Lexical for developers and AI agents

What it is for

Lexical makes small electrical home appliances in China and exports them, wholesale and under OEM and private label, to distributors in more than 100 countries. A buyer, a price-comparison tool or an AI assistant helping someone source appliances often needs the same few things: which lines Lexical sells, how a line is packed in its carton, and how to reach the export desk.

This API gives those answers as JSON, in the same words and with the same figures as the website, so a program does not have to read the pages. Each line comes with the address of its page, and every product has a link that opens the quote form with the line already filled in.

What it does not include

Prices are not published: they are quoted per order, through the export desk. Barcodes, HS codes and anything kept for verified customers are not available through this API, and neither are lines the website no longer shows.

Where to start

• /openapi.json: the OpenAPI 3.1 description of every endpoint, with its parameters and the shape of every answer. Load it into an API client or hand it to an agent as a tool definition.
• /llms.txt and /llms-full.txt: the whole website in plain text for language models, with the pages and the products.
• /.well-known/mcp: the MCP server. A browser or a script that asks for it with GET sees a short description of the server; an MCP client connects with POST.
• Base address of the API: https://project-26u0v.vercel.app/api/v1

Endpoints

Every endpoint answers with JSON and accepts an optional lang parameter (en, de, es, fr, pt, ru, zh, fa or ar) for names. The examples are shortened; the figures in them are only for illustration.

curl "https://project-26u0v.vercel.app/api/v1/products?category=Air%20Fryers&limit=1"

{
  "lang": "en",
  "total": 7,
  "count": 1,
  "offset": 0,
  "limit": 1,
  "nextOffset": 1,
  "products": [
    {
      "sku": "LAF-3068",
      "name": "Air Fryer",
      "brand": "Lexical",
      "category": "Air Fryers",
      "categoryLabel": "Air Fryers",
      "url": "https://project-26u0v.vercel.app/products/air-fryer-laf-3068",
      "photoUrl": "https://project-26u0v.vercel.app/api/image/…?w=960",
      "carton": {
        "piecesPerCarton": 4,
        "netWeightKg": 11,
        "grossWeightKg": 12,
        "volumeCbm": 0.65,
        "dimensions": null
      }
    }
  ]
}

curl "https://project-26u0v.vercel.app/api/v1/products/LAF-3068"

{
  "sku": "LAF-3068",
  "name": "Air Fryer",
  "brand": "Lexical",
  "category": "Air Fryers",
  "categoryLabel": "Air Fryers",
  "url": "https://project-26u0v.vercel.app/products/air-fryer-laf-3068",
  "photoUrl": "https://project-26u0v.vercel.app/api/image/…?w=960",
  "carton": {
    "piecesPerCarton": 4,
    "netWeightKg": 11,
    "grossWeightKg": 12,
    "volumeCbm": 0.65,
    "dimensions": null
  },
  "description": null,
  "isNew": false,
  "ceDocumented": true,
  "lang": "en",
  "quoteUrl": "https://project-26u0v.vercel.app/contact?sku=LAF-3068&product=Air%20Fryer#quote"
}

curl "https://project-26u0v.vercel.app/api/v1/categories"

{
  "lang": "en",
  "count": 2,
  "categories": [
    {
      "name": "Air Fryers",
      "label": "Air Fryers",
      "count": 4,
      "url": "https://project-26u0v.vercel.app/products?category=Air%20Fryers"
    },
    {
      "name": "Kettles",
      "label": "Kettles",
      "count": 3,
      "url": "https://project-26u0v.vercel.app/products?category=Kettles"
    }
  ]
}

curl "https://project-26u0v.vercel.app/api/v1/search?q=air%20fryer&limit=3"

{
  "q": "air fryer",
  "lang": "en",
  "count": 2,
  "results": [
    {
      "kind": "product",
      "title": "Air Fryer",
      "description": "Air Fryers · LAF-3068",
      "url": "https://project-26u0v.vercel.app/products/air-fryer-laf-3068",
      "sku": "LAF-3068"
    },
    {
      "kind": "category",
      "title": "Air Fryers",
      "description": "Category",
      "url": "https://project-26u0v.vercel.app/products?category=Air%20Fryers"
    }
  ]
}

curl "https://project-26u0v.vercel.app/api/v1/company"

{
  "name": "Lexical",
  "legalName": "Yiwu Lexical Electrical Appliances Co., Ltd.",
  "website": "https://project-26u0v.vercel.app",
  "headquarters": {
    "description": "Head office and showroom",
    "street": "571 Chouzhou North Road",
    "city": "Yiwu",
    "country": "China",
    "timeZone": "Asia/Shanghai",
    "hours": [
      {
        "days": "Monday – Saturday",
        "opens": "09:00",
        "closes": "24:00",
        "text": "09:00 – 24:00"
      }
    ]
  },
  "contact": {
    "email": "contact@lexical.com.cn",
    "quoteForm": "https://project-26u0v.vercel.app/contact#quote"
  }
}

Errors

Every error is JSON in one shape, with the HTTP status that fits. The hint says what to do next.

Example

{
  "error": {
    "code": "not_found",
    "message": "No product with SKU XYZ-0000 in the public catalogue.",
    "hint": "List the SKUs the catalogue holds with GET /api/v1/products."
  }
}

MCP server

AI agents that speak the Model Context Protocol can use the catalogue directly. The server uses the Streamable HTTP transport, needs no authentication and keeps no session. It answers every request with one JSON object.

Its address is https://project-26u0v.vercel.app/.well-known/mcp, and the same server answers at https://project-26u0v.vercel.app/api/mcp. Add it to an MCP client with a configuration like this one, or with one command:

{
  "mcpServers": {
    "lexical-catalogue": {
      "type": "http",
      "url": "https://project-26u0v.vercel.app/.well-known/mcp"
    }
  }
}

claude mcp add --transport http lexical-catalogue https://project-26u0v.vercel.app/.well-known/mcp

Tools

Each tool does what the endpoint of the same purpose does and returns its JSON as text.

Or try the handshake by hand:

curl -X POST "https://project-26u0v.vercel.app/.well-known/mcp" \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -d '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2025-11-25","capabilities":{},"clientInfo":{"name":"curl","version":"1.0.0"}}}'

The server speaks protocol versions 2025-11-25, 2025-06-18, 2025-03-26.