Skip to main content
Glama

ProofPrice

What did people actually pay for X, here, recently?

ProofPrice is a free remote MCP server and REST API that answers that question for AI agents. Every answer is a range or an official figure, never a bare number: the median, the typical range (10th to 90th percentile), the sample size, the period covered, a confidence label, and the official source with its licence. The numbers come from official open data. Asking prices are never presented as prices paid.

This repository holds the integration files only (configs, plugin manifests, docs). The hosted service code is not published here.

Add it to your AI app

The server is read-only and needs no API key or sign-in. Add the address below and you are done.

https://proofprice.com/api/mcp

Claude (claude.ai and Claude Desktop)

Open Customize > Connectors, add a custom connector, paste the address, and choose no sign-in. Switch it on for a chat from the + menu. On Team and Enterprise plans an Owner adds it under Organization settings first.

If your Claude Desktop only reads a config file, this fallback uses the mcp-remote bridge:

{
  "mcpServers": {
    "proofprice": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://proofprice.com/api/mcp"]
    }
  }
}

Claude Code

One command:

claude mcp add --transport http proofprice https://proofprice.com/api/mcp

Or install this repository as a plugin (it is also a one-plugin marketplace):

claude plugin marketplace add Screed369/proofprice-mcp
claude plugin install proofprice@proofprice

Cursor

Add to Cursor

Or paste this deeplink into a browser or terminal: cursor://anysphere.cursor-deeplink/mcp/install?name=proofprice&config=eyJ1cmwiOiJodHRwczovL3Byb29mcHJpY2UuY29tL2FwaS9tY3AifQ==

Or add to .cursor/mcp.json (one project) or ~/.cursor/mcp.json (all projects):

{
  "mcpServers": {
    "proofprice": { "url": "https://proofprice.com/api/mcp" }
  }
}

VS Code and GitHub Copilot

Install in VS Code

Or run MCP: Add Server from the Command Palette, choose HTTP, and paste the address. Or add to .vscode/mcp.json in a workspace:

{
  "servers": {
    "proofprice": { "type": "http", "url": "https://proofprice.com/api/mcp" }
  }
}

The VS Code URL handler also accepts vscode:mcp/install? followed by the URL-encoded JSON {"name":"proofprice","type":"http","url":"https://proofprice.com/api/mcp"}.

ChatGPT (developer mode)

Developer mode is available on Pro, Plus, Business, Enterprise and Education accounts on the web.

  1. Go to Settings > Apps > Advanced settings and turn on Developer mode.

  2. Choose Create app, enter https://proofprice.com/api/mcp, and select No Authentication.

  3. Start a chat, choose Developer mode from the + menu, and select the app. If ChatGPT does not call a tool, name it in your prompt, for example "use pricepaid_home_sales".

Windsurf

Open the MCP config file from Cascade (~/.codeium/windsurf/mcp_config.json in most installs) and add:

{
  "mcpServers": {
    "proofprice": { "serverUrl": "https://proofprice.com/api/mcp" }
  }
}

Any other MCP client, or plain REST

Use the address above with the Streamable HTTP transport. Without an MCP client, call the REST API directly. No key is needed and CORS is open:

curl "https://proofprice.com/api/v1/fuel?country=ES&area=42&fuel=diesel"

Related MCP server: datacommons

Tools

Tool

What it answers

pricepaid_home_sales

What homes actually sold for in a town, district or postcode: median, typical range, number of sales (or the official average where a statistics office publishes one).

pricepaid_rent_prices

Rents actually agreed or paid (registered tenancies, rental bonds, official rent surveys), never asking rents.

pricepaid_fuel_prices_now

Pump prices for petrol, diesel and other fuels near a place, or the official national average.

pricepaid_energy_prices

Household electricity and gas prices per kWh, official published averages.

pricepaid_grocery_prices

Official average shop prices of everyday items (milk, bread, eggs, coffee...).

pricepaid_medicine_prices

Regulated or reimbursement prices of medicines, and what pharmacies paid to buy them.

pricepaid_medical_prices

What Medicare paid for doctor visits, tests and hospital stays (US).

pricepaid_airfare_prices

Average airfares actually paid between two airports (US).

pricepaid_contract_prices

What public buyers paid in awarded government contracts.

pricepaid_farm_prices

Official farm-gate and wholesale prices of cereals, meat, milk, dairy, eggs, olive oil, fruit and vegetables.

pricepaid_land_prices

Average price or yearly rent per hectare of farmland.

pricepaid_car_prices

Singapore Certificate of Entitlement (COE) premiums.

pricepaid_coverage

Which countries, categories, sources, licences and update frequencies are covered right now.

