Connect Ploti to Cursor

Cursor Agent is a coding assistant that runs on your computer and writes and runs code for you. Connected to Ploti, it can download official parcel, zoning and permit records for your counties and build with them: a screening script, say, or a data pipeline. It can also show results on a Ploti map. There are two ways to connect it, and you can use both:

The CLI (Ploti’s command-line tool) and its Python library let Cursor download Ploti’s data onto your computer and work on it there.
MCP, the standard way AI tools plug in outside services, adds Ploti’s map tools, so it can save results on a Ploti map you open in the app.

CLI: download the data
MCP: results on a Ploti map

Install the CLI, sign in once, and give Cursor Ploti’s skills, short instruction files that teach it the commands.

Terminal

npm install -g @ploti/cli
ploti login
ploti skills install <folder Cursor reads skills from>

Then ask Cursor for the data you need. It finds it with ploti datasets search, reads it with ploti datasets rows, and downloads a dataset with ploti datasets download as GeoParquet, a table format most data tools read.Try: Use Ploti to pull the flood zones for Travis County, Texas, and load them into this project.Keys and setup covers the CLI and the Python library, and Find datasets, Read rows and Download a dataset explain what it can do.

Find MCP settings

Open Cursor Settings, then Tools & MCP, and select New MCP Server. Use ~/.cursor/mcp.json to make Ploti available in every project, or .cursor/mcp.json to limit it to the current project.

Add Ploti

Add the Ploti entry below under mcpServers, then save the file.

mcp.json

{
  "mcpServers": {
    "ploti": {
      "url": "https://mcp.ploti.ai/api/mcp"
    }
  }
}

Approve the connection

Return to Tools & MCP. When Ploti shows Needs login or Connect, select it. Sign in to Ploti, choose a workspace, and approve the connection.

Try it in Agent

Open the Agent panel and try: Use Ploti to find vacant parcels over 10 acres near Tampa, Florida, and show them on a map. Tools & MCP shows whether Ploti is connected.

Show results on a map explains what it can save on a map.

See Cursor’s MCP documentation for its current steps.

Start here

Chat assistants

Coding agents

Always-on assistants

Reference

Connect Ploti to Cursor