How to Connect GTIN Barcode Lookup to Claude via MCP

💡 No install? Use cloud.anythingmcp.com directly. Sign in, click Connectors → GTIN / Barcode Lookup, paste your credentials, mint an MCP API key and you're done. No Docker, no git clone, no local server.

GTIN Barcode Lookup + Claude

Let your AI turn a barcode into product data: look up food and drinks, general retail goods, cosmetics and books by EAN, UPC or ISBN, or search by name or brand to find a product's GTIN. It reads free public databases (Open Food Facts, UPCitemdb, Open Beauty Facts, Google Books), with no account and no API key.

Prerequisites

See the full setup instructions baked into the connector (visible in the in-app store when you select the connector). The required environment variables for this connector are:

(none: public data, no API key)

Step 1 — Get credentials

1. In the connector store, pick GTIN / Barcode Lookup and install it. There is nothing to paste: every source is a public database that needs no account and no API key, and there is no Authorize with Provider step.
2. Try a barcode you know: ask the AI to run gtin_lookup_food with 3017620422003 (a jar of Nutella). It returns the product name, brand, quantity, categories, Nutri-Score and an image link.
3. From then on the AI uses the tool that fits the product: gtin_lookup_retail for electronics, household goods, toys and tools, gtin_lookup_beauty for cosmetics and personal care, gtin_lookup_book for books.

Which database covers what: gtin_lookup_food and gtin_lookup_search_food read Open Food Facts (food, drinks and packaged consumer goods; millions of products, multilingual, with brand, categories, quantity, labels, Nutri-Score, ingredients and an image). gtin_lookup_retail and gtin_lookup_search_retail read UPCitemdb (electronics, household, toys, tools and other retail goods, with title, description, brand, model, images and known marketplace offers). gtin_lookup_beauty reads Open Beauty Facts (cosmetics and personal care, with ingredients and the period after opening). gtin_lookup_book reads Google Books (title, authors, publisher, publication date, description and cover). Each database knows different products, so when one has nothing the AI tries the next.

From a name to a barcode: gtin_lookup_search_food searches by product name or brand (10 results by default, up to 50) and gtin_lookup_search_retail by keyword. Both return the barcode of every match, so you can get from a description to a GTIN.

Barcode formats and check digits: give the barcode as digits and keep leading zeros. EAN-13 (GTIN-13) has 13 digits, UPC-A has 12, and a UPC-A is also found as its GTIN-13 with a 0 in front. A book's GTIN is its ISBN-13, the 13-digit number starting with 978 or 979 printed with the barcode on the back cover, and that is what gtin_lookup_book expects. Older books often carry only a 10-digit ISBN: its ISBN-13 is 978, then the first nine digits, then a recalculated check digit. The last digit of every GTIN and ISBN-13 is a check digit computed from the others, so a single mistyped digit gives a code that does not exist rather than a different product.

When a code is not found: Open Food Facts and Open Beauty Facts answer with status 0, UPCitemdb with an empty items array. That only means this database does not know the product: the AI then tries the tool for another kind of product, or searches by name and brand. Open Food Facts and Open Beauty Facts are filled in by volunteers, so some products are missing and some fields are empty.

Rate limits: these free sources limit requests per IP address. UPCitemdb's free trial tier allows about 100 requests a day per IP, Google Books has a shared anonymous quota, and Open Food Facts asks callers to keep the load light. When UPCitemdb's daily limit is used up, the retail tools fail until the next day; food, cosmetics and book lookups are not affected. Every tool goes through the instance's proxy when one is configured, which spreads the load and avoids IP bans. Look up one barcode at a time and reuse results instead of asking for the same code again.

Read-only: all six tools only read. The only thing sent to the databases is the barcode or the search term.

The connector does not add or edit products in these databases, does not create barcodes and does not decode barcode images: it works with the digits of the code.

Step 2 — Install the adapter

curl -fsSL https://raw.githubusercontent.com/HelpCode-ai/anythingmcp/main/docker-compose.quickstart.yml -o docker-compose.yml
printf 'JWT_SECRET=%s\nENCRYPTION_KEY=%s\n' "$(openssl rand -hex 32)" "$(openssl rand -hex 32)" > .env
docker compose up -d

Step 3 — Add the connector in Claude

Your server URL: in AnythingMCP, open MCP Servers → the server this connector is on, and copy its URL (https://cloud.anythingmcp.com/mcp/…). Use it wherever this guide shows YOUR_SERVER_ID.

Recommended path: works on claude.ai web without editing any config file.

1. In Claude, open Customize → Connectors.
2. Click +, then Add custom connector.
3. Enter a name, e.g. GTIN Barcode Lookup, and the remote MCP server URL https://cloud.anythingmcp.com/mcp/YOUR_SERVER_ID, then click Add.
4. Click Connect and sign in to AnythingMCP (OAuth) to allow access.

All tools below appear in your chat: start typing prompts.

Available tools

Example prompts

• "What product has the barcode 3017620422003? Give me the brand, quantity, Nutri-Score and the main ingredients."
• "Find the barcode of Coca-Cola Zero in the 1.5 litre bottle."
• "Look up the book with ISBN 9780141036144: title, author, publisher and year of publication."
• "Our supplier's delivery note lists the codes 4006381333931, 5449000000996 and 012345678905. Which products are they, and which ones could you not find?"

FAQ

Does it work with Claude Code as well as Claude Desktop? Yes, point both at https://cloud.anythingmcp.com/mcp/YOUR_SERVER_ID.

Next steps

• Connector landing page
• Connect to ChatGPT
• Connect to Cursor
• Connect to Gemini
• Connect to Meta Muse