> ## Documentation Index > Fetch the complete documentation index at: https://docs.context.dev/llms.txt > Use this file to discover all available pages before exploring further. # Extract Product Data from Webpages > Detect product pages and extract a normalized product record with name, price, billing frequency, pricing model, features, images, and SKU. Context.dev's Product API takes any product page URL (Amazon, TikTok Shop, or any SaaS/D2C website) and gives you a structured product record: * Name, description, category of the product * Pricing models, listed price and ISO currency code * List of features, the product's target audience, and more I'm integrating Context.dev's Product API into my app. Help me: 1. Install the official SDK for my language (`context.dev` on npm / PyPI / RubyGems, `context-dev/context-dev-php` on Packagist, `github.com/context-dot-dev/context-go-sdk` for Go). 2. Read the API key from the `CONTEXT_DEV_API_KEY` environment variable. Never hardcode it. 3. Call `POST /brand/ai/product` with a single `url` to extract one product. Always check that `is_product_page` is true **and** that `product` is non-null before reading product fields — non-product pages (blog posts, category grids, listing pages) return `is_product_page: false` and `product: null`. 4. Use `maxAgeMs` (body field, defaults to 7 days, max 30 days, `0` for fresh) when I want a cached result. Use `timeoutMS` (max 300000) to bound long-running calls. 5. Branch error handling on `error_code` from the response body. The common ones are `INPUT_VALIDATION_ERROR` (400), `WEBSITE_ACCESS_ERROR` (400, target unreachable or blocked), `UNAUTHORIZED` (401), `REQUEST_TIMEOUT` (408), and `INTERNAL_ERROR` (500). Docs: [https://docs.context.dev/guides/extract-product-from-websites](https://docs.context.dev/guides/extract-product-from-websites) ## Prerequisites * **A Context.dev API key.** Sign up at [context.dev/signup](https://context.dev/signup), copy the key from the [dashboard](https://context.dev/dashboard) (prefix `ctxt_secret_`), and export it: ```bash theme={null} export CONTEXT_DEV_API_KEY="ctxt_secret_..." ``` * **An SDK (optional).** Install for your language, or skip the install and call directly with `curl`: ```bash TypeScript theme={null} npm install context.dev ``` ```bash Python theme={null} pip install context.dev ``` ```bash Ruby theme={null} gem install context.dev ``` ```bash Go theme={null} go get github.com/context-dot-dev/context-go-sdk ``` ```bash PHP theme={null} composer require context-dev/context-dev-php ``` ## Extract a single product `POST /brand/ai/product` takes a single product URL and returns whether the page is a product page, the detected ecommerce platform, and the extracted product record. ```bash cURL theme={null} curl -X POST https://api.context.dev/v1/brand/ai/product \ -H "Authorization: Bearer $CONTEXT_DEV_API_KEY" \ -H "Content-Type: application/json" \ -d '{"url": "https://www.allbirds.com/products/mens-wool-runners"}' ``` ```typescript TypeScript theme={null} import ContextDev from "context.dev"; const client = new ContextDev({ apiKey: process.env.CONTEXT_DEV_API_KEY }); const response = await client.ai.extractProduct({ url: "https://www.allbirds.com/products/mens-wool-runners", }); if (response.is_product_page && response.product) { console.log( response.product.name, response.product.price, response.product.currency, ); } else { console.log("Not a product page"); } ``` ```python Python theme={null} import os from context.dev import ContextDev client = ContextDev(api_key=os.environ["CONTEXT_DEV_API_KEY"]) response = client.ai.extract_product( url="https://www.allbirds.com/products/mens-wool-runners", ) if response.is_product_page and response.product: print(response.product.name, response.product.price, response.product.currency) else: print("Not a product page") ``` ```ruby Ruby theme={null} require "context_dev" client = ContextDev::Client.new(api_key: ENV.fetch("CONTEXT_DEV_API_KEY")) response = client.ai.extract_product( url: "https://www.allbirds.com/products/mens-wool-runners", ) if response.is_product_page && response.product p = response.product puts "#{p.name} #{p.price} #{p.currency}" else puts "Not a product page" end ``` ```go Go theme={null} package main import ( "context" "fmt" "os" contextdev "github.com/context-dot-dev/context-go-sdk" "github.com/context-dot-dev/context-go-sdk/option" ) func main() { client := contextdev.NewClient( option.WithAPIKey(os.Getenv("CONTEXT_DEV_API_KEY")), ) response, err := client.AI.ExtractProduct(context.TODO(), contextdev.AIExtractProductParams{ URL: "https://www.allbirds.com/products/mens-wool-runners", }) if err != nil { panic(err) } if response.IsProductPage && response.Product != nil { fmt.Println(response.Product.Name, response.Product.Price, response.Product.Currency) } else { fmt.Println("Not a product page") } } ``` ```php PHP theme={null} ai->extractProduct( url: 'https://www.allbirds.com/products/mens-wool-runners', ); if ($response->isProductPage && $response->product) { echo $response->product->name, ' ', $response->product->price, ' ', $response->product->currency, PHP_EOL; } else { echo 'Not a product page', PHP_EOL; } ``` 10 credits per successful call ### Request Parameters | Parameter | Type | Description | | ----------- | ---------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------- | | `url` | string (URI) | **Required.