You need a list of plumbers in Manchester, dentists in Miami, or roofers in Lyon — with phones, emails, and websites. Doing it by hand takes days. Writing a scraper takes hours... until Google rate-limits you into CAPTCHA hell.

Here's the 2026 way: run a maintained Apify actor that handles proxies, pagination, and anti-blocking — then let your AI agent do the plumbing.

The stack

  • Apify — actor marketplace with ready-made Google Maps scrapers. Free tier includes monthly credits — enough to test this end-to-end.
  • Your AI agent (Claude Code, Cursor, Codex...) — runs the pipeline via a skill that knows the actor inputs.

No proxies to configure. No headless browser tuning.

Step 1 — Get your API token

  1. Create an Apify account (free credits included)
  2. Console → Settings → Integrations → copy your API token
  3. export APIFY_TOKEN=your_token

Step 2 — Run the lead-extraction actor

The lukaskrivka/google-maps-with-contact-details actor doesn't just grab businesses — it visits each website and extracts emails and social profiles automatically:

curl -X POST "https://api.apify.com/v2/acts/lukaskrivka~google-maps-with-contact-details/run-sync-get-dataset-items?token=$APIFY_TOKEN" \ -H "Content-Type: application/json" \ -d '{ "searchStringsArray": ["dentist"], "locationQuery": "Miami, FL", "maxCrawledPlacesPerSearch": 100, "language": "en" }'

You get back a JSON array with title, phone, website, emails, address, totalScore, reviewsCount, categoryName.

Step 3 — Clean lead CSV with Python 

from apify_client import ApifyClient import csv client = ApifyClient("APIFY_TOKEN") run = client.actor("lukaskrivka/google-maps-with-contact-details").call(run_input={ "searchStringsArray": ["plumber"], "locationQuery": "Austin, TX", "maxCrawledPlacesPerSearch": 200, }) with open("prospects.csv", "w", newline="") as f: w = csv.writer(f) w.writerow(["business", "phone", "email", "website", "rating", "reviews"]) for p in client.dataset(run["defaultDatasetId"]).iterate_items(): if p.get("phone") or p.get("emails"): w.writerow([ p["title"], p.get("phone", ""), (p.get("emails") or [""])[0], p.get("website", ""), p.get("totalScore", ""), p.get("reviewsCount", ""), ])

Dedupe by website or phone, filter for totalScore >= 3.5 — and you have a clean outreach list.

Bonus: let your agent run it for you

I packaged this (plus 7 other scraping patterns — jobs, Amazon, social, SERP) as Agent Skills:

npx skills add g4dr/apify-skills

After installing, tell Claude Code "get me all roofers in Birmingham" — it picks the right actor, builds the run input, and hands you the data. Full repo: g4dr/apify-skills.

For the bigger picture (proxies, headless browsers, niche actors), I maintain an awesome-web-scraping list.

Compliance note

Google Maps data is public business info — but respect GDPR and outreach laws (CAN-SPAM, opt-outs). Scrape public data, don't republish personal info, and honor platform ToS.

Questions? Drop them in the comments — happy to share run inputs for specific niches.