Back to blog

Engineering · Jun 11, 2026

MCP Spotlight: HasData — 40+ Structured Scraping Tools for Google, Amazon, Zillow, Airbnb, Indeed & More

HasData exposes 40+ site-specific MCP tools — Google Search, Maps, Trends, Flights, Amazon, Zillow, Airbnb, Indeed, Glassdoor, Yelp — returning parsed JSON instead of raw HTML. Hosted Streamable HTTP transport, one API key, no local process.

MCP ServerHasDataWeb ScrapingStructured DataMarket ResearchAI Agents

MCP Spotlight: HasData — 40+ Structured Scraping Tools for Google, Amazon, Zillow, Airbnb, Indeed & More

Server: hasdata-mcp by HasData Tools: 40+ · License: MIT · Transport: Streamable HTTP · Mode: Hosted (no local process) MCP Tracker: glama.ai/mcp/servers/HasData/hasdata-mcp Docs: docs.hasdata.com/mcp-server

Most scraping MCP servers hand you back raw HTML and call it a day. On JavaScript-heavy sites, half the time the HTML is empty because the page renders client-side. Your agent then has to wade through a soup of <div> to find the actual product price. HasData takes the opposite approach: every tool returns site-specific, parsed JSON that the model can reason about directly.

40+ tools, hosted MCP endpoint, one API key, and a transport designed for cloud agents.

Why Site-Specific Parsing Matters

Generic scrapers return markup. The model then has to:

  1. Figure out which <div> contains the price
  2. Strip currency symbols and whitespace
  3. Parse the numeric value
  4. Handle locales (€, $, £, comma vs. dot decimals)
  5. Discard ads, navigation, and recommendation widgets
  6. Extract the structured field the user actually asked for

Most "agent scrapes the web" demos fail on step 1. HasData's per-site tools skip all of that:

❌ Generic scraper: <div class="a-price"><span class="a-offscreen">$1,299.00</span>...
✅ HasData amazon_product: { "price": 1299.00, "currency": "USD", "in_stock": true, "rating": 4.7, "review_count": 8234, ... }

The model gets a JSON object with price, in_stock, rating, and review_count as top-level fields. No parsing, no guessing, no browser automation.

The 40+ Tools by Category

Web

ToolDescription
web_scraping_web_scrapingGeneric URL scraper with optional parameters (for sites without a dedicated tool)

Google (16 tools)

ToolDescription
google_serp_serpGoogle Search results — full version
google_serp_serp_lightGoogle Search results — lightweight version
google_serp_newsGoogle News results
google_serp_shoppingGoogle Shopping results
google_serp_images_imagesGoogle Image Search results
google_serp_eventsGoogle Events results
google_serp_productGoogle product details
google_serp_immersive_productGoogle immersive product details
google_serp_ai_overviewGoogle AI Overview results
google_serp_ai_modeGoogle AI Mode results
google_maps_searchGoogle Maps search by keyword and location
google_maps_placePlace details by placeId (hours, rating, coordinates, etc.)
google_maps_reviewsPlace reviews with full text
google_maps_photosPlace photos
google_maps_contributor_reviewsReviews by contributor ID
google_trends_searchGoogle Trends data
google_travel_flightsGoogle Flights results — itineraries, prices, stops, duration
bing_serpBing Search results

E-commerce

ToolDescription
amazon_searchAmazon search results
amazon_productAmazon product details by ASIN (price, variants, sellers)
amazon_reviewsAmazon product reviews
shopify_productsShopify store products
shopify_collectionsShopify store collections

Real Estate

ToolDescription
zillow_listingZillow listing search with filters (beds, baths, price)
zillow_propertyZillow property details
redfin_listingRedfin listing search
redfin_propertyRedfin property details

Jobs

ToolDescription
indeed_listingIndeed job listings
indeed_jobIndeed job details
glassdoor_listingGlassdoor job listings
glassdoor_jobGlassdoor job details

Travel & Local

ToolDescription
airbnb_listingAirbnb listings by location and dates
airbnb_propertyAirbnb listing details
yelp_searchYelp search results
yelp_placeYelp place details
yellowpages_searchYellowPages search results
yellowpages_placeYellowPages place details

Social

ToolDescription
instagram_profileInstagram public profile details

The Hosted MCP Architecture

HasData runs as a remote MCP server — no local process, no npx invocation, no version pinning. The transport is Streamable HTTP with custom headers for the API key:

{
  "mcpServers": {
    "hasdata": {
      "type": "http",
      "url": "https://mcp.hasdata.com/api/mcp",
      "headers": {
        "x-api-key": "${credentials.HASDATA_API_KEY}"
      }
    }
  }
}

Compatible with Claude Desktop, Claude Code, Claude.ai (web), Cursor, Windsurf, VS Code (GitHub Copilot), Gemini CLI, and any custom agent using the OpenAI or LangChain MCP transport.

