Tools (5)

list_subjects
browse_subject
get_table_metadata
get_data
report_feedback

list_subjects

List SCB's top-level statistical subject categories.

SCB (Statistics Sweden) publishes thousands of tables covering population,
labour market, national accounts, trade, prices, environment, education,
democracy, and more. This tool lists the main subject areas.

Use the returned IDs with browse_subject to navigate deeper.

Args:
lang: Language — 'en' for English, 'sv' for Swedish (default 'en').

Returns:
List of subject categories with id, text, and type (folder or table).

Parameter	Type	Required	Description
`lang`	`string`	optional	Language — 'en' for English, 'sv' for Swedish. (default: "en")

JSON-RPC Request

{
  "jsonrpc": "2.0",
  "method": "tools/call",
  "id": 1,
  "params": {
    "name": "list_subjects",
    "arguments": {}
  }
}

browse_subject

Browse a subject path to find sub-categories and tables.

Navigate the SCB table hierarchy by providing a path from list_subjects
or a previous browse_subject call. Returns child folders and tables.

Common top-level codes: AM (Labour market), BE (Population), BO (Housing),
EN (Energy), FM (Financial markets), HA (Trade), HE (Household finances),
HS (Health), JO (Agriculture), MI (Environment), NR (National accounts),
NV (Business activities), PR (Prices), TK (Transport), UF (Education).

When you reach a table (type='table'), use get_table_metadata with the
full path to see its dimensions and variables.

Args:
path: Subject path (e.g. 'BE', 'BE/BE0101', 'AM/AM0401').
lang: Language — 'en' or 'sv'.

Returns:
List of items at this path — folders to navigate deeper, or tables to query.

Parameter	Type	Required	Description
`path`	`string`	required	Subject path (e.g. 'BE', 'BE/BE0101', 'AM/AM0401').
`lang`	`string`	optional	Language — 'en' or 'sv'. (default: "en")

JSON-RPC Request

{
  "jsonrpc": "2.0",
  "method": "tools/call",
  "id": 1,
  "params": {
    "name": "browse_subject",
    "arguments": {
      "path": "example"
    }
  }
}

get_table_metadata

Get the structure of an SCB table — its dimensions, variables, and valid values.

Use this before querying data to understand what filters are available.
Each variable lists its valid codes and labels. Variables marked 'elimination: true'
can be omitted from queries to get aggregated totals.

Time variables are flagged with 'time: true'. Time codes follow patterns:
'2024' (annual), '2024K3' (quarterly), '2024M06' (monthly).

Args:
table_path: Full path to the table (e.g. 'BE/BE0101/BE0101A/BefolkManadCKM').
lang: Language — 'en' or 'sv'.

Returns:
Table title and list of variables with codes, labels, and valid values.

Parameter	Type	Required	Description
`table_path`	`string`	required	Full path to the table (e.g. 'BE/BE0101/BE0101A/BefolkManadCKM').
`lang`	`string`	optional	Language — 'en' or 'sv'. (default: "en")

JSON-RPC Request

{
  "jsonrpc": "2.0",
  "method": "tools/call",
  "id": 1,
  "params": {
    "name": "get_table_metadata",
    "arguments": {
      "table_path": "example"
    }
  }
}

get_data

Query data from an SCB statistical table.

Without filters, returns the latest top_n time periods for all dimensions.
With filters, you can select specific dimension values.

Each filter dict needs: 'code' (variable code from metadata), 'filter' type,
and 'values' list. Filter types:
- 'item': select specific codes, e.g. {"code": "Region", "filter": "item", "values": ["00"]}
- 'top': last N periods, e.g. {"code": "Tid", "filter": "top", "values": ["5"]}
- 'all': all values, e.g. {"code": "Kon", "filter": "all", "values": ["*"]}

Example — Sweden total population, last 5 months:
table_path='BE/BE0101/BE0101A/BefolkManadCKM'
filters=[
{"code": "Region", "filter": "item", "values": ["00"]},
{"code": "Alder", "filter": "item", "values": ["tot"]},
{"code": "Kon", "filter": "item", "values": ["1+2"]},
{"code": "ContentsCode", "filter": "item", "values": ["BE0101N1"]},
{"code": "Tid", "filter": "top", "values": ["5"]}
]

Args:
table_path: Full path to the SCB table.
filters: List of dimension filters. See get_table_metadata for valid codes.
top_n: If no filters given, fetch this many latest time periods (default 5).
lang: Language — 'en' or 'sv'.

Returns:
Parsed data with metadata (label, source, updated) and a list of records.
Each record has labeled dimension values and a 'value' field.

Parameter	Type	Required	Description
`table_path`	`string`	required	Full path to the SCB table.
`filters`	`any`	optional	List of dimension filters. See get_table_metadata for valid codes.
`top_n`	`integer`	optional	If no filters given, fetch this many latest time periods (default 5). (default: 5)
`lang`	`string`	optional	Language — 'en' or 'sv'. (default: "en")

JSON-RPC Request

{
  "jsonrpc": "2.0",
  "method": "tools/call",
  "id": 1,
  "params": {
    "name": "get_data",
    "arguments": {
      "table_path": "example"
    }
  }
}

report_feedback

Report a bug, feature request, or general feedback for this data source.

Use this when something doesn't work as expected, when you'd like
a new feature, or when you have suggestions for improvement.

Args:
feedback: Describe the issue or suggestion.
feedback_type: One of 'bug', 'feature_request', or 'general'.

Parameter	Type	Required	Description
`feedback`	`string`	required
`feedback_type`	`string`	optional	(default: "general")

JSON-RPC Request

{
  "jsonrpc": "2.0",
  "method": "tools/call",
  "id": 1,
  "params": {
    "name": "report_feedback",
    "arguments": {
      "feedback": "example"
    }
  }
}

Quick Start

Shell

curl -X POST "https://context.gnist.ai/mcp/scb/" \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -H "Gnist-API-Key: YOUR_API_KEY" \
  -d '{"jsonrpc": "2.0", "method": "tools/call", "id": 1, "params": {"name": "browse_subject", "arguments": {"path": "example"}}}'

Python

import httpx

resp = httpx.post(
    "https://context.gnist.ai/mcp/scb/",
    headers={"Gnist-API-Key": "YOUR_API_KEY", "Content-Type": "application/json"},
    json={
  "jsonrpc": "2.0",
  "method": "tools/call",
  "id": 1,
  "params": {
    "name": "browse_subject",
    "arguments": {
      "path": "example"
    }
  }
},
)
print(resp.json())