** The product page URL. | | `maxAgeMs` | integer (0-2592000000) | Return a cached result if one exists younger than this many milliseconds. Set to `0` to always scrape fresh. Max 30 days. Defaults to `604800000` (7d). | | `timeoutMS` | integer (1000-300000) | Abort with a 408 if the request exceeds this many milliseconds. Max `300000` (5 min). | ### Response ```json expandable theme={null} { "is_product_page": true, "platform": "generic", "product": { "name": "Men's Wool Runner - Natural Grey (Light Grey Sole)", "description": "The Allbirds Men's Wool Runner is a classic casual sneaker made from soft, responsibly-sourced Merino wool that is breathable, lightweight, and machine washable…", "price": 110, "currency": "USD", "billing_frequency": "one_time", "pricing_model": "flat", "url": "https://www.allbirds.com/products/mens-wool-runners", "category": "Footwear", "features": [ "Soft Merino wool upper", "Machine washable (remove insoles and laces)", "Sugarcane-based SweetFoam® midsole for support and cushioning", "Breathable and lightweight", "Responsibly sourced materials" ], "target_audience": ["Men"], "tags": ["wool", "sneaker", "men's shoes", "casual", "sustainable"], "image_url": "https://media.brand.dev/34f4b265-7198-4eaa-a4a3-e0b87f40de29.png", "images": [ "https://media.brand.dev/2177175e-aae3-4179-b59d-0df12fb1b59d.png", "https://media.brand.dev/04027441-7ec6-4928-94e5-5b74c21347b8.png", "https://media.brand.dev/f47cbb41-7c3b-4fa1-92c5-4764560dda3e.png" ], "sku": "MENS_WOOL_RUNNERS" } } ``` | Field | Type | Description | | --------------------------- | -------------- | ------------------------------------------------------------------------------------------------------------- | | `is_product_page` | boolean | Whether the URL was recognized as a product detail page. | | `platform` | string \| null | One of `amazon`, `tiktok_shop`, `etsy`, `generic`. `null` when not a product page. | | `product` | object \| null | The extracted product record. `null` when not a product page or when extraction failed (e.g. bot protection). | | `product.name` | string | Product name. | | `product.description` | string | Product description. | | `product.price` | number \| null | Numeric price for the listed currency. `null` when no price is shown. | | `product.currency` | string \| null | ISO currency code (e.g. `USD`, `EUR`). | | `product.billing_frequency` | string \| null | One of `monthly`, `yearly`, `one_time`, `usage_based`. | | `product.pricing_model` | string \| null | One of `per_seat`, `flat`, `tiered`, `freemium`, `custom`. | | `product.url` | string \| null | Canonical product page URL. | | `product.category` | string \| null | Single category label. | | `product.features` | string\[] | Bulleted feature claims, one sentence each. | | `product.target_audience` | string\[] | Audience tags (e.g. `Men`, `Small teams`). | | `product.tags` | string\[] | Free-form tags. | | `product.image_url` | string \| null | Primary image URL (CDN-hosted). | | `product.images` | string\[] | Up to 7 product image URLs (CDN-hosted). | | `product.sku` | string \| null | Stock keeping unit. `null` for SaaS, services, and other non-inventoried offerings. | ## Handle errors A non-2xx response returns an `{ message, error_code }` envelope: | Status | `error_code` | Meaning | What to do | | ------ | ------------------------ | --------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------ | | 400 | `INPUT_VALIDATION_ERROR` | Missing or malformed `url` | Validate input before the call. | | 400 | `WEBSITE_ACCESS_ERROR` | The target page was unreachable, blocked by a WAF, timed out during scraping, or served no usable content | Treat as "page cannot be extracted." Retry once; don't loop. | | 401 | `UNAUTHORIZED` | API key missing, invalid, or deleted | Re-check the env var and the dashboard. | | 408 | `REQUEST_TIMEOUT` | Page didn't finish extracting before `timeoutMS` (or 5 min default) | Retry once with backoff. | | 500 | `INTERNAL_ERROR` | Transient server error | Retry once with backoff; if it persists, contact support. | A 400 from omitting `url`: ```json expandable theme={null} { "error_code": "INPUT_VALIDATION_ERROR", "message": [ { "code": "invalid_type", "expected": "string", "received": "undefined", "path": ["url"], "message": "Required" } ] } ``` For the full catalog of error codes, see [Troubleshooting](/optimization/troubleshooting). ## Use cases * **Build a printable product catalog**: Use a website's sitemap to find all products, run `extract_product` on each page * **Procurement / ERP auto-fill**: when a buyer pastes a product URL into a procurement form, populate `sku`, `name`, `description`, `currency`, `price` and `features` automatically. ## Next steps Hide cold-hit latency from your users. Backoff strategies, client cache, and prefetch fallbacks. Caching, error handling, and key hygiene. Status codes, retry patterns, and common errors.