/**
* MCP Sampling Service
*
* Provides AI completion capabilities via connected LLM clients.
* Used for product analysis, comparisons, and recipe suggestions.
*/
import { CreateMessageResult } from "@modelcontextprotocol/sdk/types.js";
import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js";
import { z } from "zod";
export interface SamplingRequest {
messages: Array<{
role: "user" | "assistant";
content: {
type: "text" | "image";
text?: string;
data?: string;
mimeType?: string;
}
}>;
modelPreferences?: {
hints?: Array<{ name?: string }>;
costPriority?: number;
speedPriority?: number;
intelligencePriority?: number;
};
systemPrompt?: string;
includeContext?: "none" | "thisServer" | "allServers";
temperature?: number;
maxTokens: number;
stopSequences?: string[];
metadata?: Record<string, unknown>;
}
const CreateMessageResultSchema = z.object({
model: z.string(),
stopReason: z.string().optional(),
role: z.enum(["user", "assistant"]),
content: z.object({
type: z.enum(["text", "image"]),
text: z.string().optional(),
data: z.string().optional(),
mimeType: z.string().optional(),
}),
});
/**
* Request a completion from an LLM via the connected client
*/
export async function requestSampling(
mcpServer: McpServer,
request: SamplingRequest
): Promise<CreateMessageResult> {
try {
const response = await mcpServer.server.request({
method: "sampling/createMessage",
params: {
messages: request.messages,
modelPreferences: request.modelPreferences,
systemPrompt: request.systemPrompt,
includeContext: request.includeContext,
temperature: request.temperature,
maxTokens: request.maxTokens,
stopSequences: request.stopSequences,
metadata: request.metadata
}
}, CreateMessageResultSchema);
return response as CreateMessageResult;
} catch (error: unknown) {
const errorMessage = error instanceof Error ? error.message : String(error);
throw new Error(`Sampling request failed: ${errorMessage}`);
}
}
/**
* Extract text from a sampling response
*/
export function getResponseText(response: CreateMessageResult): string {
if (response.content.type === 'text') {
return response.content.text;
}
return '';
}
/**
* Creates a recipe suggestion request for a product
*/
export function createRecipeSuggestionRequest(productData: any): SamplingRequest {
const productName = productData.product_name || "this product";
const category = productData.categories || "food item";
const ingredients = productData.ingredients_text || "";
const nutriments = productData.nutriments || {};
const allergens = productData.allergens || "";
const brands = productData.brands || "";
return {
messages: [
{
role: "user",
content: {
type: "text",
text: `Generate recipe suggestions for ${productName} (${brands}). ` +
`Nutritional profile: Energy ${nutriments.energy_100g || "unknown"} kcal, ` +
`Fat ${nutriments.fat_100g || "unknown"}g, ` +
`Proteins ${nutriments.proteins_100g || "unknown"}g, ` +
`Carbs ${nutriments.carbohydrates_100g || "unknown"}g.\n\n` +
`Category: ${category}\nIngredients: ${ingredients}\nAllergens: ${allergens}`
}
}
],
modelPreferences: {
hints: [{ name: "claude-3" }, { name: "gpt-4o" }],
intelligencePriority: 0.8,
speedPriority: 0.5
},
systemPrompt: `You are a creative culinary nutritionist. Generate 4 recipe suggestions:
1. LOW-CALORIE: Light meal focusing on weight management
2. PROTEIN-RICH: Recipe for fitness enthusiasts
3. QUICK & EASY: Minimal prep and cooking time
4. FAMILY-FRIENDLY: Balanced meal for all ages
For each recipe, provide:
- Recipe name
- Ingredient list with measurements
- Brief preparation steps
- Approximate nutrition per serving
- Health benefit highlight`,
includeContext: "thisServer",
temperature: 0.7,
maxTokens: 3500
};
}
