Foursquare MCP Server
Exposes Foursquare/Swarm check-in data and user activity through 13 tools, enabling user management, check-in history, venue details, mayorships, lists, and cross-user search.
README
Foursquare MCP Server
An MCP server that exposes Foursquare/Swarm checkin data and user activity through 13 tools. Built with FastMCP, it supports stdio and HTTP transports.
Features
- User management — register multiple Foursquare accounts by OAuth token
- Checkins — fetch checkin history with date filtering and pagination, or get full detail for a single checkin
- Venue details — look up any venue by ID
- Venue history — unique venues visited with visit counts and timestamps
- Mayorships — current venues where a user is mayor
- Lists — saved, created, and followed lists with full venue details
- Tastes — user food/experience preferences
- Cross-user search — query checkins across all registered users at once
Requirements
- Python 3.11+
- uv (recommended) or pip
Setup
1. Install
cd foursquare_mcp
uv sync
Or with pip:
pip install -e .
2. Get a Foursquare OAuth Token
Option A — Interactive OAuth flow:
Create a .env file with your Foursquare app credentials (see .env.example):
FOURSQUARE_CLIENT_ID=your_client_id
FOURSQUARE_CLIENT_SECRET=your_client_secret
Then run the setup command:
foursquare-mcp setup --username alice
This opens a browser-based OAuth flow and saves the token automatically.
Option B — Manual token:
If you already have an OAuth token:
foursquare-mcp add-user --username alice --token YOUR_OAUTH_TOKEN
Tokens are stored in ~/.config/foursquare-mcp/users.json.
3. Run the Server
stdio (default, for local MCP clients):
foursquare-mcp serve
HTTP (for remote or shared access):
foursquare-mcp serve --transport streamable-http --host 0.0.0.0 --port 8006
MCP Client Configuration
Add to your MCP client config (e.g. .mcp.json):
stdio:
{
"mcpServers": {
"foursquare": {
"command": "foursquare-mcp",
"args": ["serve"]
}
}
}
Streamable HTTP:
{
"mcpServers": {
"foursquare": {
"type": "streamable-http",
"url": "http://localhost:8006/mcp"
}
}
}
Tools
User Management
| Tool | Description |
|---|---|
list_users |
List all registered Foursquare usernames |
add_user |
Register a user's OAuth token |
remove_user |
Remove a registered user |
get_user_details |
Get a user's Foursquare profile |
Checkins
| Tool | Description |
|---|---|
get_user_checkins |
Get checkins with optional date range and limit |
get_checkin_detail |
Get full detail for a single checkin (comments, photos, overlaps) |
search_all_users_checkins |
Search checkins across all registered users |
Venues
| Tool | Description |
|---|---|
get_venue_details |
Get full venue info by venue ID |
get_user_venue_history |
Get unique venues visited with visit counts and timestamps |
Social
| Tool | Description |
|---|---|
get_user_mayorships |
Get venues where the user is currently mayor |
get_user_lists |
Get all lists (saved, created, followed) |
get_list_detail |
Get full list detail with all venue items |
get_user_tastes |
Get taste preferences (food types, experiences) |
CLI Commands
foursquare-mcp serve Start the MCP server
foursquare-mcp setup Authenticate via OAuth and save token
foursquare-mcp add-user Register a token manually
foursquare-mcp remove-user Remove a registered user
foursquare-mcp list-users List registered users
Run foursquare-mcp --help for full option details.
Running as a systemd Service
# ~/.config/systemd/user/foursquare-mcp.service
[Unit]
Description=Foursquare MCP Server
[Service]
WorkingDirectory=/path/to/foursquare_mcp
ExecStart=/path/to/uv run foursquare-mcp serve --transport streamable-http --host 0.0.0.0 --port 8006
Restart=on-failure
[Install]
WantedBy=default.target
systemctl --user daemon-reload
systemctl --user enable --now foursquare-mcp
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。