gov-data

name	gov-data
description	Query the data.gov.uk CKAN catalogue — the UK government's central index of ~58,000 published datasets aggregated from every central department, most local authorities, and statutory bodies. Wraps CKAN's full-text search, dataset detail, publishing organisations, tags, and topic groups. Use when the question is about whether a department has published structured data on a topic, who publishes a given dataset, or what data is newly available. The catalogue is OGL v3.0; individual dataset licences vary and should be checked per-record.
license	Open Government Licence v3.0 (catalogue) — individual datasets vary
metadata	{"facility":"gov-data","cli-alias":"gov-data","base-url":"https://data.gov.uk/api/3/action","provenance":{"tier":3,"operator":"Government Digital Service (Cabinet Office)","service":"data.gov.uk","upstream-data":"Dataset metadata harvested from ~600 publishing organisations (central departments, local authorities, NDPBs, statutory bodies); dataset content is hosted by the publisher, not by data.gov.uk","citation-short":"via data.gov.uk","citation-formal":"data.gov.uk CKAN catalogue, retrieved {date}, OGL v3.0 (catalogue metadata)","confidence":"derived","confidence-notes":"Catalogue is authoritative for *what is published*; the catalogue record points to the publisher's hosted file. For numeric or substantive claims, follow the resource link to the source publication and cite that. Many older records carry broken resource URLs or stale schemas; treat 'last modified' as a freshness signal."}}

data.gov.uk

Base URL: https://data.gov.uk/api/3/action

data.gov.uk is operated by the Government Digital Service (Cabinet Office). It runs the standard CKAN API: ~58,000 dataset records harvested from every UK central department, ~4,000+ local authorities, and a long tail of statutory bodies (HSE, NICE, ONS, Met Office, DVLA, …).

CKAN actions wrapped

Action	Wrap	What
package_search	search	Full-text + facet search across all datasets
package_show	dataset	One dataset's full record (resources, formats, sizes)
organization_list / _show	organisations / organisation	Publishing bodies
tag_list	tags	CKAN tag vocabulary
group_list	groups	Higher-level topic groups
(helper)	recent	Recently-modified datasets (sort=metadata_modified desc)

CLI

parl gov-data search --query "spend over 500" --rows 10        # Solr full-text
parl gov-data search --fq "organization:hm-revenue-customs" --rows 50
parl gov-data dataset "uk-trade-info"                          # one dataset
parl gov-data organisations --all-fields true --take 50
parl gov-data organisation "ministry-of-justice" --include-datasets true
parl gov-data tags --query "transparency"
parl gov-data groups
parl gov-data recent --rows 20                                 # newest changes

search accepts both --query (CKAN q) and --fq (Solr filter query). Useful filters:

• organization:<slug> — datasets owned by one publisher
• groups:<slug> — datasets in one topic group
• tags:<tag> — datasets carrying one tag
• res_format:CSV — datasets with at least one CSV resource
• metadata_modified:[2025-01-01T00:00:00Z TO *] — modified since a date (Solr range syntax)

Joins to Parliament

• Transparency disclosure → committee scrutiny: "spend over £25,000" / "spend over £500" datasets are published monthly by every central department. Cross-reference with committees and written-questions-and-statements when MPs ask about specific contracts.
• Departmental publication → Estimate / Bill: datasets published by HM Treasury, DWP, MoJ etc. underpin spending rounds — pair with bills (Supply and Appropriation Bills).
• Local-authority data: data.gov.uk indexes ~4,275 council datasets. For Parliament-adjacent constituency-level analysis, pair with ons-nomis (Census) and mysoc-mapit (postcode → LAD → constituency).
• Election results & boundaries: many electoral datasets live here as resources; the canonical sources are dc-elections for election metadata and ons-geo for boundaries.

Caveats

• CKAN is a catalogue, not a hosting service. Each dataset record contains resources[]; each resource has a url pointing at the publisher's own server. Resource URLs go stale — check resources[].mimetype and resources[].url before citing as live.
• Solr behind q parameter. Full Solr query syntax works: fielded search, ranges, boosts. Use fq (filter query) for facet filters that should not affect relevance ranking.
• Mixed licences. The catalogue metadata is OGL v3.0; the datasets themselves vary — most central-government data is OGL, but some local-authority data uses CC-BY, ODbL, or bespoke terms. Check license_id / license_title per record.
• Old records can be stale. Some publisher organisations have moved on; their datasets sit in CKAN with a state: active flag even though the underlying URL 404s. Treat metadata_modified as a freshness signal.

Provenance to cite

Tier 3 — third-party (Government Digital Service / Cabinet Office); catalogue is derived metadata, dataset content is upstream of the publisher.

• Inline cite: "(via data.gov.uk)" — once per paragraph.
• For substantive claims using the content of a dataset, follow the resource link and cite the publisher's own publication: "Cabinet Office spend over £25,000, October 2025, via data.gov.uk".
• Check each dataset's license_id. OGL v3.0 is the default for central-government data but not universal.
• See ../../docs/provenance.md.