Model Context Protocol
Connect your AI agent to every ScraperCity tool. Scrape leads, find and verify emails, look up mobiles, run skip traces, and query 4.6M+ B2B contacts, all in plain language.
Works with Claude Code, Cursor, Windsurf, Cline, GitHub Copilot, Gemini CLI, and any MCP client.
1. Add the config to your MCP client
{
"mcpServers": {
"scrapercity": {
"command": "npx",
"args": ["-y", "--package", "scrapercity", "scrapercity-mcp"],
"env": {
"SCRAPERCITY_API_KEY": "your_api_key_here"
}
}
}
}Or add it to Claude Code with one command
claude mcp add scrapercity --transport stdio -e SCRAPERCITY_API_KEY=your_api_key_here -- npx -y --package scrapercity scrapercity-mcp2. Drop the config in the right file for your client
| Client | Config file |
|---|---|
| Cursor | ~/.cursor/mcp.json |
| Windsurf | ~/.codeium/windsurf/mcp_config.json |
| Claude Code | ~/.claude/mcp.json or .mcp.json in the project root |
| Cline | the Cline MCP settings file in VS Code |
3. Fully quit and relaunch your editor. Grab your API key from app.scrapercity.com/dashboard/api-docs.
All available through natural language. Your agent picks the right tool from your prompt, no parameters or documentation required. Here is every tool, grouped.
scrape_apolloPull contacts from an Apollo search URL, unlimited per run.
scrape_mapsExport Google Maps business listings to CSV.
scrape_yelpYelp business listings with reviews and ratings.
scrape_store_leadsEcommerce stores by platform, category, and installed apps.
scrape_builtwithFind sites by the technology they run.
scrape_zillow_agentsReal estate agent contact data.
scrape_crexiCommercial real estate listings and brokers.
scrape_bizbuysellBusinesses-for-sale listings.
scrape_angiHome-service provider listings.
scrape_airbnbHost and property details from Airbnb.
scrape_youtube_emailCreator emails from YouTube channels.
scrape_website_finderFind a company website from its name.
find_emailsName and company to a verified business email.
validate_emailsBulk deliverability verification with quality scores.
find_mobilesLinkedIn URL or email to a verified mobile number.
find_peopleFind anyone by name, phone, email, or address.
scrape_property_lookupProperty owner, home value, and tax history.
search_criminal_recordsCriminal, warrant, and court records.
query_lead_databaseFilter millions of B2B contacts and export to CSV.
check_run_statusPoll a run until it finishes.
download_resultsDownload a finished run as a CSV.
check_walletRead your current wallet balance.
list_runsList your recent runs.
cancel_runCancel a running job.
Your agent reads the intent from your prompt and calls the right tool. Every example is a real query, with the tool it lands on on the right.
| You ask | It calls |
|---|---|
| “Find every plumber in Denver with a website and an email, then validate them” | scrape_maps + find_emails |
| “Pull all contacts from this Apollo search and validate the emails” | scrape_apollo + validate_emails |
| “Give me Shopify apparel stores using Klaviyo that have an email” | scrape_store_leads |
| “Find the VP of Sales at these 40 companies, verified emails only” | query_lead_database + find_emails |
| “Get mobile numbers for these 200 LinkedIn profiles” | find_mobiles |
| “Find the owner and current phone for these property addresses” | scrape_property_lookup + find_people |
| “How many SaaS CTOs in the US are in the database?” | query_lead_database |
| “Enrich this CSV of contacts with emails and mobiles” | find_emails + find_mobiles |
Every tool follows one pattern: start a run, poll its status, then download the CSV. Most return in minutes; the agent handles the polling for you.