# Sommelier MCP

> Wine Pairing & Sommelier provides expert culinary advice, letting your AI agent act like a trained sommelier. You can discover ideal pairings for any dish or ingredient, get deep flavor profiles on specific wine types, and find top-rated bottle recommendations with prices.

## Overview
- **Category:** artificial-intelligence
- **Price:** Free
- **Tags:** sommelier, food-pairing, wine-recommendations, culinary-data, gastronomy, flavor-profiles

## Description

This MCP turns your AI client into an instant beverage consultant, helping you build menus, write content, or plan dinner parties around perfect food and drink matches. Instead of cross-referencing three different databases to match acidity levels with tannin structure, you simply ask the question. It handles pairing suggestions for everything from grilled salmon to pasta dishes, giving you specific bottle recommendations, ratings, and prices right away. If your workflow involves combining this MCP’s pairings data with a billing system or inventory check, remember that Vinkius ensures all credentials pass through a zero-trust proxy. This means your key never sits on a disk while the AI agent runs complex cross-platform automations for you.

It covers more than just matching: it details wine profiles—aroma notes, origin, and style—so you understand *why* something works. It's essential gear for anyone building any kind of culinary or lifestyle application.

## Tools

### get_dish_for_wine
Finds specific recipes or meals designed to pair well with a given wine type (e.g., Pinot Noir).

### get_wine_description
Retrieves detailed information covering the flavor profile, origin, and characteristics of any recognized wine variety.

### get_wine_pairing
Generates expert pairing suggestions for a dish or ingredient, including specific product recommendations with prices and ratings.

### recommend_wines
Provides targeted lists of wine products based on criteria you set, showing current ratings and costs.

## Prompt Examples

**Prompt:** 
```
What wine pairs best with grilled salmon?
```

**Response:** 
```
For grilled salmon, I recommend a Pinot Noir or a dry Chardonnay. The salmon's rich, oily texture pairs beautifully with the earthiness of Pinot Noir, while Chardonnay's buttery notes complement the grilled flavor. A top pick: Domaine Drouhin Oregon Pinot Noir 2021 — rated 4.2/5, around $32.
```

**Prompt:** 
```
What dishes pair well with an Argentinian Malbec?
```

**Response:** 
```
Argentinian Malbec is a bold, full-bodied red that pairs perfectly with:
🥩 **Grilled Steaks** (especially ribeye or sirloin)
🍔 **Gourmet Burgers** with blue cheese
🍖 **BBQ Pork Ribs** or brisket
The wine's dark fruit notes and smokiness complement charred and heavily seasoned meats.
```

**Prompt:** 
```
Give me a description of a typical New Zealand Sauvignon Blanc.
```

**Response:** 
```
Querying wine database ('get_wine_description')...
**New Zealand Sauvignon Blanc** (typically from Marlborough):
- **Profile**: Highly aromatic, crisp, and high in acidity.
- **Tasting Notes**: Overt notes of passion fruit, gooseberry, grapefruit, and fresh-cut grass or bell pepper (jalapeño descriptors are common).
- **Style**: Unoaked, best consumed young.
- **Pairings**: Goat cheese, asparagus, fresh seafood.
```

## Capabilities

### Suggest a dish based on wine
Input a specific bottle type (like Chardonnay) and receive several detailed, complementary meal suggestions.

### Get expert wine pairings
Provide an ingredient or main course and get precise recommendations for the perfect wine match, including product links.

### Detail a specific wine type
Input any recognized wine variety to retrieve its full flavor profile, origin story, and typical serving style.

### Find top-rated bottle suggestions
Get immediate recommendations for specific bottles, complete with star ratings and current retail prices.

## Use Cases

### Designing a high-end tasting menu
A restaurant manager inputs the main courses for three different dishes. The agent uses get_wine_pairing multiple times to ensure that each course is matched with an appropriate wine, creating a seamless and profitable pairing list.

### Writing educational content on terroir
A food writer needs detailed background info for a piece about Italian reds. They use get_wine_description to pull up the exact profile of Sangiovese, ensuring their article is technically accurate and rich with detail.

### Planning a dinner party budget
A host knows they are serving roasted lamb. They ask the agent to find suitable pairings using get_wine_pairing, then use recommend_wines to narrow down the options based on their total spending limit.

### Adjusting a recipe for pairing
A chef has developed a new spicy curry. They input 'spicy Indian curry' into get_wine_pairing, and the agent suggests that a crisp, high-acidity white wine will best counteract the heat.

## Benefits

- Stop guessing pairings. Use get_wine_pairing to tell your agent a dish and immediately receive specific product suggestions with prices and ratings.
- Don't waste time cross-referencing flavor charts. With get_wine_description, you pull up the complete story—from origin to tasting notes—for any wine type instantly.
- Reverse engineer menus. Need a pairing for that expensive bottle? Use get_dish_for_wine and it suggests perfect meals based on the wine's structure.
- Keep costs in check. The MCP can find specific bottles using recommend_wines, giving you immediate pricing data to make sure your recommendation is profitable.
- Build complex workflows by chaining this MCP with others. For example, pairing a meal (using get_wine_pairing) and then generating a blog post about it.

## How It Works

The bottom line is you stop guessing what works. You just ask the MCP to tell you.

1. Start by telling the MCP what you're working with: a dish (e.g., steak) or a wine (e.g., Malbec).
2. The agent sends this input to the specialized tools, which cross-reference global culinary data and flavor chemistry rules.
3. You get back an actionable list of perfect pairings, product suggestions, or detailed flavor breakdowns, ready for your final output.

## Frequently Asked Questions

**How do I use the get_wine_pairing tool with a recipe?**
Just input the dish or main ingredient into get_wine_pairing. The MCP will give you expert suggestions and help pinpoint specific bottles that match your menu's vibe.

**What if I already own a bottle of wine? Can I use get_dish_for_wine?**
Yes, absolutely. Use get_dish_for_wine. You tell it the wine you have—say, a Riesling—and it sends back several perfect recipes that complement those specific flavors.

**Can I find out about a type of wine without pairing it?**
You can use get_wine_description. Just tell the MCP the name (like 'Sauvignon Blanc') and it pulls up all the detailed information on its aroma, origin, and style.

**Does recommend_wines account for my budget?**
Yes. When you use recommend_wines, you can provide your target price range, and the MCP filters the results to only show bottles that fit both your palate and your wallet.

**If I give vague inputs, how does the `get_wine_pairing` tool refine its suggestions?**
The pairings rely heavily on detail. While simple items work, giving context—like 'seared scallops with lemon' instead of just 'scallops'—provides specific flavor cues that dramatically improve accuracy.

**When I use `get_wine_description`, is my query data stored or used for training?**
No. Your queries are processed through Vinkius’ zero-trust proxy and are never stored on disk. We process your wine information securely, maintaining absolute data privacy across all tools.

**Are there rate limits when calling `recommend_wines` for large batches of product checks?**
The MCP manages standard API limitations automatically. For high-volume tasks, you'll need to implement controlled calls or batch processing logic within your agent workflow.

**Does `get_dish_for_wine` require a full recipe, or can I just input the wine type?**
You should provide comprehensive descriptions. While you can reference the wine itself, adding details about how it's served (e.g., 'served with crusty bread') helps create more accurate dish suggestions.

**What wines are covered?**
The database covers all major wine varieties including Cabernet Sauvignon, Merlot, Pinot Noir, Chardonnay, Sauvignon Blanc, Riesling, Malbec, Prosecco, Champagne, and dozens more from wine regions worldwide.