Skip to main content
An AI assistant connected to Ploti can reach only what you can reach in the Ploti app: the official records in your workspace’s coverage areas (the counties your plan covers), and the maps and files saved in your workspace (your team’s Ploti account). It follows the same access rules as the app, with no separate path for assistants.

​
Signing in

You connect an assistant by signing in on Ploti’s own page and approving the connection. You never paste a Ploti password into the assistant.
  • The approval screen names the assistant and the workspace before you grant access.
  • The assistant receives short-lived access tokens, and its refresh token is replaced every time it is used.
  • Ploti’s tools never return passwords or tokens.
For engineers: the connection is OAuth 2.1 over HTTPS, with dynamic client registration and authorization-code exchange with PKCE (S256). Redirect URLs must match the registered value exactly, and authorization codes are single-use.

​
What an assistant can reach

Every connection belongs to one Ploti member and one workspace. Ploti’s servers check that identity on every call.
  • A connection reaches only the workspace you approved.
  • It reads data only in the workspace’s coverage areas.
  • Ploti’s servers enforce what your plan includes and which datasets you may read. The assistant is not trusted to.
  • A map belongs to the member who made it, not only to the workspace. The map’s worker, the private database that runs its analysis, belongs to the map.
  • Saved maps and files follow the same access rules as everything else in the workspace.

​
What is sent to Ploti

When you ask an assistant to use Ploti, the assistant’s provider sends Ploti the name of the tool and the inputs for that request. Ploti sends back the result, such as a short preview of the data, the source, or a link to a map or file. Ploti does not ask for the rest of your conversation. The assistant’s provider decides what it sends with each call, so read that provider’s privacy terms as well. Don’t put passwords, full payment card numbers, health information, government ID numbers or other sensitive data you don’t need into prompts or files.

​
What Ploti stores, and for how long

For the full terms, read the Ploti Privacy Policy, Terms of Service, Acceptable Use Policy, and Subprocessor List.

​
Disconnect an assistant

Disconnect Ploti in the assistant’s connector settings at any time. Workspace admins can also remove access, in Ploti or in the assistant’s admin settings. Once disconnected, the assistant can make no further calls. Maps and files it saved stay in the workspace. If you suspect someone else has access, disconnect the assistant and contact support@ploti.ai.

​
Intended use

Ploti is for lawful property, land-use, ordinance, site-selection and geospatial research. It must not be used for unlawful surveillance, harassment or discrimination, to get around access controls, or to decide whether a person is eligible for housing, lending, insurance, employment or similar. Property and ordinance data can be incomplete, delayed or inconsistent between sources. Ploti’s results support research. They are not legal, surveying, title, environmental, engineering or investment advice. Confirm important findings with the cited source and the right professional.

​
Technical compatibility

Ploti’s connector uses MCP (Model Context Protocol) over Streamable HTTP. Every tool returns a plain text result, so it works in assistants that can’t show interactive maps or tables. Each tool has a title and is labeled as reading, saving or deleting, so an assistant can ask you to confirm before it changes anything. For the standards Ploti follows, see the Model Context Protocol specification, Anthropic Software Directory Policy, and OpenAI plugin guidelines.