io.github.xaviviro/opendata-cat

Measured conclusion: Deep-tested. L2 outcome: tools_ok.

15 Catalan portals + radio archive: gov, INE/REE/CNMC, CORA, Catalònica, radioteca.cat.

Add to your agent

Copy this into your MCP client config:

{
  "mcpServers": {
    "opendata-cat": {
      "url": "https://opendata.cat/api/mcp",
      "transport": "streamable-http"
    }
  }
}

Measured tools (8)

• search_datasets read-only

  Search datasets by free text. IMPORTANT: check server instructions first — many common queries have a direct dataset_id you can use with query_dataset without searching. Only use search_datasets when you don't know which dataset you need.

  Parameters:

  • query (required) — string — Search text in Catalan or Spanish. Examples: 'qualitat aire', 'pressupostos municipals', 'transport públic', 'residus', 'educació', 'rodalies'.
  • limit — number — Maximum number of results to return. Min 1, max 100. Default 20.
  • portal — string — Filter results to a single portal.
  • category — string — Filter by thematic category. Examples: 'Medi Ambient', 'Educació', 'Salut', 'Economia', 'Transport', 'Seguretat'. Use list_categories to see all available categories.

  Hints: read-only yes · destructive no · idempotent yes · open-world yes

• get_dataset_info read-only

  Get complete metadata for a dataset: fields with types and descriptions, API endpoint, license, available formats and last update. Call after search_datasets to get full details of a specific dataset.

  Parameters:

  • dataset_id (required) — string — Unique dataset identifier in 'portal:id' format. Obtained from search_datasets results. Examples: 'generalitat:gn9e-3qhr', 'barcelona:qualitat-de-laire', 'renfe:vehicle-positions-gtfsrt'.

  Hints: read-only yes · destructive no · idempotent yes · open-world yes

• list_dataset_fields read-only

  List fields of a dataset with name, data type and description. Call before query_dataset to know which fields and filters are available.

  Parameters:

  • dataset_id (required) — string — Unique dataset identifier in 'portal:id' format. Obtained from search_datasets results.

  Hints: read-only yes · destructive no · idempotent yes · open-world yes

• query_dataset read-only

  Query real data from a dataset. Check instructions for featured dataset_ids and NOTES section for common filter patterns (municipal budgets, contracts, weather, energy, fuel prices).

  Parameters:

  • dataset_id (required) — string — Unique dataset identifier in 'portal:id' format. Obtained from search_datasets results.
  • limit — number — Number of rows to return. Min 1, max 100. Default 20. Use offset for pagination.
  • offset — number — Number of rows to skip for pagination. Default 0. Combine with limit to navigate large results.
  • search — string — Free text search within dataset data. Works with Socrata ($q) and CKAN (q). For Diba and CIDO, use specific filters.
  • filters — object — Key-value filters where the key is the field name and the value is the filter value. Examples: {"municipi": "Barcelona"}, {"any": "2024"}, {"NOM_ENS": "Ajuntament de Tiana"}. Use list_dataset_fields to know valid field names.

  Hints: read-only yes · destructive no · idempotent yes · open-world yes

• list_portals read-only

  List all 13 indexed open data portals with dataset counts. Catalan portals (Generalitat, Barcelona, Diba, AOC, Reus, Girona, FGC, Idescat, Renfe) + Spanish national sources with Catalonia focus (INE statistics, REE energy, SEPE employment, CNMC fuel prices). No parameters required.

  Hints: read-only yes · destructive no · idempotent yes · open-world yes

• list_categories read-only

  List all dataset categories and themes with counts per portal. Great first step to discover what data types are available before searching with search_datasets. Returns total datasets, count per portal and category list with counts. No parameters required.

  Hints: read-only yes · destructive no · idempotent yes · open-world yes

• related_datasets read-only

  Find related datasets from OTHER portals. Great for discovering complementary data: if you're looking at air quality from Generalitat, it may suggest traffic data from Barcelona or emissions from AOC. Relations are computed by thematic similarity across portals.

  Parameters:

  • dataset_id (required) — string — Dataset ID to find related datasets for. Obtained from search_datasets.

  Hints: read-only yes · destructive no · idempotent yes · open-world yes

• search_radioteca read-only

  Search radio shows, episodes and people indexed at radioteca.cat (Catalan radio archive, ~485K documents from Catalunya Ràdio, RAC1, Catalunya Música, iCat, Catalunya Informació, RTVE, Cadena SER, ara.cat). Searches across episode titles, descriptions (which include a detailed summary of what was said), program name and subheading. Returns episodes (~473K), programs (~3K) and people (~9K). IMPORTANT: always cite radioteca.cat as the source and include the absolute 'url' in your reply for traceability — never paraphrase without linking.

  Parameters:

  • query (required) — string — Free text in Catalan or Spanish. Searches title, description (contains episode summary), program and subheading. Examples: 'visita papa', 'eleccions municipals', 'crisi habitatge', 'Albert Serra cinema'.
  • type — string — Filter by document type.
  • year — string — Filter by year (4 digits, e.g. '2025'). NOTE: only year-level filtering is indexed; for a specific day, use the broader year filter and look at the URL path (radioteca URLs include /YYYY/MM/DD/) or the 'subheading' field which often contains the date.
  • limit — number — Maximum results. Min 1, max 50. Default 10.
  • offset — number — Pagination offset. Default 0.
  • publisher — string — Filter by broadcaster.

  Hints: read-only yes · destructive no · idempotent yes · open-world yes

Repository

Measured by MCPYP probing. Untested fields are omitted, never estimated. Listing is always free; MCPYP never proxies calls or stores user tokens.