Try the Search or browse Adidas products API

- Keyword search is best-effort relevance, not a guaranteed keyword match: an obscure `query` returns whatever Adidas's own search index surfaces (there is no reliable signal to distinguish a true match from Adidas's fallback results). - Requesting a `page` beyond the available result pages — or an unknown `category` — returns a not-found error rather than an empty page. - `products[].id` is the Adidas SKU to pass as `product_id` to `adidas-product` for full product-detail data. - `products[]` carries rich per-SKU metadata: `division` (brand line, e.g. Originals/Sportswear/Performance), `category` (e.g. shoes/clothing), `sport[]`, `generic_product_type[]`, `surface[]`, `available_sizes[]`, `color_variations[]`, and `orderable`/`preorderable`/`personalizable` availability flags. - `filters[]` are the facet groups Adidas's own search exposes (e.g. Shipping, Width, Brand, Price), each with `values[]` carrying a live `count`. `sort_options[]` lists the four sort orders with the currently selected one flagged. `breadcrumbs[]` is populated only for category browse (empty for keyword search). - **Applying a facet.** There is no separate filter parameter: facets are applied by composing them into the `category` value. Each `filters[].values[].slug` is the token to splice in — `category=women-black-athletic_sneakers` applies Color=Black to `women-athletic_sneakers`, and `category=women-athletic_sneakers-prime` applies Shipping=PRIME. The resulting `total_products` matches the facet value's advertised `count`. The token's **position within the slug varies by facet** (Color sits in the middle, Shipping at the end), so compose from a slug you have actually seen rather than assuming a fixed order, and read `filters[].values[].selected` to see what is already applied. - Use `filters[].key` (the facet's stable name, e.g. `searchcolor`, `multi_age_gender_en_us`) when reasoning about facets. `filters[].id` is Adidas's opaque per-deployment UUID — it is not stable across categories and cannot be used to build a request. - The facet set is **category-dependent**: `women-athletic_sneakers` exposes 10 groups, `men-clothing` 15, and some categories add a Size group that others omit. Always read the facets from the response rather than assuming a fixed list. - `total_pages` is derived from `total_products` and the 48-per-page size.