibung |
| Suche nach Produkten anhand von Schlüsselwörtern, unterstützt Filter für Kategorien, Preise und Bewertungen |
| Listet alle Produktkategorien und deren Anzahl auf |
| Ruft vollständige Informationen zu einem einzelnen Produkt anhand der ID ab |
Related MCP server: MCP Merchant Scout
git clone <your-repo-url>
cd mcp-product-searchmacOS / Linux:
python3 -m venv .venv
source .venv/bin/activate
pip install "mcp[cli]"Windows:
python -m venv .venv
.venv\Scripts\activate
pip install "mcp[cli]"Erfordert Python 3.10 oder höher. Überprüfbar mit
python3 --version.
mcp dev server.pyDer Browser öffnet automatisch den MCP Inspector. Falls er sich nicht automatisch öffnet, rufen Sie die in der Konsole angezeigte Adresse manuell auf (normalerweise http://localhost:6274).
Verbindungsschritte:
Ändern Sie links unter Command den Pfad zum absoluten Pfad des Pythons in Ihrer virtuellen Umgebung:
macOS/Linux: /absoluter/pfad/mcp-product-search/.venv/bin/python
Windows: C:\absoluter\pfad\mcp-product-search\.venv\Scripts\python.exe
Geben Sie bei Arguments server.py ein.
Klicken Sie auf Connect. Wenn unten links Connected erscheint, war die Verbindung erfolgreich.
Klicken Sie oben auf den Tab Tools, wählen Sie ein Tool aus, geben Sie die Parameter ein und klicken Sie auf Run Tool.
Suchen Sie die Konfigurationsdatei:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
Fügen Sie den folgenden Inhalt hinzu (ersetzen Sie ihn durch den tatsächlichen absoluten Pfad):
{
"mcpServers": {
"product-search": {
"command": "/绝对路径/mcp-product-search/.venv/bin/python",
"args": ["/绝对路径/mcp-product-search/server.py"]
}
}
}Aktuellen absoluten Pfad anzeigen:
# macOS / Linux
pwd
# Windows
cdSpeichern Sie die Konfigurationsdatei und beenden Sie Claude Desktop vollständig, bevor Sie es neu starten.
claude mcp add product-search \
/绝对路径/mcp-product-search/.venv/bin/python \
/绝对路径/mcp-product-search/server.pyNach der Einbindung in Claude können Sie Fragen wie diese stellen:
帮我搜索苹果品牌的笔记本电脑
找一款评分 4.8 以上、价格不超过 300 美元的耳机
列出所有产品分类
查询产品 P003 的详细信息Claude entscheidet automatisch, welches Tool aufgerufen werden soll, und gibt basierend auf den zurückgegebenen strukturierten Daten eine Antwort.
searchParameter | Typ | Erforderlich | Standardwert | Beschreibung |
| string | Ja | — | Sucht in Produktnamen, Beschreibungen, Marken, Kategorien und Tags |
| string | Nein | — | Filtert nach Kategorie, z. B. "Laptops", "Headphones" |
| float | Nein | — | Maximalpreis (USD) |
| float | Nein | — | Mindestbewertung (0–5) |
| int | Nein |
| Obergrenze der Ergebnisse (maximal 50) |
Die Ergebnisse werden nach Bewertung (absteigend) und Preis (aufsteigend) sortiert.
Beispiel für Rückgabe:
{
"keyword": "apple",
"filters": { "category": "Laptops", "max_price": null, "min_rating": null },
"total_results": 1,
"products": [
{
"id": "P001",
"name": "Apple MacBook Pro 14-inch M3",
"category": "Laptops",
"brand": "Apple",
"price": 1999.0,
"currency": "USD",
"stock": 42,
"rating": 4.8,
"description": "14-inch Liquid Retina XDR display, M3 chip, 18GB RAM, 512GB SSD.",
"tags": ["laptop", "apple", "macbook", "m3", "professional", "portable"]
}
]
}list_categoriesKeine Parameter. Gibt alle Kategorienamen und die entsprechende Produktanzahl zurück.
get_productParameter | Typ | Erforderlich | Beschreibung |
| string | Ja | Produkt-ID, z. B. "P001" (Groß-/Kleinschreibung wird ignoriert) |
mcp-product-search/
├── server.py # MCP 服务器，定义工具
├── products.py # 产品目录与搜索逻辑
├── pyproject.toml # 项目依赖
└── README.mdDer aktuelle Produktkatalog besteht aus 12 Beispieldaten in products.py. Um diese durch echte Daten zu ersetzen, müssen Sie lediglich die Funktion search_products() anpassen; server.py muss nicht geändert werden.