More tools and countries are added often. pricepaid_coverage and https://proofprice.com/sources always show the current list, with every source's publisher, licence and limits. All tools are read-only; parameters are described in each tool's schema and in the OpenAPI file.

Example answer

Illustrative figures, not real data. The shape is what the server returns.

Prompt: "What did terraced houses in Bristol actually sell for recently?" (tool: pricepaid_home_sales, arguments country=GB, town=Bristol, property_type=terraced)

{
  "query": { "country": "GB", "town": "Bristol", "postcode_district": null, "property_type": "terraced", "months": 12 },
  "status": "ok",
  "price": {
    "currency": "GBP",
    "unit": "per home",
    "value": 395000,
    "measure": "median",
    "median": 395000,
    "typical_low": 265000,
    "typical_high": 575000,
    "min": 150000,
    "max": 1250000
  },
  "sample_size": 1240,
  "period": { "from": "2025-09-01", "to": "2026-08-29" },
  "confidence": "high",
  "price_type": "transaction",
  "as_of": "2026-10-14T09:00:00.000Z",
  "source": {
    "id": "gb-hmlr-ppd",
    "name": "Price Paid Data",
    "publisher": "HM Land Registry",
    "licence": "Open Government Licence v3.0",
    "licence_url": "https://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/",
    "attribution": "Contains HM Land Registry data © Crown copyright and database right. This data is licensed under the Open Government Licence v3.0.",
    "url": "https://www.gov.uk/government/statistical-data-sets/price-paid-data-downloads"
  },
  "method": "Median and 10th–90th percentile of standard-price residential sales (category A) recorded by HM Land Registry in the selected area and period. Repossessions, buy-to-let and transfers to companies (category B) are excluded.",
  "method_url": "https://proofprice.com/method",
  "summary": "Terraced homes in Bristol sold for a median of £395,000; typical range £265,000–£575,000. Based on 1240 records from HM Land Registry (2025-09-01 to 2026-08-29). Confidence: high.",
  "notice": "Information, not a valuation or offer. Ranges describe past recorded prices."
}

Confidence is low for 5 to 14 records, medium for 15 to 49 and high for 50 or more. With fewer than 5 records the server returns status: "insufficient_data" and no figure.

Some sources publish one official figure instead of a list of records, for example household energy prices from Eurostat. Those answers have price_type: "official_statistic", a statistic.reference_period (such as a half-year), and empty range fields and sample size, because the publisher gives one average and ProofPrice does not invent a range.

Data sources and licences

Every answer comes from an official publisher (land registries, statistics offices, regulators, ministries, the European Commission and Eurostat) under an open licence, and carries its attribution line. Please keep that line when you show a figure. The live list of sources, with each licence, update frequency and known limits, is at https://proofprice.com/sources and https://proofprice.com/api/v1/coverage.

How to use the numbers honestly

  • Cite the source and attribution line returned with each answer.

  • Every answer says what kind of figure it is (price_type): recorded sale prices, official published averages, posted shelf or pump prices, contract award values or regulated reimbursement prices. Never present one kind as another.

  • A range is not an asking price, a valuation, an offer or advice.

  • Home-sale answers are aggregates only. No individual address is ever returned.

  • No figure is returned from fewer than 5 records.

Privacy, terms and contact

Contributing

Want a country or category added? See CONTRIBUTING.md or email hello@proofprice.com.

Licence

The integration files in this repository are released under the MIT licence, copyright 2026 Simon Creedon. The licence covers only these files. It does not cover the hosted service's source code or the third-party data, which stays under its publishers' licences.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      D
      maintenance
      Cost of living, CPI, PPP, and regional price data for 190+ countries. Semantic endpoints with natural language parameters. MCP server for AI agents.
      MIT
    • A
      license
      Not graded
      quality
      C
      maintenance
      Enables AI agents to resolve places and statistical variables, then retrieve harmonised public statistics such as population, income, health, employment, and emissions with publisher provenance and child-place comparisons.
      90 npm
      MIT
    • F
      license
      Not graded
      quality
      A
      maintenance
      Enables AI agents to turn natural-language questions into ranked, authority-tagged primary data sources and hand-by-hand navigation instructions for reaching the exact data on official government, international organization, and research sites. Delivers a traceable evidence chain from query to original source, avoiding hallucinated figures and secondhand citations.
      -
    • A
      license
      Not graded
      quality
      B
      maintenance
      Real prices people actually paid (homes, rents, fuel, energy, groceries, medicines) from official open data, with source, period, sample size and licence on every answer. Read-only, no API key.
      MIT