The Mine Works MCP Server: Add 29 Data Tools to Claude Desktop in 2 Minutes

The Model Context Protocol lets AI clients like Claude Desktop and Cursor call external tools directly from the chat interface. Instead of running a scraper separately and pasting results in, you ask Claude to find LinkedIn employees at a company and it returns them in the same conversation.

The Mine Works MCP server puts 29 of our data tools behind a single endpoint. One connection, your Apify token, and all 29 tools are available wherever you work.

What is MCP

MCP (Model Context Protocol) is an open standard published by Anthropic that lets AI clients connect to external servers and call their tools. It works like a plugin system: you add a server URL to your AI client, and the AI can call that server’s tools as part of its reasoning process.

Claude Desktop, Cursor, Windsurf, and most modern AI coding tools support MCP out of the box. The tools appear natively in the interface: no copy-pasting, no separate terminal, no context-switching.

The Mine Works MCP: what’s included

29 tools covering B2B prospecting and lead generation, recruiting and job market intelligence, company and financial verification, real estate, developer intelligence, and academic research:

Tool	What it does
search_linkedin_employees	Find people at any company using Google indexing
search_linkedin_jobs	LinkedIn job listings by keyword and location
search_linkedin_posts	Search posts by keyword via Google indexing
find_b2b_leads	Business emails and LinkedIn profiles for target companies
scrape_reddit	Posts and comment trees by subreddit or keyword
scrape_ats_jobs	Greenhouse, Lever, Workday, and Ashby job boards
search_google_trends	Interest over time and geographic breakdown
search_sec_filings	Full-text search across SEC EDGAR filings
search_indiamart_suppliers	B2B suppliers and manufacturers in India
search_naukri_jobs	Job listings in India with salary data
search_pubmed	36M+ biomedical articles with abstracts
search_arxiv	AI, CS, and science preprints with PDF links
search_nih_grants	NIH grant awards, PIs, and funding amounts
scrape_trustpilot	Customer reviews with ratings and dates
get_linkedin_profiles	Experience, education and skills by profile URL
get_linkedin_company	Employee count, industry, website, followers
search_linkedin_employees	Employee lists and candidate sourcing by company and role
search_google_maps	Business listings: address, phone, website, rating, hours
find_website_contacts	Emails, phones and socials from any list of domains
verify_emails	MX, SMTP, disposable and role-based detection in bulk
search_crunchbase	Funding totals, last round, investors, firmographics
analyze_github_repos	Stars, languages, README, releases, commit recency
lookup_lei	Official GLEIF Legal Entity Identifier records
resolve_company_identity	Resolve a company to its LEI and SEC CIK in one call
search_zillow	For-sale and rental listings by city, ZIP or neighborhood
get_zillow_property_details	Price history, tax history, schools, Zestimate
search_redfin	For-sale and sold listings with agent, broker and MLS ID
search_realtor	Listings with agent and brokerage on every record
get_run_results	Collects results from longer-running tools

No LinkedIn login. No cookies. No API keys beyond your Apify token.

How billing works

Every tool call triggers an Apify actor run on your account. You pay Apify directly at pay-per-result rates, the same pricing as using the actors directly from the Apify console.

There is no subscription or credit system on our end. The MCP server is a pass-through to your Apify account. Monitor usage at Apify Billing.

Every actor charges from the first result returned, and most charge nothing when a run comes back empty.

Connect to Claude Desktop

Open ~/.claude.json (or go to Settings › Developer › Edit Config) and add:

{
  "mcpServers": {
    "themineworks": {
      "url": "https://the-mine-works-mcp.hatchable.site/api/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_APIFY_TOKEN"
      }
    }
  }
}

Replace YOUR_APIFY_TOKEN with your token from Apify Account › Integrations. Restart Claude Desktop. All 29 tools appear immediately.

Connect to Cursor

In Cursor settings, go to Features › MCP and add a new server:

{
  "themineworks": {
    "url": "https://the-mine-works-mcp.hatchable.site/api/mcp",
    "headers": {
      "Authorization": "Bearer YOUR_APIFY_TOKEN"
    }
  }
}

Example prompts

Once connected, you can use natural language:

LinkedIn prospecting:

“Find the Head of Marketing and VP of Sales at Notion, Figma, and Linear. Return their LinkedIn profiles.”

Competitive job intelligence:

“Search LinkedIn for machine learning engineer jobs in Bangalore posted in the last week. Which companies are hiring most aggressively?”

Research synthesis:

“Search PubMed for recent clinical trials on GLP-1 receptor agonists and cardiovascular outcomes. Summarize the key findings from the abstracts.”

SEC filing analysis:

“Search SEC filings for mentions of ‘artificial intelligence risk’ in 10-K filings from the past year. Which companies are disclosing the most AI-related risks?”

B2B lead research:

“Find the CEO, CTO, and Head of Product at stripe.com, linear.app, and vercel.com. Include their LinkedIn profiles and guessed business emails.”

The transport

The server uses the MCP Streamable HTTP transport (protocol version 2024-11-05). It works with any MCP client that supports URL-based servers with custom headers. If your client supports command + args style stdio servers, you can also run it as a local proxy. Contact us for the CLI wrapper.

See all actors individually

Each of the 14 tools is also available as a standalone Apify actor at apify.com/themineworks. Use them directly if you prefer the Apify console, the Python SDK, or the JavaScript client.

The MCP server is a convenience layer for AI-first workflows. The underlying actors are the same.

---

MCP endpoint: https://the-mine-works-mcp.hatchable.site/api/mcp

Full setup guide: themineworks.com/mcp