mcp-outremer
# mcp-outremer
MCP server for French overseas territories public-data discovery and official source navigation.
## Scope
French overseas territories data discovery across national portals and local open-data portals.
## Tools
- `outremer_get_sources`
- `outremer_search_data_gouv`
- `outremer_get_dataset`
- `outremer_fetch_source_excerpt`
- `outremer_explain_scope`
- `outremer_list_reference_items`
- `outremer_list_territories`
- `outremer_get_territory`
- `outremer_list_communes`
- `outremer_search_territory_datasets`
- `outremer_compare_departments`
## Install
```bash
npm install
npm run build
npm test
npm run dev
```
## Claude Desktop
```json
{
"mcpServers": {
"outremer": {
"command": "npx",
"args": ["mcp-outremer"]
}
}
}
```
## Sources
- data.gouv.fr: https://www.data.gouv.fr/
- data.gouv.fr API reference: https://doc.data.gouv.fr/api/reference/
- geo.api.gouv.fr: https://geo.api.gouv.fr/
- La Réunion open data: https://data.regionreunion.com/
- New Caledonia open data: https://data.gouv.nc/
- Météo-France Outre-mer: https://meteofrance.re/
## Example Prompts
- "Search national datasets about overseas territories."
- "Find official data portals for La Réunion, Mayotte, and New Caledonia."
- "Explain which national APIs can resolve overseas commune codes."
## 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
Scored across 10 tools
Most tools have clear, distinct purposes: listing vs. fetching territories, sources vs. excerpts, and global vs. territory-scoped dataset search. However, potential confusion exists between search_data_gouv and search_territory_datasets, and get_territory also returns 'search hints' which could overlap.
All tools share the 'outremer_' prefix and follow a consistent verb_noun snake_case pattern (list_, get_, search_, fetch_, explain_). There are no mixed conventions or unexpected casing.
With exactly 10 tools covering territory lookup, dataset search, source management, and reference utilities, the count feels well-scoped for a French overseas territories data assistant. It is within the ideal 3–15 range and each tool addresses a distinct need.
The server covers the core workflows of discovering territories, searching and inspecting datasets, listing reference sources, and fetching source excerpts. Minor gaps include no tool for full source content retrieval (only excerpts) and no direct commune search beyond department codes, but these are reasonable limitations for a focused utility.