# Introduction

> What Cadastre is, and what it offers people, their tools and AI agents.

Cadastre is a data source for real-estate and site-selection data. It brings together the public datasets a site-selection decision rests on (markets, labor, demographics, infrastructure, hazards and health) for U.S. metro areas, with each figure traced to its publisher, its licence and the method that produced it. Property and parcel data will join them as it comes online. Cadastre does not carry listings.

The same data drives the app, where consultants and developers screen and rank metros and hand clients a report, and it is offered to your own tools and AI agents through a read-only API, a remote MCP server and a command-line tool.

## What's available today

- **Metro-level data.** 19 datasets of measures for U.S. metro areas, from U.S. federal agencies and OpenStreetMap: see [Datasets](https://getcadastre.com/docs/datasets). Each dataset page lists its publisher, licence, required credit, period, geographies, measures with their units and how each is assembled for a metro, and its known gaps.
- **Screening and ranking in the app.** Requirements and weighted criteria over those measures, a ranked register, saved snapshots, and reports as PDF, PowerPoint, CSV and Excel.
- **Documentation an agent can read.** Every page here is also Markdown at the same address with `.md` added (this page is [https://getcadastre.com/docs.md](https://getcadastre.com/docs.md)), [/llms.txt](https://getcadastre.com/llms.txt) lists every page, and [/llms-full.txt](https://getcadastre.com/llms-full.txt) is all of them in one file.
- **A read-only API** for the data, screening and ranking, in beta and free while it is, with no account needed for the public datasets; sign in for higher limits, with API access turned on for the account (write to support@getcadastre.com to ask for it). See [API](https://getcadastre.com/docs/api), and [Connect an AI agent](https://getcadastre.com/agents) for setting up each agent.
- **Open data.** The OpenStreetMap-derived transmission-line dataset, offered under the ODbL on the [Open data](https://getcadastre.com/open-data) page.
- **A remote MCP server** at `https://getcadastre.com/mcp`, so AI assistants such as Claude Code, Codex and Cursor can use the same data with the same access, with no sign-in, or signed in through the browser for higher limits. See [MCP server](https://getcadastre.com/docs/mcp).
- **A command-line tool**, `cadastre`, for the terminal and scripts: screens, rankings, comparisons and metro profiles with their attribution, and a command that prints, or writes, your agent's MCP setup. See [CLI](https://getcadastre.com/docs/cli).

## Coming next

- **Parcel and property data**, once its sources' rights are cleared. Parcel-level data is not available yet: the launch is metro-level.

## What these docs cover

- [Getting started](https://getcadastre.com/docs/getting-started): using the API without an account, signing in, API keys and connecting an agent.
- [API](https://getcadastre.com/docs/api), [MCP server](https://getcadastre.com/docs/mcp) and [CLI](https://getcadastre.com/docs/cli): the ways in for your tools and agents.
- [Datasets](https://getcadastre.com/docs/datasets): one page per dataset, generated from the code that imports it, so a page cannot say something different from what is imported.
- [Methodology](https://getcadastre.com/docs/methodology): how a metro value is made from county data, which vintage is read, and where coverage is missing.
- [Rights and attribution](https://getcadastre.com/docs/rights): how each source is licensed, which sources are restricted and why, and the credit each one requires.
- [Data freshness](https://getcadastre.com/docs/freshness): the period each dataset describes, when it was last refreshed and when it is next checked.

> Docs pages describe data; they do not publish data values. Values come from the app and the API, always with their source's credit.