Food Intel

Google indexed web pages. We index the shelf.
The first unified AI food intelligence platform. Nutrition, ingredients, allergens, real-time prices, FDA recalls, commodity markets, and supply chain data — with full origin provenance on every record. 11 MCP tools for any AI to query.
8,000+Products
22,000+Nutrient Records
5,000FDA Recalls
11MCP Tools
10Data Sources
9AI Shelf Layers

The Problem

AI cannot answer "what's in my food?" today because the data is split across 15+ databases that don't talk to each other.

SourceHasMissing
USDA FoodData CentralNutrition for 380K+ foodsNo prices, no stores, no real-time
Open Food Facts4M+ barcodes, crowdsourcedSparse US coverage, stale data
openFDAAll US food recalls since 2004No products, no nutrition
Kroger APIReal-time prices, stock, aisle locationKroger-only, no MCP
BLS / USDA AMSCommodity & retail food pricesNo products, monthly lag
Instacart85K+ retailers, fused catalogCentralized, not sovereign, closed
New products can take months to appear in any public database. USDA: manufacturer submits voluntarily (weeks-months). Open Food Facts: someone scans the barcode (days-never). FDA: only if recalled. The gap between a product hitting the shelf and AI knowing about it is the core food transparency problem.

The Solution: Food Intel

Aggregate every free public source, add retailer API data, maintain origin provenance on every record, serve unified MCP tools to any AI.

Origin Policy: Every record carries its data source. Every API response includes full attribution — source URL, license, when it was last updated. No data without provenance.

Data Sources

SourceRecordsUpdatesLicense
Kroger Product API6,346 productsReal-timeFree (OAuth, 10K/day)
USDA FoodData Central593 productsQuarterlyPublic domain (CC0)
Open Food Facts1,045 productsContinuousOpen Database License
openFDA Recalls5,000 recallsWeeklyPublic domain
USDA AMS Market News20 pricesWeeklyPublic domain
BLS CPI Average Prices39 pricesMonthlyPublic domain

The 9-Layer AI Shelf

A physical grocery store has one layer — the box on the shelf. The AI shelf has nine. Every layer is a surface where value flows — to the consumer, from the brand, verified by the data.

Layer 1

Verified Facts

Nutrition, ingredients, allergens, country of origin, certifications, SNAP eligibility. Provable, from authoritative sources.

20 cal | 1g protein | Ingredient: Romaine Lettuce | Origin: USA | SNAP: Yes
Layer 2

Brand Message

The box we "deleted." Brand story, product positioning, sustainability claims, comparison messaging, new product launches. Brands pay to syndicate.

"Taylor Farms — family-owned since 1995, Salinas Valley. Hydroponically grown for year-round freshness."
Layer 3

Social Proof

Customer ratings, purchase patterns, repeat purchase rate, trending products. Community-generated, brand-amplified.

3.15/5 (39 reviews) | Frequently bought with: Caesar dressing, croutons | Purchases up 40% this week
Layer 4

Chef / Cookbook / Restaurant

Professional endorsement. "Featured in Ina Garten's Caesar Salad." Same romaine used by Sweetgreen. 47 recipes across NYT Cooking.

Used by: 3 Michelin-starred restaurants in NYC | Appears in 47 published recipes
Layer 5

Influencer / Creator

Social media endorsements, viral recipe trends, dietitian recommendations, fitness content. Paid and organic.

@healthyeats (2.3M followers): "my go-to romaine" | Featured in 12M-view TikTok lettuce wrap trend
Layer 6

Health / Diet Context

AI-computed diet compatibility, health goal matching, drug interactions, glycemic impact. Generated from nutrition data.

Keto: YES | Whole30: YES | Low-FODMAP: YES | High Vitamin K — monitor if on blood thinners
Layer 7

Supply Chain / Provenance

Farm identity, harvest date, cold chain verification, miles traveled, carbon footprint. The FSMA 204 / data wallet layer. What nobody else has.

Farm: Taylor Farms, Salinas CA | Harvested: May 20 | Cold chain: 34-38°F verified | 287 miles to store
Layer 8

Merchandising / Deals

Promotions, digital coupons, Ibotta cashback, bundle deals, loyalty points, price comparison across stores.

$0.50 off with Kroger Plus | Earn $1.00 Ibotta cashback | Walmart: $2.49 (Kroger matches)
Layer 9

Personal Health Context

The only layer that flows FROM the consumer. Active medications (GLP-1), recent surgery (gallbladder), chronic conditions, per-person allergies, life stage. Lives in your data wallet, not the retailer's cloud.

Steven: post-gallbladder → LOW FAT | Household member: GLP-1 → HIGH PROTEIN | Shared meal must satisfy BOTH
Layers 1-8 flow TO the consumer. Layer 9 flows FROM the consumer. The wallet is the boundary. The AI reads all 9 layers but Layer 9 requires YOUR permission every time. Eight layers of product data that brands pay to populate. One layer of personal data that you own and control.

Why This Is Not Google Search

DimensionGoogle SearchFood Intel
Data unitWeb pageProduct at a specific store on a specific shelf
PriceMaybe mentioned on a page somewhere$2.99 right now at Kroger On the Rhine
AvailabilityUnknownIn stock: HIGH
LocationStore addressAisle: VEGETABLES, Bay 16, Shelf 3, Right side
NutritionGeneric article about romaine20 cal, 1g protein, per this specific product
SafetyNews article about a recallReal-time recall check against FDA database
Comparison10 blue links to blog postsSide-by-side nutrition + price for 5 products
The query Google can't answer"Find me every SNAP-eligible, gluten-free, sub-$4 protein source in stock at my store sorted by protein-per-dollar"

MCP Tools

Any AI can discover and use these tools via /.well-known/mcp.json

ToolWhat It Does
food_searchSearch products by name, brand, or UPC barcode across all sources
food_nutritionFull nutrition panel + ingredients + allergens for a specific product
food_compareSide-by-side nutrition comparison of 2-5 products
food_recall_searchSearch FDA food recalls by product, company, or reason
food_recall_checkCheck if a product or brand has active recalls
food_commodity_pricesAgricultural commodity prices (terminal market + farm gate)
food_retail_pricesBLS average retail prices for staple grocery items
food_ingredients_lookupSearch products by ingredient or allergen
food_data_sourcesFull transparency: all data sources, record counts, last update times
food_new_productsRecently added products — shows the new product discovery lag
food_supply_chainFSMA 204-style traceability: farm-to-shelf journey

The Food Data Stack

The grocery industry has a 5-layer data infrastructure with no cryptographic spine. Every layer runs on trust-by-contract.

LayerWhoWhatRootz Opportunity
IdentifiersGS1GTIN / UPC / Digital LinkSigned product identity wallets
Syndication1WorldSync, Syndigo, SalsifyBrand-authored product contentVerified, signed brand claims
RetailersKroger, WalmartStore-specific price + stockWe already pull from Kroger API
AggregatorsInstacart Universal CatalogCross-retailer fused catalogSovereign per-product identity dissolves this moat
ConsumerApps, AI agentsWhat the shopper seesFood Intel — this is what we built
GS1 Digital Link is the wedge. QR codes on packaging already resolve to manufacturer-controlled URLs. That URL is where .well-known/ai endpoints belong — signed, agent-readable product identity. The food-product analog of Origin's company profiles.

API Access

Every endpoint returns origin provenance. Every response tells you exactly where the data came from.

GET /.well-known/mcp.json — MCP discovery

POST /mcp/tool — Execute any tool

GET /api/search?q=romaine+lettuce — Quick search

GET /api/sources — Data source transparency

GET /ai-context — AI orientation & skills