SuufScan MCP Server
Converts the land-office directory (SuufScan Explorer) from OpenStreetMap-derived location data — accepts arrays, wrapped objects, or GeoJSON FeatureCollections and returns ODbL attribution (© OpenStreetMap contributors) with office results.
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
SuufScan for Alexa+ — land buyer protection for Senegal and its diaspora
Track: Alexa+ · self-hosted MCP server (Streamable HTTP, MCP spec 2025-11-25) + a simulated Alexa+ voice experience in the browser.
Millions of Senegalese live abroad, and buying land back home is one of the biggest decisions a family makes. Most of them buy from thousands of kilometres away, through relatives or brokers, and land fraud is widely reported in the Senegalese press: the same plot sold twice, fake titles, inherited land sold without the other heirs, plots in unapproved subdivisions.
SuufScan is a Senegalese PropTech that helps buyers protect themselves before they buy. This repository exposes SuufScan's buyer-protection knowledge and its directory of land offices as an MCP server, so a voice assistant like Alexa+ can answer questions such as:
"Alexa, I live in New York and want to buy land in Mbour for my mother. The seller has a délibération. What should I do before I pay?"
The answer is a set of concrete steps, the documents to ask for, and the offices to go to. It never says a plot is "safe" and never gives a price: it helps the buyer ask the right questions to the right people.
What's inside
Path | What it is |
| Express app. |
| The 4 MCP tools (below). Every result has a short |
| Buyer-protection steps per land status, warning signs, glossary of Senegalese land documents. |
| Loads the land-office directory (SuufScan Explorer) from a URL or a local file, ranks offices by distance. |
| Approximate town-centre coordinates for communes of the Dakar–Thiès–Mbour corridor and the Saint-Louis region. |
| Simulated Alexa+ experience: voice in (Web Speech API), deterministic intent routing, real MCP JSON-RPC calls to |
| End-to-end test with the official MCP TypeScript SDK client over Streamable HTTP. |
MCP tools
Tool | Use it when the user asks… |
| "What should I do / ask for before buying?" — |
| "Where should I go?" — nearest tax and Domains centre (with its cadastre desk), town hall, prefecture, urban planning service |
| "What are the scams to watch out for?" |
| "What is a NICAD / an état des droits réels / a délibération?" |
All tools are read-only (readOnlyHint: true).
Related MCP server: ECRVSP Documentos: Transferência de Proprietário
Run it
Requires Node.js 18+.
npm install
npm start # MCP server on http://localhost:3333/mcp, simulator on http://localhost:3333/
npm test # end-to-end MCP testOpen http://localhost:3333/ in Chrome, tap the ring and speak (English), or click an example question.
Configuration
Variable | Default | Purpose |
|
| HTTP port |
| — | JSON source for the office directory (SuufScan Explorer). Accepts an array, |
|
| Local JSON file, used when no URL is set |
| SuufScan Explorer + © OpenStreetMap contributors, licensed under ODbL | Attribution returned with office results |
data/places.example.json contains placeholder entries only (labelled "Example…", located at town centres) so the project runs out of the box. Real office data is served by SuufScan and is not committed here.
Connect an MCP client
Any MCP client that speaks Streamable HTTP can use http://<host>:3333/mcp, for example the MCP Inspector:
npx @modelcontextprotocol/inspector
# Transport: Streamable HTTP · URL: http://localhost:3333/mcpFor Alexa+, the same endpoint is the integration point: deploy it behind HTTPS and register it as an MCP server in the Alexa+ developer Preview.
How it works
voice ──► Alexa+ (simulated in the browser)
│ intent routing → tool + arguments
▼
POST /mcp (JSON-RPC 2.0, MCP 2025-11-25, Streamable HTTP)
│ initialize → tools/list → tools/call
▼
SuufScan MCP server ──► buyer-protection content
│ └► land-office directory (SuufScan Explorer)
▼
{ speech, steps / offices / warning_signs, disclaimer } ──► spoken answerThe server is stateless: each HTTP request gets its own MCP server instance and transport, so it scales horizontally and needs no session store.
Scope and limits
General information to help a buyer protect themselves before purchase. Not legal advice. For a binding answer on a specific plot, ask a notary.
The tools never return a verdict on a plot or a price.
French legal designations ("titre foncier", "état des droits réels", "Certificat d'Urbanisme"…) are kept as-is: they are the names buyers will hear at the offices.
Voice input in the simulator is English (Chrome Web Speech API).
License
MIT — see LICENSE. Office directory data, when loaded from SuufScan, carries its own attribution (OpenStreetMap contributors, ODbL).
The SuufScan name and logo are © SuufScan and are not covered by the MIT license.
This server cannot be deployed
Maintenance
Related MCP Connectors
Beninese ANDF cadastral parcels by NUP, titre foncier or coordinates, with declared zones.
Sourced answers on FR/EU rules + French company search (SIREN, NAF) for agents, quoted and budgeted
Instant pricing for French land-registry documents (etat hypothecaire, deeds). Read-only, no auth.
French real estate data: cadastre, DVF sales, DPE energy ratings, price estimates, parcel context
Related MCP Servers
- AlicenseAqualityDmaintenanceProvides AI access to Korea's land use planning, urban regulations, and permit information through natural language queries. It integrates tools for land analysis, regulation verification, and public notice scraping to prevent hallucinations in property-related AI applications.819MIT
- AlicenseNot gradedqualityDmaintenanceProvides read-only access to official property transfer documents from ECRVSP, allowing AI agents to query document information via natural language.MIT
- AlicenseAqualityFmaintenanceEnables AI assistants to assess any parcel for micro-farm viability (climate, water, farmland context, access, flood risk) and dwelling feasibility, and to find real land for sale from free, keyless public data sources.51MIT
- AlicenseAqualityCmaintenanceGives AI assistants read-only, key-free access to Sentinel-2 NDVI time series, ESA WorldCover land-cover shares, and Copernicus 30 m elevation and slope for a point or small area, returning the scene IDs, units, and caveats behind every number. Built-in sanity checks and a verify tool recompute key values through an independent path so the assistant can flag cloudy, low-confidence, or mismatched results instead of guessing.6MIT