The hosted model means:

  • No cold starts on your machine
  • No version pinning — HasData maintains compatibility
  • No rate-limit management — HasData handles proxy rotation, JavaScript rendering, and anti-bot bypass
  • Same billing as the REST API — if you're an existing HasData customer, your existing balance and rate limits apply

The x-api-key header is sent on every call. There's no session or token exchange to manage.

The "Run It All In One Prompt" Promise

The product page makes a specific promise — and the tool design backs it up:

"Find the cheapest flights from NYC to London next month and compare prices." "Monitor iPhone 16 prices on Amazon across the last 7 days." "Pull all 5-star reviews for this product ASIN and summarize the common themes."

None of these require scraping code, proxy management, or parsing logic. The agent chains google_travel_flights, amazon_product, and amazon_reviews directly. The structured output flows into the model's reasoning, not into a regex pipeline.

Facio Integration

{
  "mcpServers": {
    "hasdata": {
      "type": "http",
      "url": "https://mcp.hasdata.com/api/mcp",
      "headers": {
        "x-api-key": "${credentials.HASDATA_API_KEY}"
      }
    }
  }
}

Facio's audit trail captures every data extraction call: which site was queried, what parameters, what was returned. For competitive intelligence and market research workflows, this creates a traceable record of how a market analysis was assembled — which Amazon ASINs were checked, which Glassdoor jobs were retrieved, which Google Maps places were inspected.

For HITL workflows, the read-side (data extraction) is fully agent-driven. The write-side — actually placing an order, sending a message, applying for a job — stays with the human. The agent surfaces the data, the human takes the action. Facio captures both phases.

Quickstart

# 1. Get an API key from the HasData dashboard
#    https://app.hasdata.com/

# 2. Add to your MCP client config
{
  "mcpServers": {
    "hasdata": {
      "type": "http",
      "url": "https://mcp.hasdata.com/api/mcp",
      "headers": {
        "x-api-key": "<your-api-key>"
      }
    }
  }
}

# Or with Claude Code:
# claude mcp add hasdata -t http https://mcp.hasdata.com/api/mcp \
#   --header "x-api-key: <your-api-key>"

# 3. First prompts
# "Find the cheapest flights from NYC to London next month and compare prices"
# "Monitor iPhone 16 prices on Amazon across the last 7 days"
# "Pull all 5-star reviews for ASIN B0CHX1W1XY and summarize the common themes"
# "Get me all software engineer job postings in Berlin from Indeed and Glassdoor"
# "Find the top-rated Italian restaurants in Chicago using Google Maps"
# "What are people searching for related to 'AI agents' on Google Trends this week?"

Use Cases

Competitive intelligence: Point amazon_search and google_serp_shopping at the same query. The model compares pricing, ranking, and review counts across both — in one prompt, no spreadsheet.

Lead generation: google_maps_search by keyword and city returns business names, addresses, phone numbers, and ratings. Stack with yellowpages_search for broader coverage. Your agent builds a lead list with structured fields ready for CRM import.

Market research: google_trends_search + google_serp_news in the same session gives you both the search volume signal and the editorial context. The agent reasons across both to identify trending topics.

Real estate analysis: zillow_listing accepts filters for beds, baths, price range, and listing type. The agent pulls a filtered list and summarizes it — no manual Zillow browsing.

Recruiting: indeed_listing and glassdoor_listing both take location and keyword. Run both and the model deduplicates and ranks by recency.

Travel planning: google_travel_flights returns structured itineraries with prices, stops, and duration. Combine with airbnb_listing for a full trip plan in one conversation.

Price monitoring: Track the same Amazon ASIN over time. The agent calls amazon_product repeatedly and surfaces price changes, stock status, and review count drift.

Bottom Line

HasData is what "agent scrapes the web" should look like: 40+ site-specific tools that return parsed JSON, not raw HTML. Hosted MCP endpoint, no local process, one API key, and a transport designed for cloud agents.

For competitive intelligence, lead generation, market research, recruiting, travel planning, and any workflow where your agent needs structured data from the public web — this is the missing layer between "ask the agent" and "build a custom scraper."

At https://mcp.hasdata.com/api/mcp, your agent can answer data questions about Google, Amazon, Zillow, Airbnb, Indeed, Glassdoor, Yelp, and more in a single prompt.


MCP Spotlight is a series covering servers that give AI agents real capabilities. Every server is evaluated for tool quality, data quality, and integration fit with Facio's HITL-first agent runtime.