OpenStreetMap Scraper - Places, POIs & Geocoding

Search OpenStreetMap (via the official Nominatim API) and get clean rows for any named place or address: coordinates, full structured address, category/type, OSM ids and bounding box. Geocode addresses in bulk or look up landmarks, cities and points of interest. No API key, no browser.

What you get per place

Field	Meaning
displayName / name	Full display name and short name
lat / lon / boundingBox	Coordinates and bounding box
category / placeType / addressType	OSM class, type and address type
houseNumber / road / city / county / state / postcode / country / countryCode	Structured address parts
osmType / osmId / placeId / placeUrl	OSM identifiers and the map page

Input

{ "searchQueries": ["Eiffel Tower", "1600 Pennsylvania Ave"], "countryCodes": "us", "maxItemsPerQuery": 10 }

• Search terms - named places or addresses, one per line. Optionally restrict to countryCodes (e.g. us,ca).
• Import from a file - paste a list, or link a public .txt/.csv, a Google Sheet/Drive link, or an Apify key-value-store record.
• Output fields - trim every record to exactly the columns you need.

Leave everything empty and the run returns a small free sample so you can see the shape first.

Reliability

Reads the official Nominatim API with a descriptive User-Agent and a built-in delay that respects Nominatim's usage policy (about one request per second). A search that matches nothing, or an import that cannot be read, bills $0 and says why.

Honest limits

• Matches named places and addresses, not "category in a city" phrases. Use Eiffel Tower, Berlin, or a street address - a query like "coffee shops in Berlin" is not a Nominatim geocoding query and returns nothing.
• Street-level fields (houseNumber, road, postcode) fill only for address queries - city and landmark searches return the place without a house number by design.
• Throttled to ~1 request/second per Nominatim policy, so large lists take proportionally longer.

Pricing

$0.001 per place on the FREE tier (tiered pricing lowers it with volume). Only places actually saved are billed; empty searches and unreadable imports cost nothing.

Output views

• Places - name, full name, category, coordinates and the OSM URL.

Use with AI assistants (MCP)

Available through the Apify MCP server - an agent can geocode an address, resolve a landmark to coordinates, or enrich a list of place names with lat/lon and structured addresses in one call.

Agent-ready: autonomous payments (x402 & Skyfire)

This actor is agent-ready - AI agents can discover it, run it, and pay for it autonomously, with no Apify account and no human in the loop. It uses pay-per-event pricing and limited permissions, so it qualifies for Apify's agentic-payment standards:

• x402 - an open, HTTP-native payment protocol. Agents pay per run in USDC on the Base network directly through the Apify MCP server - no account, no API key.
• Skyfire - agent-to-service payments for fully autonomous AI-agent workflows.

Building an AI agent, MCP tool, or autonomous data pipeline? This scraper is ready to plug in and pay as it goes.