Skip to main content
Glama
README.md
# mcp-mayotte

MCP server for Mayotte public data, vigilance, administrative references, and local risk discovery.

## Scope

Mayotte-focused public-data discovery: communes, vigilance, risks, population, administrative and open-data sources.

## Tools

- `mayotte_get_sources`
- `mayotte_search_data_gouv`
- `mayotte_get_dataset`
- `mayotte_fetch_source_excerpt`
- `mayotte_explain_scope`
- `mayotte_list_reference_items`
- `mayotte_list_communes`
- `mayotte_find_commune`
- `mayotte_get_commune_risks`
- `mayotte_search_datasets`
- `mayotte_get_vigilance`

## Install

```bash
npm install
npm run build
npm test
npm run dev
```

## Claude Desktop

```json
{
  "mcpServers": {
    "mayotte": {
      "command": "npx",
      "args": ["mcp-mayotte"]
    }
  }
}
```

## Sources

- data.gouv.fr Mayotte search: https://www.data.gouv.fr/fr/datasets/?q=Mayotte
- Météo-France vigilance Mayotte: https://vigilance.meteofrance.fr/fr/mayotte
- geo.api.gouv.fr: https://geo.api.gouv.fr/
- Géorisques: https://www.georisques.gouv.fr/
- INSEE Mayotte: https://www.insee.fr/fr/statistiques?debut=0&theme=1&geo=DEP-976

## Example Prompts

- "List official Mayotte source links."
- "Search data.gouv.fr for Mayotte school or health datasets."
- "Show Mayotte commune reference hints."

## Safety

This MCP helps agents discover and summarize public sources. It is not an official authority. For emergency, legal, or administrative decisions, follow the competent public service.

## Glama / Docker

The repo includes `Dockerfile` and `glama.json`.

Publishing notes: [`docs/publishing.md`](docs/publishing.md).

Build steps:

```json
["npm install", "npm run build"]
```

CMD arguments:

```json
["node", "dist/index.js"]
```

## License

MIT

TDQS

A3.7/5.0

Scored across 10 tools

Disambiguation3/5

mayotte_search_data_gouv and mayotte_search_datasets both target the same data.gouv.fr dataset search and could be confused, and list_reference_items is vague next to get_sources. Otherwise the tools separate fairly clearly into dataset, commune, source, and vigilance functions.

Naming Consistency5/5

All tools consistently use the mayotte_ prefix with lowercase snake_case and clear verb-object combinations such as list_communes, find_commune, get_vigilance, and fetch_source_excerpt. The naming pattern is uniform and predictable.

Tool Count5/5

Ten tools is well within the ideal range for a regional information server. Each tool contributes to a distinct area such as datasets, communes, sources, or vigilance without unnecessary bloat.

Completeness4/5

The tool surface covers the core read-only workflows: searching and retrieving datasets, looking up communes, listing curated sources, and getting vigilance information. Minor gaps exist, such as no dedicated commune detail tool and some vagueness around reference items, but agents can work around them.

Maintenance

ActivityMaintained
ResponsivenessNo issues