# ShopSavvy for AI Agents - MCP Server & Skills

> Connect your AI assistant to ShopSavvy's database of 100M+ products. Works with Claude, Claude Code, ChatGPT, Codex, Gemini, and any MCP-compatible agent. Free to use.

*Source: https://shopsavvy.com/agents*

---

## Hero

**Badge:** Agent Skills & MCP Server

**Headline:** Shop Smarter with Any AI Agent

**Subheadline:** Connect your favorite AI assistant to ShopSavvy's database of 100M+ products. Get instant price comparisons from 40,000+ retailers in any conversation.

**Trust Indicators:**
- No signup required
- 100% free to use
- 100M+ products
- 40,000+ retailers

---

## Supported Agents

### Primary AI Assistants

Works with leading AI assistants:

| Agent | Provider |
|-------|----------|
| Claude Desktop | Anthropic |
| Claude Code | Anthropic |
| ChatGPT | OpenAI |
| Codex | OpenAI |
| Gemini | Google |
| Copilot | GitHub |

### AI-Powered IDEs & Editors

Plus AI-powered IDEs & editors:

| IDE/Editor | Provider |
|------------|----------|
| Cursor | AI-first IDE |
| Windsurf | Codeium |
| VS Code | Microsoft |
| JetBrains | IntelliJ & more |
| Zed | Fast editor |
| Continue | Open source |

---

## How It Works

Three simple steps to smarter shopping with any AI:

### Step 1: Connect the MCP Server
Add ShopSavvy to your AI agent using our MCP server URL or install the skill/plugin.

### Step 2: Ask About Products
Just ask naturally! "What's the best price for AirPods Pro?" or "Compare prices for this TV."

### Step 3: Get Instant Answers
Your AI searches our database of 100M+ products and shows you the best prices from 40,000+ stores.

---

## Research Workflows (What You Can Ask)

Four common research workflows your AI can help with:

### Find Best Price
- **Query:** "What's the best price for AirPods Pro?"
- **Tools used:** `product_lookup`, `product_offers`
- **Result:** Current prices from all retailers with deal assessment

### Research a Category
- **Query:** "Best gaming laptops under $1500"
- **Tools used:** `content_best_picks`, `universal_search`
- **Result:** Top picks with expert reviews and current prices

### Timing Advice
- **Query:** "Is now a good time to buy a TV?"
- **Tools used:** `product_price_history`, `content_buying_guide`
- **Result:** Price trends and when-to-buy recommendations

### Compare Products
- **Query:** "Compare iPhone 15 vs Samsung S24"
- **Tools used:** `product_lookup_batch`, `content_tldr_review`
- **Result:** Side-by-side specs, prices, and pros/cons

---

## Features (What Your AI Can Do)

Powerful shopping tools available through any connected AI agent:

### Find the Best Price
Search any product and instantly see prices from thousands of retailers. Never overpay again.

### Check Price History
See how prices have changed over time. Know if you're getting a good deal or if you should wait.

### Get Buy/Wait Advice
Find out if a product is at a good price now or if you should wait for a better deal. Timing matters.

### Research Any Product
Look up products by name, barcode, URL, or model number. Your AI finds it all.

### Expert Buying Guides
Access curated "Best Of" articles, TLDR reviews with pros/cons, and when-to-buy timing advice.

### Smart Recommendations
Get AI-powered product recommendations based on rankings, reviews, and price analysis.

---

## Example Conversations

See how AI agents help you shop smarter:

### Conversation 1: Best Price Lookup
**You:** What's the best price for AirPods Pro right now?

**AI Assistant:** I found AirPods Pro 2nd Gen at several retailers. The best price is $189.99 at Amazon (normally $249). This is $10 above the all-time low of $179.99 from Black Friday. Good deal if you need them now!

### Conversation 2: Buy or Wait
**You:** Is this a good time to buy a PS5?

**AI Assistant:** Based on price history, PS5 prices are currently stable at $449. The best prices typically appear during Black Friday ($349-$399) and occasionally during major sales. If you can wait, I'd suggest holding off until the next major shopping event.

### Conversation 3: Price Comparison
**You:** Compare prices for the Samsung S24 Ultra

**AI Assistant:** Samsung S24 Ultra (256GB) prices:
- Samsung.com: $1,299 (with trade-in: $799)
- Amazon: $1,199
- Best Buy: $1,199 (plus $100 gift card)
- Walmart: $1,199

Best value: Best Buy with the gift card bonus.

### Conversation 4: Best Picks
**You:** What are the best wireless earbuds right now?

**AI Assistant:** Based on expert reviews and current prices, here are the top picks:

1. AirPods Pro 2 - Best for iPhone ($189 at Amazon)
2. Sony WF-1000XM5 - Best sound quality ($248 at Best Buy)
3. Samsung Galaxy Buds 3 Pro - Best for Android ($179 at Samsung)

All three have excellent noise cancellation. The AirPods Pro 2 are at a great price right now, down from $249.

