mcpField
R
RuralCheck

RuralCheck

Remote
ruralcheck.euBusiness & Marketing0Updated 1d ago

About RuralCheck

RuralCheck MCP server lets AI agents analyse rural land plots in Europe using RuralCheck data. Reports provide selected checks with computed figures, caveats and sources, spanning risks, local services, planning, climate and infrastructure.

Key features

  1. 01Analyse selected plot checks
  2. 02Assess hazards and local conditions
  3. 03Check coverage and available checks
  4. 04Save and organize plot reports
  5. 05Return figures with caveats and sources

Use cases

  1. 01Assess flood, wildfire or landslide exposure
  2. 02Review noise, sunshine or local services
  3. 03Check planning, land-register and tax details
  4. 04Compare plots using sourced reports

MCP support

The MCP server exposes RuralCheck tools for place lookup, coverage checks, plot analysis and report retrieval, as well as saving and organizing plots. Results include figures, caveats and data attribution.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://ruralcheck.eu/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

Details

From the project README

RuralCheck MCP server

This server lets an AI agent analyse a plot of rural land in Europe with RuralCheck. Ask Claude, ChatGPT or any other MCP client whether a field floods, how loud the road is, how many hours of sun the house gets in January, or which land-register plot it is. The agent calls RuralCheck and gets back computed figures with their sources.

What it can answer

A report is a set of checks. You ask only for the ones you need, and each comes back with its figure, its caveats and where it came from.

GroupChecks
Viewhow far the view runs, and what is in view
Local areanearest village and town, people nearby, shops and services, bodies of water, emergency services, drive times, local economy, visitors, sales nearby, town websites, hunting
Annoyancesroad and rail noise, aircraft, nuisances, bites and stings, wild animals, burglary
Plotland register, paths on the plot, municipality, planning zone, rules here, tax when you buy, yearly property tax, terrain, protected areas
Weather and climateweather, sun, wind, snow, night sky, air quality, climate in 2050
Risksflooding, coastal flooding, wildfire, landslides, avalanches, earthquake, termites and dry rot
Infrastructuresolar panels, groundwater, water restrictions, internet
Vegetationtrees, pollen, what grows here
Accesspublic transport, access road

list_checks returns the current list with what each check covers, how long it takes and whether the free plan runs it.

Every number is derived from official data. A report is not a land survey, a valuation or legal advice. It covers the parts of Europe where the ground is measured finely enough, and check_coverage says whether a point is one of them.

Tools

Every tool needs a RuralCheck account. The free plan runs the free checks, with the same limits as on the map, and a paid plan runs every check. An account can start one new report a minute through analyse_plot. Fetching one with get_report does not count. More than one a minute is a Custom solution.

Nothing here deletes a plot for good. A deleted plot goes to the Trash on the account, and is removed 30 days later or when the reader empties the Trash on the website.

ToolWhat it does
find_placeA village, town, region, country or postcode to coordinates. An area comes with its extent and whether reports are made there. Street addresses are found in Norway, Belgium and Denmark, and elsewhere reduced to their town.
check_coverageWhether a point can be analysed, and how finely its ground is measured.
list_checksEvery check, with what it covers, how long it takes and whether the free plan runs it.
analyse_plotRuns the checks you name for one point and returns the results.
get_reportFetches a report that took longer than one tool call waits.
sample_reportA real full report for 4 ha of land in the Dordogne valley, to see what each check returns.
upload_listingsProfessional plan. Sends an agent's listings: each becomes a saved plot with a full report and a share link, run one after another.
get_listing_uploadAn upload's progress: each listing's status, its share link, or why it was not run.
list_listing_uploadsThe account's uploads and how far each has got.
list_plotsThe plots the account keeps, with the key figure of each check. Or the Trash.
get_plotOne kept plot with its report, as it was when it was kept.
save_plotKeeps a finished report in My plots. Only a report that ran every check of its plan.
update_plotRenames a kept plot or replaces its note.
file_plotPuts a kept plot in a folder, or takes it out of one.
delete_plotMoves a kept plot to the Trash. It can be restored for 30 days.
restore_plotTakes a plot out of the Trash.
list_foldersThe account's folders.
create_folder, rename_folderMakes or renames a folder.
delete_folderDeletes a folder. Its plots stay, in no folder.

Answers are JSON, returned as text: the figures, their caveats, and the attribution for the data behind them. Map outlines and pictures are left out.

Use it

Hosted

The server is at https://ruralcheck.eu/mcp. It signs you in with OAuth, so Claude and ChatGPT need nothing but the address:

  • Claude: Customize, Connectors, +, Add custom connector, enter the address, then Connect and allow it on the RuralCheck page.
  • ChatGPT: turn on Developer mode under Settings, Security and login. Then open Plugins, select +, enter the address, and allow it on the RuralCheck page.

Clients that take a header can use an API key from your account page instead. In Claude Code:

claude mcp add --transport http ruralcheck https://ruralcheck.eu/mcp \
  --header "Authorization: Bearer $RURALCHECK_API_KEY"
On your own machine

Python 3.10 or newer. With uv:

uvx --from git+https://github.com/cmosig/ruralcheck-mcp ruralcheck-mcp

Claude Desktop, Cursor and most other clients take a block like this:

{
  "mcpServers": {
    "ruralcheck": {
      "command": "uvx",
      "args": ["--from", "git+https://github.com/cmosig/ruralcheck-mcp", "ruralcheck-mcp"],
      "env": { "RURALCHECK_API_KEY": "rc_..." }
    }
  }
}

ruralcheck-mcp --http --port 8765 serves streamable HTTP instead of stdio, with the OAuth sign-in in front of it. A request without a token gets a 401 that points at /.well-known/oauth-protected-resource/mcp, which names ruralcheck.eu as the authorization server. The token a client comes back with is an ordinary API key. The server keeps none: it checks each token with the API and passes it on.

VariableDefaultMeaning
RURALCHECK_API_KEYnoneYour key, for stdio. Required.
RURALCHECK_BASE_URLhttps://ruralcheck.euWhere the API is.
RURALCHECK_WAIT_S45How long one tool call waits for a report before handing back an id.
RURALCHECK_PUBLIC_URLhttps://ruralcheck.euWhere the hosted server is reached, for its OAuth metadata.
RURALCHECK_AUTH_URLthe public URLThe OAuth authorization server.

Attribution

The data behind each answer is published under licences that require credit. Every report carries an attribution list. Show it to whoever reads the results.

Develop

python -m venv .venv && .venv/bin/pip install -e . pytest
.venv/bin/python -m pytest

tests/test_server.py runs against a fake API and needs no network. tests/test_live.py drives every tool against a real instance and is skipped unless RURALCHECK_LIVE_URL and RURALCHECK_API_KEY are set. Run it whenever the API changes.

The contract lives under /api/v1/ on ruralcheck.eu: checks, coverage, search, sample, reports with a long-polling ?wait= and ?drawing=false, and plots and folders for My plots.

Licence

The code in this repository is MIT licensed. The reports it fetches are covered by the RuralCheck terms and by the licences listed in each report's attribution.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.