Search skills/

Google Maps Scraper Claude skill

Agent skill that runs the open-source gosom Google Maps scraper locally in Docker to collect business leads, reviews, emails and contact details.

gosom/google-maps-scraper 1 skill Browser automation & scrapingData & analytics
Our verdict Recommended with caveats

A practical, well-guarded wrapper around an established scraper, but it scrapes a Google product and the README contains sponsor and affiliate-style promotions.

3/5 Depth expertise shipped ! Security Broad access 8d Maintenance Updated 8 days ago 7.0K Adoption stars OK License MIT, commercial OK

What Google Maps Scraper does

This skill wraps gosom's Google Maps scraper, a Go tool that extracts place data such as name, address, phone, website, rating, review counts, coordinates, reviews and optionally emails. The skill turns a plain request like find dentists in Berlin into a validated local crawl: it checks the request, guides nontechnical users through setup, starts the scraper in Docker, monitors the job with a status script and helps you work with the results. It also asks you to choose a proxy path (use your own, see sponsor options, or continue without) and its rules forbid asking for, printing or logging proxy credentials. The SKILL.md is about 7 KB with seven reference files and a handful of Node scripts, and it is allowed to run node, bash and docker commands. The underlying project offers a CLI, web UI, REST API and an optional self-hosted multi-user edition, plus a LeadsDB export option that takes an API key. It requires Docker and Node.js on macOS, Linux or Windows through WSL. Scraping Google Maps may conflict with Google's terms, and collecting emails can raise privacy-law issues.

Good for

  • Local lead lists and business directories
  • Market research on competitors in a city
  • Nontechnical users who want an agent to run the scraper

Not for

  • Use cases that must stay within Google's terms of service
  • Environments without Docker

What's inside depth 3/5

A 7 KB skill with seven reference files and a few Node scripts, wrapping a mature separate scraper; no skill-level tests or evals.

Skills
1
Reference docs
7
Scripts
8
Tests / evals
3
Templates
0
Avg SKILL.md
7.1 KB

Counted from the repository file tree. How we score depth.

Dependencies and costs

  • Docker
  • Node.js
  • Optional proxy service for larger crawls
  • Optional LeadsDB API key for export

Security review

Broad access

This skill asks for more access than most, or sends data to another service. That is usually a design choice rather than bad intent, but you should know before you install it.

  • Pre-approves broad shell commandsFrontmatter allowed-tools: Bash(node:*) Bash(bash:*) Bash(docker:*) Read Write pre-approves any bash, node and docker command, and the body tells the agent not to ask permission before routine Docker runs.skills/google-maps-scraper/SKILL.md:15
  • Loads instructions from a server at run timeAt the start of every workflow the agent runs npx --yes skills update google-maps-scraper --yes without asking, then re-reads SKILL.md so the updated instructions apply immediately.skills/google-maps-scraper/scripts/ensure-latest.sh:12

A Google Maps scraping skill that runs a Docker image to collect business data, with helper scripts for proxy setup and monitoring. The agent is pre-approved for any bash, node and docker command and, at the start of each workflow, silently updates the skill from the network and reloads its instructions. The scraper sends anonymous usage telemetry unless you set DISABLE_TELEMETRY=1, and the repo also contains an optional self-hosted SaaS provisioner (curl-pipe-sh install, SSH and cloud-token handling) that the skill itself does not use. Scraping Google Maps may breach Google's terms of service.

