import { z } from "zod";
import type { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js";
export function registerAllPrompts(server: McpServer): void {
// Product Analysis Prompt
server.registerPrompt(
"analyze-product",
{
title: "Analyze Product",
description:
"Generate a detailed analysis of a product including its variants, pricing, inventory status, and recommendations for optimization.",
argsSchema: {
productId: z
.string()
.describe("The product ID (GID or numeric) to analyze"),
},
},
async ({ productId }) => ({
messages: [
{
role: "user" as const,
content: {
type: "text" as const,
text: `Please analyze the product with ID "${productId}". Use the get_product tool to fetch its details, then provide:
1. **Product Overview**: Title, vendor, type, and current status
2. **Pricing Analysis**: Current prices across variants, compare-at prices if set
3. **Inventory Status**: Stock levels, which variants are low or out of stock
4. **SEO Review**: Handle, title length, description quality
5. **Recommendations**: Actionable suggestions to improve the product listing
Start by fetching the product data.`,
},
},
],
})
);
// Order Summary Prompt
server.registerPrompt(
"summarize-orders",
{
title: "Summarize Recent Orders",
description:
"Generate a summary of recent orders including revenue, fulfillment status, and trends.",
argsSchema: {
timeframe: z
.enum(["today", "week", "month"])
.default("week")
.describe("The timeframe to analyze"),
limit: z
.number()
.default(50)
.describe("Number of orders to fetch"),
},
},
async ({ timeframe, limit }) => {
const dateFilter =
timeframe === "today"
? "created_at:>=today"
: timeframe === "week"
? "created_at:>=7_days_ago"
: "created_at:>=30_days_ago";
return {
messages: [
{
role: "user" as const,
content: {
type: "text" as const,
text: `Please provide a summary of orders from the ${timeframe}. Use the get_orders tool with query "${dateFilter}" and first=${limit}.
Include in your analysis:
1. **Total Orders**: Count and total revenue
2. **Fulfillment Status**: Breakdown by unfulfilled, partially fulfilled, fulfilled
3. **Payment Status**: Breakdown by paid, pending, refunded
4. **Top Products**: Most ordered items
5. **Average Order Value**: Calculate and compare to typical benchmarks
Start by fetching the order data.`,
},
},
],
};
}
);
// Inventory Health Check Prompt
server.registerPrompt(
"inventory-health",
{
title: "Inventory Health Check",
description:
"Analyze inventory levels across products and identify items that need attention (low stock, overstock, etc.).",
argsSchema: {
threshold: z
.number()
.default(10)
.describe("Low stock threshold to flag items"),
},
},
async ({ threshold }) => ({
messages: [
{
role: "user" as const,
content: {
type: "text" as const,
text: `Please perform an inventory health check. Use the get_inventory tool to fetch inventory levels.
Analyze and report:
1. **Low Stock Items**: Products with inventory below ${threshold} units
2. **Out of Stock**: Items with zero inventory
3. **Overstock Candidates**: Items with unusually high inventory (if detectable)
4. **Inventory Distribution**: Summary by location if multiple locations exist
5. **Action Items**: Prioritized list of items needing restocking
Start by fetching inventory data.`,
},
},
],
})
);
// Customer Insights Prompt
server.registerPrompt(
"customer-insights",
{
title: "Customer Insights",
description:
"Analyze customer data to identify patterns, VIP customers, and engagement opportunities.",
argsSchema: {
segment: z
.enum(["all", "repeat", "high-value", "recent"])
.default("all")
.describe("Customer segment to analyze"),
},
},
async ({ segment }) => {
const queryMap: Record<string, string> = {
all: "",
repeat: "orders_count:>1",
"high-value": "total_spent:>500",
recent: "updated_at:>=30_days_ago",
};
return {
messages: [
{
role: "user" as const,
content: {
type: "text" as const,
text: `Please analyze the "${segment}" customer segment. Use the get_customers tool${queryMap[segment] ? ` with query "${queryMap[segment]}"` : ""}.
Provide insights on:
1. **Customer Count**: Total customers in this segment
2. **Engagement Metrics**: Average orders, total spent
3. **Geographic Distribution**: Where customers are located
4. **VIP Identification**: Top customers by order count or spend
5. **Opportunities**: Suggestions for engagement or retention
Start by fetching customer data.`,
},
},
],
};
}
);
// Custom GraphQL Query Prompt
server.registerPrompt(
"custom-query",
{
title: "Custom GraphQL Query",
description:
"Help construct and execute a custom GraphQL query against the Shopify Admin API.",
argsSchema: {
intent: z
.string()
.describe("What data do you want to fetch or what action do you want to perform?"),
},
},
async ({ intent }) => ({
messages: [
{
role: "user" as const,
content: {
type: "text" as const,
text: `I need help with a custom Shopify Admin API query. My goal is: "${intent}"
Please:
1. Determine if this requires a query (read) or mutation (write)
2. Construct the appropriate GraphQL operation
3. Use the run_graphql_query tool to execute it
4. Explain the results
If you're unsure about the exact schema, start with a simpler query and iterate based on what fields are available.`,
},
},
],
})
);
}
