Overview
An MCP server that exposes access to the Foursquare Places API so AI agents can perform local search for places. The server leverages Foursquare's Places API, GeoTagging API, and Search & Data APIs to provide rich location data, including category filters, hours, photos, reviews, and real-time popularity. By adhering to the Model Context Protocol, it exposes functions that an LLM can call directly to query places, retrieve metadata, and refine search results based on user context. The project emphasizes local deployment: MCP is currently supported for local use, requiring a Foursquare Service API Key and the Claude Desktop App for development. To set up the server, follow the Python-based setup described in fsq-server-python/README.md, which walks through configuring a Python MCP server using uv. This setup enables AI agents to become situationally aware and provide personalized recommendations based on user surroundings.
Features
Access to Foursquare Places API via MCP
Exposes functions and endpoints to query places and retrieve metadata from the Foursquare Places API through MCP.
LLM-callable MCP functions
Exposes functions that an LLM can understand and call directly to access Foursquare data.
Local deployment support
MCP is currently focused on local use, requiring Claude Desktop for development.
Python-based setup
Guides developers to set up a Python MCP server using uv, as described in fsq-server-python/README.md.
Foursquare API key authentication
Requires a Foursquare Service API Key to authenticate and access API endpoints.
Rich place data access
Provides access to place data with metadata such as categories, hours, photos, reviews, and real-time popularity via Foursquare APIs.
Who Is This For?
- AI developers:Enable AI agents to perform localized place searches using Foursquare data via MCP.
- Local business apps:Enable local business apps to present Foursquare-based place recommendations to users.