Datenquelle | Anpassung |
Lokale JSON/CSV | Datei beim Start lesen und |
SQLite / PostgreSQL | Durchlauflogik durch SQL-Abfragen ersetzen |
E-Commerce API | HTTP-Anfragen innerhalb der Funktion senden |
Elasticsearch | ES-Volltextsuche-Schnittstelle aufrufen |
Retrieve full details for a single product by its ID (e.g. "P001").
Args: product_id: The product's unique identifier.
| Name | Required | Description | Default |
|---|---|---|---|
| product_id | Yes |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations exist; description implies read-only but fails to disclose any behavioral details beyond 'retrieve'.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is extremely concise, front-loading the core purpose and listing the parameter clearly.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the simple single-parameter input and presence of an output schema, the description is sufficiently complete for the task.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The description adds meaning to the product_id parameter with an example and context, compensating for 0% schema coverage.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states it retrieves full details for a single product by ID, distinguishing it from siblings like list_categories and search.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance on when to use this tool versus siblings like search or list_categories is provided.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Return all available product categories and their item counts.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description only states the basic return value, omitting additional behavioral details such as ordering, limits, or side effects.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
A single, clear sentence with no unnecessary words, perfectly front-loaded.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a simple read operation with no parameters and an output schema present, the description sufficiently covers what the tool returns.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Since there are no parameters, the description does not need to add parameter information. Baseline 4 for 0 parameters.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description explicitly states the verb 'Return' and resource 'all available product categories and their item counts', distinguishing it from sibling tools like 'get_product' and 'search'.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance is provided on when to use this tool versus alternatives like 'search' or 'get_product', leaving the agent to infer context.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Search products by keyword with optional filters.
Args: keyword: Search term matched against name, description, brand, category, and tags. category: Filter by exact category name (e.g. "Laptops", "Headphones"). max_price: Upper price limit in USD. min_rating: Minimum rating (0–5). limit: Maximum number of results to return (default 10, max 50).
| Name | Required | Description | Default |
|---|---|---|---|
| keyword | Yes | ||
| category | No | ||
| max_price | No | ||
| min_rating | No | ||
| limit | No |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries full burden. It discloses that keyword matches multiple fields (name, description, brand, category, tags) and explains filters, but does not mention that the operation is read-only, result ordering, or error behavior. Adequate but missing some context.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is concise, with a clear introductory sentence followed by structured 'Args' bullet points. It front-loads the purpose and efficiently details parameters without unnecessary text.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given 5 parameters and existence of an output schema, the description covers input well but omits output format details. It could mention response structure or behavior like empty results. Acceptable but not fully complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, so description must compensate. It fully explains each parameter beyond the schema: keyword matches specific fields, category is exact name, max_price is in USD, min_rating is 0-5, limit has default and max. This adds significant meaning.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states 'Search products by keyword with optional filters.' It specifies the verb 'search' and the resource 'products', distinguishing it from sibling tools 'get_product' (retrieves a single product) and 'list_categories' (lists categories).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage for searching products but does not explicitly contrast with siblings or provide when-not-to-use guidance. The context is clear from names, but the description lacks explicit exclusions or alternative recommendations.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.1.0get_productlist_categoriessearchScored across 3 tools
Each tool has a distinct purpose: get_product retrieves by ID, list_categories returns category overview, search does keyword-based filtering. No ambiguity in their roles.
All tool names follow a consistent verb_noun pattern (get_product, list_categories, search) using snake_case, making them predictable.
With 3 tools, the server is well-scoped for a search-focused service. Each tool earns its place without bloat or deficiency.
The tool set covers core search and retrieval operations (get by ID, list categories, keyword search). Missing a general 'list all products' operation, but search likely covers this with an empty keyword, though not documented; minor gap.
AMZScout Skill + MCP gives AI agents live access to real Amazon marketplace data across 14 Amazon marketplaces. Analyze any ASIN, validate product ideas, research niches, compare competitors, discover profitable keywords, and build data-driven PPC strategies using trusted Amazon insights instead of AI assumptions. Works with Claude, ChatGPT, Cursor, and any other MCP-compatible AI client. To connect, you'll need an AMZScout API plan and authorize your account. Get access and view pricing here: https://learn.amzscout.net/amazon-product-api-for-ai-agents
Search multi-merchant supply, checkout, and track orders via MCP.
Hosted Amazon Seller Central and Amazon Ads MCP server for Claude, ChatGPT, Cursor, and agents.
Connect Amazon Seller Central to Claude or ChatGPT via MCP. Orders, inventory, pricing, fees, FBA.
