NY Data Registry — Claude Code MCP server: features and setup
The NY Data Registry MCP server for Claude Code locates New York open-data portals by place or subject. Its index search can return datasets, column names, and formats.
MIT ★ 14 updated 2026-11-16
claude mcp add --env NY_DATA_INDEX_AUTO_DOWNLOAD=<NY_DATA_INDEX_AUTO_DOWNLOAD> --env SOCRATA_APP_TOKEN=<SOCRATA_APP_TOKEN> --transport stdio ny-data-registry -- npx -y @betanyc/ny-data-registry-mcpWhat NY Data Registry does
NY Data Registry identifies which New York portal covers a location or subject and explains how to reach its data. Registry tools locate sources and API details, and can check whether a portal responds. Dataset and column searches rely on a harvested SQLite index.
Index tools search Socrata and ArcGIS portals, with filters for topics, places, formats, and more. Results can be grouped or exported as CSV; a separate MCP, such as socrata, handles data queries. The index tools need a harvested index, and the install commands include SOCRATA_APP_TOKEN in their configuration.
What it helps with
- locate an open-data portal for a place or subject
- find the API details for a portal
- search datasets and inspect their column names
- group dataset matches by portal
- export matching datasets to CSV
Example prompts for the agent
- Find New York open-data portals covering housing.
- Which datasets contain tax-lot information?
- Show the columns and formats for datasets from this portal.
How to install NY Data Registry
Platform CLI
claude mcp add --env NY_DATA_INDEX_AUTO_DOWNLOAD=<NY_DATA_INDEX_AUTO_DOWNLOAD> --env SOCRATA_APP_TOKEN=<SOCRATA_APP_TOKEN> --transport stdio ny-data-registry -- npx -y @betanyc/ny-data-registry-mcpReplace <…> with your own values
Platform CLI
codex mcp add ny-data-registry --env NY_DATA_INDEX_AUTO_DOWNLOAD=<NY_DATA_INDEX_AUTO_DOWNLOAD> --env SOCRATA_APP_TOKEN=<SOCRATA_APP_TOKEN> -- npx -y @betanyc/ny-data-registry-mcpReplace <…> with your own values
Config file
{
"mcpServers": {
"ny-data-registry": {
"command": "npx",
"args": [
"-y",
"@betanyc/ny-data-registry-mcp"
],
"env": {
"NY_DATA_INDEX_AUTO_DOWNLOAD": "<NY_DATA_INDEX_AUTO_DOWNLOAD>",
"SOCRATA_APP_TOKEN": "<SOCRATA_APP_TOKEN>"
}
}
}
}Replace <…> with your own values
Config file
{
"servers": {
"ny-data-registry": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@betanyc/ny-data-registry-mcp"
],
"env": {
"NY_DATA_INDEX_AUTO_DOWNLOAD": "<NY_DATA_INDEX_AUTO_DOWNLOAD>",
"SOCRATA_APP_TOKEN": "<SOCRATA_APP_TOKEN>"
}
}
}
}Replace <…> with your own values
Platform CLI
gemini mcp add -e NY_DATA_INDEX_AUTO_DOWNLOAD=<NY_DATA_INDEX_AUTO_DOWNLOAD> -e SOCRATA_APP_TOKEN=<SOCRATA_APP_TOKEN> ny-data-registry npx -y @betanyc/ny-data-registry-mcpReplace <…> with your own values
Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about NY Data Registry
How do I install NY Data Registry in Claude Code?
Run claude mcp add ny-data-registry-mcp -e SOCRATA_APP_TOKEN=<SOCRATA_APP_TOKEN> -- npx -y @betanyc/ny-data-registry-mcp. To use index tools, also configure NY_DATA_INDEX_AUTO_DOWNLOAD and SOCRATA_APP_TOKEN.
What does NY Data Registry do?
It finds New York open-data portals by location or subject, explains how to access them, and searches datasets through an index. Another MCP server handles the actual data query.
What does NY Data Registry need to run?
Its setup uses SOCRATA_APP_TOKEN, and index tools need a harvested SQLite index. The Claude Code command runs the package with npx.