SB OGC MCP
Official# SB OGC MCP — Studio Bereikbaar Mobility Data for Claude
MCP server that gives Claude access to Dutch mobility data via Studio Bereikbaar's OGC API.
## What you get
13 tools for Claude Desktop:
| Tool | What it does |
|------|-------------|
| `list_collections` | List 15 data collections (boundaries, NRM networks, demographics) |
| `describe_collection` | Get schema/properties of a collection |
| `list_processes` | List 7 analysis processes |
| `get_features` | Fetch GeoJSON features with filters |
| `search_boundaries` | Find gemeente/wijk/buurt/pc4 by name |
| `run_odin_query` | ODIN travel survey analysis (20 years, modal split, trends) |
| `run_odin_compare` | Compare two locations side-by-side |
| `run_modal_split` | Generate modal split chart (returns PNG image) |
| `run_odin_spider` | Origin-destination desire lines (GeoJSON) |
| `run_odin_profile` | 7-cluster mobility profiling |
| `run_odin_spider_profile` | Demographics of travelers to/from a municipality |
| `run_accessibility_map` | BOP accessibility map (returns PNG image) |
## Setup voor collega's
### 1. Installeer uv (eenmalig)
```bash
curl -LsSf https://astral.sh/uv/install.sh | sh
```
### 2. Configureer Claude Desktop
Open `~/Library/Application Support/Claude/claude_desktop_config.json` en voeg toe:
```json
{
"mcpServers": {
"sb-ogc-api": {
"command": "uvx",
"args": ["sb-ogc-mcp"]
}
}
}
```
### 3. Herstart Claude Desktop
De tools verschijnen automatisch. Probeer:
> "Welke data collections heeft Studio Bereikbaar?"
> "Laat de modal split zien voor Amsterdam"
> "Vergelijk Rotterdam en Utrecht qua vervoerswijzen"
> "Toon de bereikbaarheidskaart voor fietsen naar supermarkten in Den Haag"
## Data
All data comes from `tools.studiobereikbaar.nl/oapi` (OGC API):
- **ODIN** — CBS travel survey, 20 years (2004–2023), ~3M trips, enriched by Roland Kager
- **NRM** — National traffic model networks (2022 + 4 future scenarios)
- **CBS boundaries** — gemeente, wijk, buurt, pc4, provincie
- **BOP** — accessibility maps (Move Mobility)
- **V500** — 500m population + urbanization grids
TDQS
Scored across 16 tools
Each tool targets a distinct operation: collection metadata, data retrieval, thematic mapping, accessibility, modal split, ODIN queries, profiles, spider diagrams, and boundary search. No two tools have overlapping purposes; descriptions clearly differentiate their outputs and inputs.
All tools follow a verb_noun pattern (e.g., describe_collection, generate_thematic_map, run_odin_query). The verbs are appropriate for the actions (list, describe, get, run, search), and the pattern is consistent across the entire set.
16 tools cover the necessary functionality for a geospatial analysis server without being excessive. The count is well-scoped for discovery, data access, and multiple analysis types.
The tool set provides a full lifecycle: discover collections/processes, retrieve features, generate thematic maps, and run various analyses (accessibility, modal split, ODIN surveys, spider diagrams). There are no obvious gaps for the intended domain.