mcp-publix
# mcp-publix
An MCP (Model Context Protocol) connector for Publix grocery stores. Provides tools for searching products, finding store locations, and browsing weekly deals.
## Tools
### `product_search`
Search for products available at Publix grocery stores.
**Input:**
- `query` (string, required): The product name or category to search for.
**Example:**
```json
{ "query": "chicken" }
```
**Returns:** A list of matching products with name, price, and aisle location.
---
### `store_finder`
Find Publix store locations near a zip code or city.
**Input:**
- `location` (string, required): A zip code or city name.
**Example:**
```json
{ "location": "Atlanta" }
```
**Returns:** A list of nearby Publix stores with address, phone number, and hours.
---
### `weekly_deals`
Get the current weekly deals and specials at Publix.
**Input:**
- `category` (string, optional): Filter deals by category. Available categories: `produce`, `meat`, `dairy`, `snacks`, `beverages`, `bakery`. Omit to retrieve deals across all categories.
**Example:**
```json
{ "category": "produce" }
```
**Returns:** A list of sale items with original price, sale price, and savings amount.
---
## Installation
### From npm (once published)
```bash
npm install -g @striderlabs/mcp-publix
```
### From source
```bash
git clone <repo-url>
cd mcp-publix
npm install
npm run build
```
## Usage
### As a standalone MCP server (stdio transport)
```bash
npx @striderlabs/mcp-publix
# or after global install:
mcp-publix
```
### Claude Desktop configuration
Add the following to your `claude_desktop_config.json`:
```json
{
"mcpServers": {
"publix": {
"command": "npx",
"args": ["-y", "@striderlabs/mcp-publix"]
}
}
}
```
### Claude Code / MCP CLI
```bash
mcp add publix -- npx -y @striderlabs/mcp-publix
```
## Development
```bash
npm install # install dependencies
npm run build # compile TypeScript with esbuild
npm pack # create distributable tarball
```
## License
MIT
TDQS
Scored across 3 tools
Each tool has a clearly distinct purpose: product_search focuses on product details, store_finder on store locations, and weekly_deals on promotions. There is no overlap in functionality, making tool selection straightforward for an agent.
All tool names follow a consistent snake_case pattern with clear, descriptive verb_noun combinations (product_search, store_finder, weekly_deals). This uniformity enhances readability and predictability.
With 3 tools, this server is well-scoped for a grocery store assistant, covering key areas like product lookup, store location, and deals. Each tool earns its place without being overly sparse or bloated.
The toolset covers core grocery shopping needs—finding products, stores, and deals—with no obvious dead ends. A minor gap might be the lack of tools for managing a shopping list or checking inventory, but the existing tools support basic workflows effectively.