I built an MCP server for Chinese fast-food nutrition data — and learned why nobody ships this

Описание и применение

Создан сервер MCP, предоставляющий данные о калорийности китайских фаст‑фудов, решая проблему отсутствия структурированных источников и снижая риск «галлюцинаций» при запросах к агенту.

Задача

Agents cannot reliably retrieve nutrition data for Chinese fast‑food chains because the data is scattered across brand mini‑programs, PDFs, and marketing pages in Chinese, with no machine‑readable field for per‑100ml or per‑cup values.

Результат

The MCP server provides a reliable, agent‑callable interface to Chinese fast‑food nutrition data, reducing hallucinations and enabling accurate calorie queries.

Рабочий процесс

• 1. Install the foodcalorie‑mcp package via pip or use the web lookup page.
• 2. Configure the server with a JSON file of brand‑menu data via the FOODCALORIE_EXTRA_DATA environment variable.
• 3. The server merges the free 103 items with the Pro 116 items, yielding 219 records with no ID collisions.
• 4. Agents call the six tools (search_food_by_name, get_daily_nutrition_summary, compare_foods, get_health_advice, browse_by_category, search_food_by_barcode) to retrieve nutrition information.
• 5. The tools expose coverage information and source URLs for each record, allowing audit and verification.

Доказательства

Ask any agent about Western chain food and it does fine: OpenFoodFacts, USDA, MyFitnessPal dumps are everywhere.

Источник

I built foodcalorie-mcp — an MCP server that exposes Chinese food and chain-restaurant nutrition as six tools any agent can call.

Источник

The free build ships 103 general Chinese food items compiled into the package. Brand-menu data is not in the package — it's a JSON file pointed to by an environment variable.

Источник

Every nutrition dataset has holes. The question is whether the vendor tells you where they are.

Источник

Pro records include the URL the number came from — brand nutrition pages, official menu PDFs, mini-program screens.

Источник