Mealary
Powered by USDA FoodData Central

Recipe in.
Nutrition facts out.

Paste any recipe and get accurate per-serving nutrition and a print-ready FDA Nutrition Facts label in seconds — computed from real USDA data, with an API and MCP server for AI agents.

No credit card · 7,800+ USDA foods · Free forever tier

Your FDA Nutrition Facts panel and per-serving breakdown will appear here — computed from USDA FoodData Central.

FDA 21 CFR 101.9 rounding2016 Daily ValuesFALCPA big-9 allergensEvery number cites a USDA source

The nutrition math, done right

Not a chatbot estimate — a deterministic engine that food businesses and agents can trust for compliance-grade output.

Accurate USDA analysis

Ingredient parsing, household-unit → gram conversion with food densities, and per-serving aggregation across 7,800+ USDA foods.

Real FDA labels

A standards-compliant Nutrition Facts panel with %DV, Daily Values, allergen 'Contains' line and a descending-weight ingredient statement.

Cited, verifiable sources

Each ingredient links to its exact USDA FoodData Central record — the trust surface humans and LLMs both look for.

Allergen detection

Automatic screening for the FALCPA big-9 allergens (incl. sesame) straight from your ingredient list.

API + MCP for agents

A single-key JSON API, OpenAPI spec, llms.txt and a hosted MCP server so any AI agent can compute nutrition mid-task.

Made for scale

Structured errors, idempotency keys, clear rate limits and an inspectable request log — usage-metered so agents can pay per call.

For AI agents & developers

The nutrition tool your agent reaches for

Meal-planning, recipe and food-tech agents need real nutrition mid-task. Mealary is built to be that capability: predictable schema, single API key, a public sandbox key, and a hosted MCP server. Docs are the onboarding.

curl -X POST https://mealary.com/api/v1/analyze \
  -H "Authorization: Bearer mly_test_sandbox" \
  -H "content-type: application/json" \
  -d '{
    "ingredients": [
      "2 cups all-purpose flour",
      "1 cup sugar",
      "2 large eggs"
    ],
    "servings": 12
  }'

# → per-serving nutrition_facts, %DV,
#   allergens, and cited USDA sources

Free

For home cooks, bloggers & trying things out.

$0

  • ✓ Unlimited nutrition analysis
  • ✓ Watermarked FDA label preview
  • ✓ 100 API calls / month · sandbox key
Start free
Most popular

Pro

For food businesses & agent builders.

$29/mo

  • ✓ Print-ready label export (PDF/PNG/SVG)
  • ✓ Unlimited saved recipes
  • ✓ 10,000 API calls / month · metered overage
  • ✓ MCP server + priority support
See pricing

Frequently asked

Where does Mealary's nutrition data come from?

Every value is computed from the USDA FoodData Central database — the U.S. government's authoritative food-composition dataset. Each ingredient links back to the exact USDA food record so your numbers are verifiable, not an AI guess.

Are the labels FDA-compliant?

Mealary applies the FDA rounding rules from 21 CFR 101.9 and the 2016 Daily Values to produce a standard Nutrition Facts panel with %DV, an allergen 'Contains' statement, and an ingredient list in descending order of weight. It's a computation tool — always verify against your final formulation before printing.

What's the difference between free and Pro?

Free gives you unlimited on-screen nutrition analysis and a watermarked label preview. Pro ($29/mo) unlocks clean, print-ready label export (PDF/PNG/SVG), unlimited saved recipes, and the metered API + MCP server for building automations and AI agents.

Can an AI agent use Mealary?

Yes — Mealary ships a public JSON API, an OpenAPI spec, an llms.txt, and a hosted MCP server with analyze_recipe, generate_label, search_foods and get_food tools. There's a public sandbox key so an agent can try it with zero signup.