socrata-mcp
An MCP server that provides tools to search, query, and aggregate data from any Socrata open data portal.
README
Socrata MCP Server
An MCP server that gives Claude tools to explore and query any Socrata open data portal.
Tools
| Tool | Description |
|---|---|
search_datasets |
Search for datasets by keyword, optionally scoped to a specific portal domain |
get_dataset_metadata |
Get column definitions and metadata for a dataset |
query_dataset |
Query rows using SoQL WHERE / SELECT / ORDER |
aggregate_dataset |
GROUP BY aggregation with count/sum/avg etc. |
list_portal_categories |
List dataset categories on a specific portal |
Installation
Prerequisites
You need Python 3.10 or newer. Check your version:
python3 --version
If you need to install or upgrade Python:
- macOS:
brew install python(requires Homebrew) - Ubuntu/Debian / WSL:
sudo apt update && sudo apt install -y python3 python3-venv python3-pip - Fedora/RHEL:
sudo dnf install python3 - Windows 11 native: Download from python.org and check Add to PATH during install
1. Clone or copy this folder
git clone https://github.com/your-username/socrata-mcp.git
cd socrata-mcp
Or just copy the server.py and requirements.txt files into a folder of your choice.
2. Create a virtual environment
python3 -m venv .venv
3. Activate the virtual environment
| Platform | Command |
|---|---|
| macOS / Linux / WSL | source .venv/bin/activate |
| Windows (cmd.exe) | .venv\Scripts\activate.bat |
| Windows (PowerShell) | .venv\Scripts\Activate.ps1 |
Your prompt will change to show (.venv) when it is active.
4. Install dependencies
pip install -r requirements.txt
5. Verify the install
python3 server.py --help
You should see an error like:
RuntimeError: SOCRATA_DOMAIN environment variable is required
That means the server loaded correctly — it just needs your credentials configured (see below).
Configuration
Both environment variables are required. The server will not start without them.
| Variable | Description | Example |
|---|---|---|
SOCRATA_DOMAIN |
Hostname of the Socrata portal | data.delaware.gov |
SOCRATA_APP_TOKEN |
API token from your Socrata account | abc123xyz... |
To get an API token: log in to your Socrata portal → Developer Settings → Create New App Token.
Add to Claude Desktop
Claude Desktop reads its MCP server list from a JSON config file. Edit the file for your platform:
| Platform | Config file location |
|---|---|
| macOS | ~/Library/Application Support/Claude/claude_desktop_config.json |
| Windows 11 | %APPDATA%\Claude\claude_desktop_config.json |
| WSL | Use the Windows path above — Claude Desktop runs on Windows, not inside WSL |
macOS / Linux (native)
{
"mcpServers": {
"socrata": {
"command": "/absolute/path/to/socrata-mcp/.venv/bin/python3",
"args": ["/absolute/path/to/socrata-mcp/server.py"],
"env": {
"SOCRATA_DOMAIN": "data.delaware.gov",
"SOCRATA_APP_TOKEN": "your-api-key-here"
}
}
}
}
Replace /absolute/path/to/socrata-mcp with the real path. Find it by running pwd inside the project folder.
Windows 11 (native Python)
{
"mcpServers": {
"socrata": {
"command": "C:\\Users\\YourName\\socrata-mcp\\.venv\\Scripts\\python.exe",
"args": ["C:\\Users\\YourName\\socrata-mcp\\server.py"],
"env": {
"SOCRATA_DOMAIN": "data.delaware.gov",
"SOCRATA_APP_TOKEN": "your-api-key-here"
}
}
}
}
Windows 11 with WSL
Claude Desktop runs on Windows but can launch a WSL process via wsl.exe:
{
"mcpServers": {
"socrata": {
"command": "wsl.exe",
"args": [
"--exec",
"/home/your-wsl-username/socrata-mcp/.venv/bin/python3",
"/home/your-wsl-username/socrata-mcp/server.py"
],
"env": {
"SOCRATA_DOMAIN": "data.delaware.gov",
"SOCRATA_APP_TOKEN": "your-api-key-here"
}
}
}
}
Replace your-wsl-username with your WSL username (run whoami inside WSL to check).
After editing the config, restart Claude Desktop and you should see the Socrata tools available.
License
This project is licensed under the MIT License.
This code was generated using Claude by Anthropic.
Example queries
- "What datasets are available on the Delaware open data portal?"
- "Show me the top vendors by contract spend in FY2025"
- "What columns are in dataset sifm-293u?"
- "Which agencies spent the most on cloud solutions last year?"
- "Search for education datasets on data.delaware.gov"
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。