Zara Search Autocomplete Scraper

Get real-time search autocomplete suggestions from Zara.com for any keyword, in any of Zara's 100+ country/language locales. This Actor replicates the suggestions that appear in Zara's own search box as a shopper types — perfect for keyword research, search-UX testing, trend monitoring, and powering AI shopping assistants with valid, high-signal Zara search terms.

No proxies, no cookies, no reverse-engineering required — just call the Actor with a query and get clean, structured JSON back in seconds.

Why use this Actor?

• Instant autocomplete data – fetch the exact suggestions Zara's own search bar shows for any seed keyword.
• Global coverage – works across 100+ Zara locales (US, UK, Spain, France, Germany, Japan, Brazil, and many more).
• Section filtering – scope suggestions to Home, Woman, Man, or Kids.
• AI-agent and LLM/RAG ready – returns clean, minimal JSON that's easy to feed into vector stores, RAG pipelines, or use as a live tool call from an AI agent that needs to know what search terms are actually valid on Zara.
• MCP-compatible – works as a callable tool inside Model Context Protocol (MCP) servers and AI agent frameworks (LangChain, CrewAI, AutoGen, LlamaIndex, custom LLM tool-use setups) so your agent can query real-time Zara search suggestions on demand.
• No maintenance – built and maintained on the Apify platform, so it keeps working as Zara updates its site.

What data do you get?

For each run, the Actor returns the list of autocomplete suggestions Zara's search returns for your query — the same strings a shopper would see appear beneath the search box while typing. Example output for the query "shoes":

[
  { "query": "shoes" },
  { "query": "women shoes" },
  { "query": "red shoes" },
  { "query": "men shoes" },
  { "query": "white shoes" },
  { "query": "sale shoes" },
  { "query": "leather shoes" },
  { "query": "gold shoes" },
  { "query": "black shoes" },
  { "query": "barefoot shoes" }
]

A run summary (query list plus completion timestamp) is also saved to the key-value store under the SEARCH_SUMMARY key, so you can quickly check what was fetched and when without paging through the full dataset.

Input parameters

Parameter	Type	Required	Default	Description
query	string	Yes	–	The seed search term to get autocomplete suggestions for (e.g. "dress", "shoes", "jacket").
locale	string	Yes	en_US	Language/country locale, in language_COUNTRY format (e.g. en_US, es_ES, en_GB, fr_FR, ja_JP). 100+ locales supported.
section	string	No	HOME	Zara section to search within: HOME (all), WOMAN, MAN, or KID.

Example input

{
  "query": "dress",
  "locale": "en_US",
  "section": "WOMAN"
}

Use cases

• Keyword research & SEO – discover the exact long-tail search phrases real Zara shoppers use, straight from Zara's own autocomplete engine.
• AI shopping agents – give an LLM-powered shopping assistant a live tool for validating or expanding a user's vague query ("shoes" → "white shoes", "leather shoes", "sale shoes") before searching or recommending products.
• RAG pipelines – enrich a retrieval-augmented generation system with up-to-date, locale-specific Zara search vocabulary.
• Competitive & trend monitoring – track how Zara's suggested searches shift across seasons, sales, and locales.
• Search UX QA – validate what your own or a competitor's autocomplete should surface for a given seed term.

How to use this Actor

1. Provide a query, choose a locale, and optionally a section.
2. Run the Actor via the Apify Console, Apify API, scheduler, or an MCP/AI-agent integration.
3. Read the results from the dataset (items) or the SEARCH_SUMMARY record in the key-value store.

Run programmatically (Apify API)

curl "https://api.apify.com/v2/acts/pintostudio~zara-search-autocomplete/run-sync-get-dataset-items?token=<YOUR_TOKEN>" \
  -X POST \
  -H "Content-Type: application/json" \
  -d '{ "query": "dress", "locale": "en_US", "section": "WOMAN" }'

Use as an AI agent / MCP tool

This Actor can be wired into any AI agent framework or MCP server as a callable tool ("get_zara_search_suggestions"), letting your agent fetch live, accurate Zara search terms instead of hallucinating product keywords. Combine it with the Zara Product Search Actor for a full search → validate → fetch pipeline.

Output schema

Results are stored in two places:

• Dataset – one row per autocomplete suggestion, with a single query field.
• Key-value store (SEARCH_SUMMARY) – a JSON summary of all suggestions returned plus the run's completion timestamp.

Frequently asked questions

Does this use Zara's official API? It calls the same autocomplete endpoint Zara's website itself uses, so results match exactly what shoppers see in the search box.

How many suggestions are returned per query? Up to 10 suggestions per query/locale/section combination, matching Zara's own autocomplete limit.

Can I use this for markets outside the US? Yes — pass any of the 100+ supported locale values (e.g. es_ES for Spain, en_GB for the UK, fr_FR for France, de_DE for Germany, ja_JP for Japan).

Is this Actor suitable for AI agents and LLM tool-use? Yes. The output is minimal, predictable JSON designed to be easy to parse and reason over inside an LLM prompt, RAG context window, or agent tool response.

Related Actors

• Zara Product Search – full product search results by keyword.
• Zara Products by Category – scrape all products from a category page.
• Zara Similar Product – find visually/algorithmically similar products.

Support

Found an issue or have a feature request? Open an issue on the Actor's Issues tab and we'll get back to you.