# Praesago MCP Server

[![praesago-mcp MCP server](https://glama.ai/mcp/servers/marlybone/praesago-mcp/badges/score.svg)](https://glama.ai/mcp/servers/marlybone/praesago-mcp)

UK property area intelligence inside Claude, ChatGPT, Cursor or any MCP client.

Ask your AI assistant "is Walthamstow up and coming?" or "find areas in the
North West under £250k with strong fundamentals" and get live, validated
answers from Praesago's data — 2,292 England & Wales postcode districts scored
monthly from 30+ government data sources (Land Registry, ONS, police.uk, EPC,
planning registers and more).

This is a **hosted (remote) MCP server** — no install, no local process.

```
https://praesago.com/api/mcp
```

Transport: Streamable HTTP (JSON-RPC 2.0, protocol `2025-03-26`).

## Tools

| Tool | What it answers |
|---|---|
| `get_area_score` | Trajectory Score (0-100), percentile, gentrification stage, backtested calibration |
| `get_market_pulse` | **Live asking market**: active listings, price reductions, median asking, time on market |
| `get_area_context` | Full district profile: demographics, income, tenure, IMD, health, education, broadband, flood risk, prices |
| `search_areas` | Screen districts by score, region and price ceiling |
| `compare_areas` | Two districts side by side |
| `search` / `fetch` | ChatGPT deep-research compatible aliases |

`get_market_pulse` is collected weekly from property portals and exists
nowhere else: Land Registry reports completions months after the fact, and
the portals themselves do not publish reduction rates or time-on-market.

## Connect

**Claude (web / desktop):** Settings → Connectors → Add custom connector →
paste the URL above.

**Claude Code:**

```bash
claude mcp add --transport http praesago https://praesago.com/api/mcp
```

**Cursor / other clients:** add a remote server with transport `http` and the
URL above.

**Local stdio (any MCP client, or air-gapped configs):** this repo ships a
zero-dependency stdio proxy to the hosted endpoint, in both runtimes:

```bash
node server.js                         # Node 18+, no npm install needed
python server.py                       # Python 3.9+, no pip install needed
# or
docker build -t praesago-mcp . && docker run -i praesago-mcp
```

Set `PRAESAGO_API_KEY` in the environment to unlock full area context.

## Tools

| Tool | Access | What it does |
|---|---|---|
| `get_area_score` | Free | Trajectory Score (0–100), national percentile, gentrification early-warning stage and measured historical calibration for any postcode district or full postcode |
| `search_areas` | Free | Screen all 2,292 districts by minimum score, region and maximum median price |
| `compare_areas` | Free | Two districts side by side |
| `get_area_context` | API key | Full area payload: demographics, income, tenure, deprivation (IMD), health, education, broadband, flood risk, council tax, politics, 12-month prices |
| `search` / `fetch` | Free / key | Aliases for ChatGPT deep-research connector compatibility |

Free tier: 30 calls/day, no key needed. Full access: add a
[Praesago API key](https://praesago.com/pricing) as `Authorization: Bearer`
or `X-API-Key`.

## Honesty notes

The Trajectory Score measures the strength of signals that have **historically
preceded** property price growth. It is a screening aid, not a prediction —
and the server's instructions tell your assistant exactly that. Praesago
publishes its full validation record, including the windows where the model
failed: [praesago.com/validation](https://praesago.com/validation).

Calibration (measured across ~36,700 backtested district-windows): districts
scoring 60+ beat their county's median price growth in 63% of three-year
windows; districts under 40 managed 37%.

## Example prompts

- "Is Walthamstow up and coming?"
- "Find postcode districts in Greater Manchester under £250k with strong fundamentals"
- "Compare E17 and N15 for a buy-to-let"

## Links

- Docs: https://praesago.com/docs/mcp
- Methodology: https://praesago.com/methodology
- Validation: https://praesago.com/validation
- REST API: https://praesago.com/developer

Data sources: HM Land Registry, ONS, police.uk, DfE, Environment Agency,
Ofcom, DLUHC and more (Crown copyright / OGL). Coverage: England & Wales.