### Conversation 5: Product Reviews
**You:** What are people saying about the Sony WH-1000XM5?

**AI Assistant:** The Sony WH-1000XM5 headphones get excellent reviews:

Pros:
- Best-in-class noise cancellation
- Exceptional 30-hour battery life
- Very comfortable for long sessions
- Great call quality

Cons:
- No folding design (less portable)
- Premium price point

Bottom line: The best overall noise-canceling headphones for most people.

### Conversation 6: Sale Timing
**You:** When do TVs usually go on sale?

**AI Assistant:** TVs have predictable sale cycles:

Best times to buy:
- Super Bowl (late Jan/early Feb) - 15-25% off
- Amazon Prime Day (July) - 20-30% off
- Black Friday (November) - 25-40% off

Pro tip: New TV models release in spring, so last year's models drop significantly in March-April. Consider previous-year models for the best value.

---

## Pricing

### Free for Everyone

Generous limits for casual use, more power for heavy users.

### Free Tier - $0
No signup required.

- 10 requests per minute
- 100 requests per hour
- All tools included
- Perfect for casual shopping

### Premium Tier - $49/mo
Data API subscription. For Power Users.

- 100 requests per minute
- No hourly limit
- Priority support
- For developers & power users

View Data API plans at https://shopsavvy.com/data

---

## Installation Guide

Select your AI platform for step-by-step setup instructions.

### Quick Install (One Command)

No signup required. Works immediately.

**Claude Code:**
```
claude mcp add --transport sse shopsavvy https://api.shopsavvy.com/mcp/sse
```

**Any MCP Agent (NPX):**
```
npx shopsavvy-mcp
```
npm package: https://www.npmjs.com/package/shopsavvy-mcp

**Direct SSE URL:**
```
https://api.shopsavvy.com/mcp/sse
```

Or scroll down for detailed setup instructions for your specific platform.

### Or Just Tell Your AI

If you're using Claude Code, you can simply ask Claude to install it:

> "Please install the ShopSavvy MCP server so I can research product prices"

Claude will run the install command for you automatically.

---

### Claude Desktop Setup

**Step 1: Open your config file**

Find and open your Claude Desktop configuration file:

- **macOS:** `~/Library/Application Support/Claude/claude_desktop_config.json`
- **Windows:** `%APPDATA%\Claude\claude_desktop_config.json`
- **Linux:** `~/.config/Claude/claude_desktop_config.json`

**Step 2: Add ShopSavvy server**

Add this configuration (create the file if it doesn't exist):

```json
{
  "mcpServers": {
    "shopsavvy": {
      "command": "npx",
      "args": ["shopsavvy-mcp"]
    }
  }
}
```

**Step 3: Restart Claude Desktop**

Completely quit and reopen Claude Desktop. Look for the MCP indicator in the bottom-right corner of the chat input.

**Note:** Requires Node.js to be installed. The `shopsavvy-mcp` package auto-downloads on first run via npx.

---

### Claude Code Setup

**One-Command Install**

Run this command in your terminal:

```
claude mcp add --transport sse shopsavvy https://api.shopsavvy.com/mcp/sse
```

**Verify Installation**

Check that ShopSavvy is connected:

```
claude mcp list
```

Or type `/mcp` in Claude Code to see connection status.

**Scopes:** Add `--scope user` to make it available across all projects, or `--scope project` to share via `.mcp.json` with your team.

---

### ChatGPT Setup

**Requires:** ChatGPT Desktop app with Developer Mode enabled (added March 2025)

**Step 1: Enable Developer Mode**

In ChatGPT Desktop, go to **Settings > Connectors > Advanced** and enable **Developer Mode**.

**Step 2: Add MCP Connector**

Add a new MCP connector with the ShopSavvy server URL:

```
https://api.shopsavvy.com/mcp/sse
```

**Step 3: Start Using**

ChatGPT will now have access to ShopSavvy tools. Ask about products and prices naturally!

---

### OpenAI Codex Setup

**One-Command Install**

Run this in your terminal:

```
codex mcp add shopsavvy -- npx shopsavvy-mcp
```

**Or configure manually**

Add to your `~/.codex/config.toml`:

```toml
[mcp_servers.shopsavvy]
command = "npx"
args = ["-y", "mcp-remote", "https://api.shopsavvy.com/mcp/sse"]
```

---

### Google Gemini Setup

**Step 1: Open settings file**

Edit `~/.gemini/settings.json`:

**Step 2: Add ShopSavvy**

```json
{
  "mcpServers": {
    "shopsavvy": {
      "command": "npx",
      "args": ["shopsavvy-mcp"]
    }
  }
}
```

**Step 3: Restart Gemini**

Restart Gemini CLI or your Gemini-powered application.

---

### Cursor IDE Setup

**Step 1: Open MCP settings**

Press `Ctrl + Shift + P` (or `Cmd + Shift + P` on Mac), type "cursor settings", and find the MCP option in the sidebar.

