The Mine Works

Crawl any website into clean, pre-chunked Markdown with per-chunk token counts for RAG pipelines, vector DBs (Pinecone, Qdrant) and LLM context. MCP native for Claude & ChatGPT. SPA support via Playwright. Pay only for pages that crawl. A Firecrawl alternative.

Get Semrush Authority Score, backlinks, referring domains, and organic traffic estimates for any domain via Semrush's free Website Authority Checker. No Semrush account, login, or paid API needed. Use it as an MCP server in Claude, ChatGPT & AI agents.

Search and pull datasets from any Socrata government portal (CDC, HHS, CMS, NY, NYC, Texas, 100s more). SoQL filtered queries, structured JSON, no API key, no anti bot. Works in Claude, ChatGPT and any MCP agent for instant public data access.

Scrape Bilibili (China's dominant video platform) by keyword, video URL or creator ID: views, danmaku, likes, coins, favorites, shares, duration and creator stats. Public API, no login, no browser needed.

Scrape every US federal contract, grant, loan & award from USAspending.gov as clean JSON. Filter by keyword, agency, NAICS, recipient, amount & state. GovCon & B2G intelligence. No API key. Works in Claude, ChatGPT & any MCP agent.

Resolve a company name to its official website domain, filtering out Wikipedia, LinkedIn, social media and directory sites. No login, no API key.

Scrape Similarweb's free public website overview: estimated monthly visits, engagement, traffic-source split, top countries & similar sites. No login, no paid Similarweb seat. Use it as an MCP server in Claude, ChatGPT & AI agents.

Look up a US importer's public sourcing profile: total shipments, date range covered, and the supplier list with country, location and shipment volume per supplier. Sourced from public US Customs bill of lading manifest data. No login. Works in Claude, ChatGPT and any MCP agent.

Scrape Google News by keyword or topic: headline, article URL, source, publish date & snippet. No API key, no login, up to 1,000 results per run. Works in Claude, ChatGPT & any MCP AI agent for live news monitoring.

Live job-market data for recruiting agents: ATS postings (Greenhouse/Lever/Workday/Ashby), LinkedIn jobs, Naukri, candidate sourcing, and employer snapshots. No result, no charge.

Find public Telegram channels matching a keyword: handle, title and canonical t.me URL. Optional monitor mode bills only for new channels since the last run. No login, no bot token.

Scrape FDA 510(k) medical device clearances by company, device, or product code. Returns applicant, decision date, decision type & panel for medtech competitive intel and regulatory affairs. No API key. Use it as an MCP server in Claude, ChatGPT & AI agents.

Scrape Google Maps business listings by search query: name, category, address, phone, website, rating, review count, hours, coordinates. No login, no email guessing. Pay only per place delivered. Works in Claude, ChatGPT & any MCP agent.

Export any Instagram account's followers or following list to structured JSON/CSV: username, name, ID, verified & private flags, profile link. Instagram gates these lists behind login, so this actor uses YOUR OWN session cookie (never ours) to paginate safely. MCP-ready.

Scrape Trustpilot business reviews by domain: reviewer, star rating, title, text, date, company reply, trust score & total count. No login, beats the AWS WAF via residential proxies. Works in Claude, ChatGPT & any MCP agent.

Scrape Weibo (China's dominant Twitter-equivalent) by user ID: follower count, verification, bio, and full post timeline with reposts, comments and likes. China residential proxy, no login, no browser.

Scrape Carsales.com.au vehicle listings by make and model: price, year, odometer, listing URL and photos. Australia residential proxy, no login, pay per vehicle.

Scrape Etsy search listings: title, price, currency, shop name and image. Handmade and vintage marketplace research. No login, no API key.

Scrape Google Trends: interest over time, related & rising queries, related topics, geo breakdowns. Degraded since 2026-07-30 by Google-side rate-limiting, see README — zero charge on failure either way. Multi keyword compare, all timeframes.

Discover public LinkedIn newsletters matching a keyword: title, URL and subscriber count where available. Optional monitor mode bills only for new newsletters. No login, no API key.

Scrape Norway's official BRREG business register: company status, NACE industry, address, employees, and full officer roster (CEO, board, signatories) by name or org number. 1.16M entities, no API key. Works in Claude, ChatGPT & any MCP agent.

Scrape US healthcare providers from the official CMS NPI Registry by specialty, name, org, city or state. Get NPI, credentials, specialty, license, address, phone as clean JSON. No API key. Works in Claude, ChatGPT and any MCP agent.

Search the full text of every SEC EDGAR filing since 2001 across all companies. Find any phrase or risk factor in 10-K, 10-Q, 8-K, S-1 filings. Filter by form and date. Clean JSON, no API key. Works in Claude, ChatGPT and any MCP agent.

Scrape the full product catalogue of any Shopify store — title, vendor, type, tags, variants, price range, SKUs, stock and images. Uses each store's public products feed: no login, no API key, no browser. Competitor and dropshipping research. Use it as an MCP server in Claude, ChatGPT & AI agents.