> ## Documentation Index
> Fetch the complete documentation index at: https://ploti.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# The map agent

> Ask Ploti's map agent about the data in your coverage areas, and get maps, filters, snapshots, ordinance answers and exports back

The map agent is the AI assistant built into the Ploti app. You ask it questions in a chat beside a map. It reads the official records in your coverage areas (the counties your plan covers), highlights matching parcels on the map, answers zoning and ordinance questions from the local code, and saves results you can download. Every answer names its source.

To use it, open **Map** in the sidebar, or open a result from a finished research request, and ask in the chat panel. The chat header calls it the **General Agent**.

Use the map agent when the data is there and you want to explore it, follow up on a result, or refine it. Use a [research request](https://ploti.ai/docs/research/overview.md) when the answer needs data Ploti may not have yet, or a longer run you can walk away from.

![Ploti map showing McDonald's locations across the Orlando area, with the research outputs on the left and the result layer in the Map Key.](https://cdn.ploti.ai/static/docs/ploti-app/ca070795c4c17e5706d558014e775a3e9cfa16362e7d11175d974dad55d55c9e/result-on-map.webp)

A completed research result mapped across Orange County with its outputs and map key available for inspection.

## What you can ask for

> [card] Find and read data
>
> The agent searches the layers, tables and documents in your coverage areas
> and reads them in full, by place, wherever the map is looking.

> [card] Maps and filters
>
> A filter draws a layer with conditions, such as parcels over 20 acres owned
> by an LLC, and stays live as the data updates.

> [card] Snapshots
>
> A snapshot saves a fixed set of places, such as your shortlist, as a map
> layer and a table you can come back to.

> [card] Ordinances and documents
>
> Ask what a zoning code allows, or what a plan or meeting record says. The
> agent cites the section it read.

> [card] Analysis
>
> Rankings, comparisons, statistics and spatial joins, run on Ploti's servers.

> [card] Exports and files
>
> Export a snapshot's or table's visible rows as CSV, or ask the agent to save
> a file, such as a CSV, a PDF or an image.

The agent can also search the web for context, read files you attach with **Attach files**, and start a research request from the chat when the data you need isn't in Ploti yet.

## Ask well

Name the place, the criteria that matter, and what a useful answer looks like:

```text theme={null}
Compare these three parcels for a 60,000 square foot industrial building.
Account for access, flood exposure, nearby utilities and zoning. Return a
comparison table and call out what still needs verification.
```

Follow-ups can refer to what's on the map:

```text theme={null}
Tighten this to parcels with at least 10 usable acres.
```

```text theme={null}
Why were these properties excluded?
```

```text theme={null}
What does the city's code require for parking at a self-storage facility here?
```

## Inspect the result

Pan and zoom, search for a place, and select any feature to see its attributes. The map key lists what is drawn and lets you show, hide and restyle each layer. Some layers, such as parcels, appear only once the map is zoomed in close enough; the agent reads them by place either way.

## Credits

The map agent uses credits, Ploti's unit for paid work, as it works. Everyone in your workspace (your team's Ploti account) shares one balance. See [Coverage areas and pricing](https://ploti.ai/docs/getting-started/coverage-areas.md).

> [!WARNING]
>
> Public records can be incomplete or out of date. Confirm material facts with
> the source the agent cites, or with the relevant agency, before you rely on
> them.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.