MCPSERV.CLUB
foursquare

Foursquare MCP Server

MCP Server

Enable AI agents with real‑time, category‑rich local place search

Stale(55)
8stars
1views
Updated Sep 17, 2025

About

This MCP server connects an LLM to the Foursquare Places API, providing accurate geolocation, searchable categories, photos, reviews, and real‑time popularity data. It empowers AI agents to offer situationally aware, personalized place recommendations.

Capabilities

Resources
Access data sources
Tools
Execute functions
Prompts
Pre-built templates
Sampling
AI model interactions

Overview

The Foursquare MCP server bridges the gap between conversational AI agents and real‑world place data. By exposing the Foursquare Places API through Model Context Protocol, it lets an LLM translate a user’s raw GPS coordinates or natural‑language queries into actionable insights about nearby venues. This solves the long‑standing problem of turning geographic coordinates into meaningful, context‑rich information that an AI can reason about and present to a user.

At its core, the server offers high‑accuracy geotagging through Foursquare’s Place Snap technology and a rich search layer that filters by category, operating hours, popularity, photos, reviews, and ratings. Developers can build agents that ask for “the best coffee shop near me open now” or “find a park with Wi‑Fi within 5 km,” and the server returns structured results that include not only names and addresses but also multimedia, sentiment scores, and real‑time footfall data. This capability turns a simple location query into a nuanced recommendation engine powered by AI.

Key features of the server include:

  • Direct MCP integration – functions are exposed in a format that Claude (and other LLMs) can call without custom adapters.
  • Rich metadata retrieval – every place result comes with photos, reviews, ratings, and current popularity metrics.
  • Advanced filtering – developers can specify categories, amenities, hours, and other criteria to narrow results.
  • Real‑time data – the API provides up‑to‑date footfall and popularity scores, enabling agents to suggest places that are currently less crowded.
  • Global coverage – the underlying database spans 100 million venues across more than 1,500 categories worldwide.

Use cases abound in both consumer and enterprise domains. A travel assistant can generate day‑plans that adapt to a user’s current location, recommending restaurants, museums, and parks with real‑time crowd information. A delivery service can route drivers to the nearest pickup point that is open and not overloaded. In retail, an AI concierge can guide shoppers to the nearest store with desired products or promotions.

Integrating this server into an AI workflow is straightforward: after obtaining a Foursquare Service API key, the MCP server exposes functions that an LLM can invoke during a conversation. The LLM can decide when to call the server based on user intent, and the returned structured data feeds directly into response generation. This tight coupling eliminates latency between query and answer, delivering a seamless, situationally aware user experience.

What sets this MCP server apart is its focus on situational awareness combined with a developer‑friendly interface. By abstracting the complexities of the Foursquare API behind MCP, it empowers developers to build highly personalized, context‑aware agents without deep knowledge of REST or OAuth flows. The result is a powerful tool that turns raw geographic data into actionable, AI‑driven insights for the next generation of location‑aware applications.