SuperAPI AI Directory
Partner with us中文AI newsFavorites
All toolsAI developer platformshasdata-mcp

hasdata-mcp

Structured data APIs for Google Search, Maps, Trends and Flights, Amazon, Airbnb, Zillow, Yelp and more, returning clean JSON.

Open website

About this tool

Purpose

HasData MCP is a server based on the Model Context Protocol (MCP) that wraps HasData's web scraping and search APIs into data tools that AI clients can call directly. Users do not need to write scraping code, manage proxies, or handle parsing logic; they only need to make a request in the conversation to obtain structured data. The tools cover multiple domains such as search, e-commerce, maps, travel, real estate, recruitment, and social, returning JSON parsed for each site.

Main Capabilities

The server provides tool groups across search, e-commerce, maps, travel, real estate, recruitment, social, and other directions, and each group is also provided as a standalone server, one server per site, using the same tools with shorter URLs.

  • Web scraping: the standalone server is web-scraping-mcp.
  • Google series: google-search-mcp (all google_serp_* tools), google-images-mcp, google-maps-mcp, google-scholar-mcp, google-trends-mcp, google-flights-mcp, google-flights-deals-mcp, google-hotels-mcp.
  • Other search engines: bing-mcp, duckduckgo-mcp.
  • AI answer engines: chatgpt-mcp.
  • E-commerce: amazon-mcp, walmart-mcp, shopify-mcp.
  • Real estate: zillow-mcp, redfin-mcp.
  • Recruitment: indeed-mcp, glassdoor-mcp.
  • Travel and local: airbnb-mcp, booking-mcp, yelp-mcp, yellowpages-mcp.
  • Social: facebook-mcp, instagram-mcp, tiktok-mcp, youtube-mcp.

The tools are site-specific: for example, amazon_product understands ASIN, variants, and seller data, and google_maps_place understands business hours, coordinates, and ratings, rather than being a generic scraper. The server also comes with 20 prompts, allowing clients to directly provide ready-made workflows; and 151 resources, corresponding to parameters with fixed values among the 28 services, making it easy for clients to understand the vocabulary without consuming quota. Each URI has the form hasdata://<service>/<parameter>, and resources/list returns the current collection. These two types of lists can be read without an API key, so clients can obtain them before the user registers.

Usage

There are two authentication methods:

  1. OAuth (recommended): after connecting to the server, log in to your HasData account in the browser, with no need to copy a key.
  2. API key: send the key in the x-api-key request header; the key can be obtained from the HasData console.

Client setup examples (original documentation text):

  • Claude Desktop / Claude.ai: go to Settings → Connectors → Add custom connector and enter the address; it will prompt you to log in with OAuth. If you prefer an API key, you can instead write it into claude_desktop_config.json.
  • Claude Code: supports OAuth (you can also run /mcp in the session and choose Authenticate), or use an API key.
  • Cursor: add it to ~/.cursor/mcp.json or .cursor/mcp.json.
  • Cline: add the hasdata server in the Cline MCP settings file.
  • Other MCP clients that support streamable HTTP can also use it.

Requests without OAuth or a valid key return 401 Unauthorized. If you want the client to expose fewer tools, you can use the apis query parameter to limit the APIs, for example https://mcp.hasdata.com/mcp?apis=amazon,google_maps。各站点的 the apis key is listed in the "One site per server" section. The remote URL for any standalone server is https://mcp.hasdata.com/mcp?apis=<key>, multiple keys can be combined with commas. Each standalone server is published simultaneously as both an npm and PyPI package, for clients that need a local stdio launcher, and is listed in the official MCP registry as com.hasdata/<name>.

Use Cases

  • Competitive intelligence: call amazon_search and google_serp_shopping simultaneously for the same query, letting the model compare prices, rankings, and review counts.
  • Lead generation: use google_maps_search to search by keyword and city, returning business names, addresses, phone numbers, and ratings, and stack yellowpages_search to expand coverage.
  • Market research: combine google_trends_search and google_serp_news in the same session to obtain both search volume signals and editorial context.
  • Real estate: zillow_listing supports filtering by bedrooms, bathrooms, price range, and listing type, allowing the model to pull filtered results directly and summarize them.
  • Recruiting: indeed_listing and glassdoor_listing both accept location and keywords; after running both, the model can deduplicate and sort by time.
  • Travel: google_travel_flights returns structured itineraries with prices, stops, and durations, and can be combined with airbnb_listing to complete trip planning in a single conversation.

Limitations and Notes

  • The server is remotely hosted and does not run locally: there is no local Node process, no version pinning, and no local cold start.
  • Each tool call consumes HasData credits in the same way as directly calling the REST API; for specific credits and concurrency rules, see the official Credits and Concurrency documentation.
  • Each tool is site-specific, not a general-purpose scraper.
  • Pricing, specific credit amounts, and plan information are not yet specified in the provided materials.

Contact: support@hasdata.com; for API documentation, see HasData API Docs.

Related tools

Browse category