Worth knowing

  • Runs remote code at an unpinned versionRuns the unpinned npm package skills via npx --yes, and run-local.sh pulls the unpinned Docker image gosom/google-maps-scraper (latest) before each crawl.skills/google-maps-scraper/scripts/ensure-latest.sh:12
  • Sends usage data by default (can be turned off)The scraper sends anonymous usage events (job count, duration, error text, OS info, an ID hashed from your public IP) to PostHog (eu.i.posthog.com) unless DISABLE_TELEMETRY=1; run-local.sh does not set that variable.runner/runner.go:252
  • Installs with curl | shREADME tells the user to run curl -fsSL .../PROVISION | sh for the optional self-hosted SaaS setup; PROVISION runs a :latest Docker image with the Docker socket and ~/.ssh mounted.README.md:339
  • Makes network requestsThe scraper loads Google Maps pages with Playwright, looks up the public IP via ipify/ifconfig.me-type services for the telemetry ID, and optionally exports to LeadsDB, S3 or a Postgres server.tlmt/tlmt.go:90
  • Needs an API key for its own serviceOptional LEADSDB_API_KEY / -leadsdb-api-key is used only for the LeadsDB export client; proxy credentials go to a local file read by the scraper for the proxy.leadsdb/leadsdb.go:17
  • Installs packages or toolsPulls a Docker image, creates a persistent Docker volume for the Playwright browser cache, and installs a Chromium browser through Playwright; the SaaS provisioner installs Docker and Caddy on remote servers.skills/google-maps-scraper/scripts/run-local.sh:148
  • Changes files outside your projectWrites proxy URLs entered at a masked prompt to ~/.config/google-maps-scraper/proxies.txt (mode 600, plaintext); the SaaS provisioner uses ~/.gmapssaas and ~/.ssh/known_hosts-style files.skills/google-maps-scraper/scripts/configure-proxy.sh:8
  • Relies on a CLI or package we could not reviewRelies on Docker and Node.js/npx, and on a prebuilt Docker image whose contents are not the files we reviewed.skills/google-maps-scraper/scripts/run-local.sh:123

What we checked

The repository (260 files) on Oct 29, 2026. Not a full review: we read 13 of 146 code files; it relies on code outside the repository: Docker image gosom/google-maps-scraper (pulled at its latest tag on each run); Docker image ghcr.io/gosom/google-maps-scraper-saas:latest used by PROVISION; npm package `skills` run by `npx --yes skills update`.

NVIDIA SkillSpector static scan: 183 pattern hits (Memory Poisoning 55, Privilege Escalation 28, MCP Rug Pull 28, Tool Misuse 27, Data Exfiltration 12, Rogue Agent 9). Hits are Dockerfile/Makefile cleanup commands, a vendored minified xterm.js and pico.css (normal libraries), the SaaS provisioner's SSH-key, token and docker.sock handling, test fixtures with long encoded Google Maps payloads, and documentation. The telemetry, self-update and pre-approved shell items above are real behaviors, not scanner noise.

A review, not a certification. Skills change after we check them — read the files before installing.

License

MIT — commercial use allowed. Scraped data may carry its own legal restrictions.

Can it run unattended?

Yes. Crawls run as background Docker jobs that the skill monitors with a status script and can resume, so long jobs can run unattended.

Alternatives to Google Maps Scraper

Apify Agent Skills

apify/agent-skills

Apify's official skills for scraping with its Actor marketplace and for building, converting and integrating Actors with the apify CLI.

With caveats Check first Depth 3/5

Firecrawl CLI

firecrawl/cli

Firecrawl's command-line client and 13 agent skills for web search, scraping, crawling, interacting with pages and parsing files through the paid Firecrawl API.

With caveats Check first Depth 2/5

Browser Use

browser-use/browser-use

Seven skills around the Browser Use browser agent: drive local Chrome over CDP, use the paid cloud API, run QA scoring, or pay by crypto wallet.

With caveats Broad access Depth 3/5

FAQ

What is the Google Maps scraper skill?

It is an agent skill from the gosom/google-maps-scraper project. You describe the businesses you want and the agent runs the scraper locally in Docker and helps you read the results.

How do I install it?

Run npx skills add gosom/google-maps-scraper. You need Docker and Node.js, and on Windows the README says to use WSL.

Is scraping Google Maps allowed?

The tool itself is open source, but Google's terms restrict automated scraping, and collecting personal contact data can fall under privacy laws. Check your own obligations before using it.

Does it cost anything?

The scraper is free and MIT licensed. Proxies, the HasData service and LeadsDB mentioned in the README are optional third-party services with their own pricing.

Sources: GitHub repository (stars, license, file tree, last push Oct 21, 2026). Review based on the README and SKILL.md files as published.