**Step 2: Add configuration**

Or edit `~/.cursor/mcp.json` directly:

```json
{
  "mcpServers": {
    "shopsavvy": {
      "command": "npx",
      "args": ["shopsavvy-mcp"]
    }
  }
}
```

**Step 3: Restart Cursor**

Restart Cursor to activate the MCP server.

---

### VS Code / GitHub Copilot Setup

**Requires:** VS Code 1.99+ with GitHub Copilot. MCP is GA in VS Code 1.102+.

**Step 1: Open settings**

Go to **Settings > Extensions > GitHub Copilot > MCP Servers**, or edit your `settings.json`.

**Step 2: Add MCP server**

```json
"github.copilot.chat.mcpServers": {
  "shopsavvy": {
    "command": "npx",
    "args": ["-y", "mcp-remote", "https://api.shopsavvy.com/mcp/sse"]
  }
}
```

**Step 3: Use in Agent Mode**

Enable Copilot Agent Mode to use MCP tools. ShopSavvy tools will be available in your Copilot chat.

---

### Windsurf IDE Setup

**Step 1: Open config file**

Edit your Windsurf MCP configuration:

- **macOS:** `~/.codeium/windsurf/mcp_config.json`
- **Windows:** `%USERPROFILE%\.codeium\windsurf\mcp_config.json`

**Step 2: Add configuration**

```json
{
  "mcpServers": {
    "shopsavvy": {
      "command": "npx",
      "args": ["shopsavvy-mcp"]
    }
  }
}
```

**Or use the MCP Marketplace**

Go to **Settings > Cascade > MCP Servers** to browse and add servers.

---

## Available Tools

Once connected, your AI will have access to these shopping tools:

### Product Research

| Tool | Description |
|------|-------------|
| `product_lookup` | Find by barcode, ASIN, or URL |
| `product_search` | Search by keyword |
| `universal_search` | Enhanced search with rankings |

### Price Intelligence

| Tool | Description |
|------|-------------|
| `product_offers` | Prices from all retailers |
| `product_price_history` | Historical pricing trends |
| `product_schedule` | Set up price alerts |

### Expert Content

| Tool | Description |
|------|-------------|
| `content_best_picks` | Curated buying guides |
| `content_tldr_review` | Quick pros/cons verdicts |
| `content_qa_answers` | Product Q&A answers |
| `content_buying_guide` | When-to-buy timing advice |

---

## Other MCP-Compatible Platforms

ShopSavvy works with any platform that supports the Model Context Protocol, including:

- Continue.dev
- JetBrains IDEs
- Sourcegraph Cody
- Zed Editor
- Cline
- LibreChat
- MindPal
- Chainlit

For these platforms, use the MCP Server URL above and follow their documentation for adding SSE-based MCP servers.

---

## Frequently Asked Questions

### Which AI platforms are supported?
ShopSavvy works with any platform supporting the Model Context Protocol (MCP). This includes Claude Desktop, Claude Code, ChatGPT (with Developer Mode), OpenAI Codex, Google Gemini, GitHub Copilot, Cursor, Windsurf, VS Code, JetBrains IDEs, Zed, Continue.dev, and many more. MCP has become the standard for AI tool integrations, with 17,000+ community servers and widespread adoption.

### Do I need Node.js installed?
For most platforms, yes. Node.js is required to run the mcp-remote bridge that connects our SSE server to your AI tool. Claude Code is the exception since it has native SSE support and can connect directly without extra dependencies.

### Do I need a ShopSavvy account?
Nope! The MCP server works immediately without any signup. Just connect using the server URL and start asking about products. Anonymous users get generous rate limits.

### Is it really free?
Yes! The free tier includes 10 requests per minute and 100 requests per hour, which is plenty for most users. No credit card or account required. We believe everyone should have access to price comparison tools.

### What can I ask my AI about?
Anything about shopping! Ask for price comparisons across retailers, check if now is a good time to buy, view price history trends, look up products by barcode/URL/ASIN, set up price alerts, and get deal recommendations. Your AI will search our database of 100M+ products to help you.

### How do I get higher rate limits?
Power users and developers can subscribe to our Data API starting at $49/month. This gives you 100 requests per minute with no hourly limit, plus priority support. Visit our Data API page for all plans.

### Which retailers do you track?
Over 40,000 retailers including Amazon, Walmart, Best Buy, Target, eBay, Costco, Home Depot, Newegg, B&H Photo, and thousands more. If it's a major retailer in the US, we almost certainly track it.

### What is MCP?
The Model Context Protocol (MCP) is an open standard developed by Anthropic that lets AI assistants connect to external tools and data sources. It was donated to the Linux Foundation in December 2025 and has been adopted by major platforms including OpenAI, Google, GitHub, and many AI-powered IDEs.

---

## Call to Action

**Ready to Shop Smarter?**

Connect your favorite AI agent to ShopSavvy and start finding better deals.

Get Started Free at https://shopsavvy.com/agents#install
