GOV.UK Guidance to Markdown with Citations

AI agents asked about UK rules too often answer from memory. This Actor gives them the current GOV.UK page instead, as clean Markdown with what a citation needs: title, publishing organisation, last-updated date, change history, withdrawn status and a ready-made citation line. It reads GOV.UK's official Content API 1, which anyone may use for any purpose with no key 1, so there's no scraping and nothing to break when the page design changes.

What you get for each page

• Markdown:
  • headings, lists, tables (as Markdown tables), and callout boxes as quotes;
  • links made absolute, so they work outside GOV.UK;
  • no navigation, cookie banners or feedback forms.
• Citation: Title. Organisation, GOV.UK. Last updated 10 October 2024. URL (accessed 3 October 2026). Dates are in UK time, as GOV.UK shows them.
• Status:
  • ok;
  • withdrawn, with the date and GOV.UK's explanation (often a link to what replaced it);
  • not-found (404) or gone (410) 2;
  • redirected-elsewhere;
  • invalid: not a www.gov.uk page, since subdomains and attachments aren't in the API 1.
• Redirects: GOV.UK's own redirects are followed (exact and prefix rules), and the old URLs are listed.
• Guides: a link to one part of a multi-page guide returns just that part, or every part with wholeGuide.
• Also: attachments (HTML and PDF links), the change history, description, documentType and locale.
• Long pages: maxCharacters cuts at a paragraph break and says so.
• Rate limit: requests stay under the API's limit of 10 a second 1.
• Licence: the OUTPUT record carries the Open Government Licence v3.0 attribution statement 3 to include when you reuse the text.

Checking:

• Unit tests replay real Content API responses recorded on 3 October 2026: an answer, a guide part, a redirect, a withdrawn publication with attachments, and a 404.
• tests/check_live.py compares the Markdown with the pages as people see them on www.gov.uk, using Python only. On 3 October 2026, 7 pages agreed:
  • 97% to 100% of the Markdown's 5-word sequences were on the rendered page;
  • titles matched the page headings;
  • "Last updated" dates matched where the page shows one.

Input

{
  "urls": [
    "https://www.gov.uk/vat-rates",
    "https://www.gov.uk/register-for-vat/how-register-for-vat",
    "https://www.gov.uk/vat-registration",
    "https://www.gov.uk/government/publications/covid-19-stay-at-home-guidance"
  ],
  "maxCharacters": 3000
}

Example output

One dataset row per page:

{
  "input": "https://www.gov.uk/vat-rates",
  "url": "https://www.gov.uk/vat-rates",
  "status": "ok",
  "title": "VAT rates",
  "partTitle": null,
  "description": "Current VAT rates - standard 20% and rates for reduced rate and zero-rated items",
  "documentType": "answer",
  "organisations": [
    "HM Revenue & Customs"
  ],
  "firstPublished": "2011-09-12",
  "lastUpdated": "2014-12-12",
  "withdrawn": false,
  "withdrawnAt": null,
  "withdrawnExplanation": null,
  "locale": "en",
  "redirectedFrom": [],
  "markdown": "The standard VAT rate is *20%*\n\n> This guide is also available [in Welsh (Cymraeg)](https://www.gov.uk/cyfraddau-taw).\n\n## VAT rates for goods and services\n\n|  | % of VAT | What the rate applies to |\n| --- | --- | --- |\n| Standard rate | 20% | Most goods and services |\n| Reduced rate | 5% | Some goods and services, for example, children’s car seats and home energy |\n| Zero rate | 0% | Zero-rated goods and services, for example, most food and children’s clothes |\n\nThe standard rate of VAT increased to 20% on 4 January 2011 (from 17.5%).\n\nSome things are exempt from VAT, such as postage stamps, financial and property transactions.\n\nThe [VAT rate businesses charge](https://www.gov.uk/vat-businesses/vat-rates) depends on their goods and services.\n\nCheck the [rates of VAT](https://www.gov.uk/rates-of-vat-on-different-goods-and-services) on different goods and services.",
  "characters": 876,
  "attachments": [],
  "changeHistory": [],
  "citation": "VAT rates. HM Revenue & Customs, GOV.UK. Last updated 12 December 2014. https://www.gov.uk/vat-rates (accessed 3 October 2026).",
  "note": null,
  "error": null
}

Integrations

Ready-made files in integrations/README.md: a Microsoft Power Platform custom connector, OpenAPI 3 (AI agents, n8n, Postman), an Excel / Power BI Power Query script, a Google Sheets function, and curl, Python and JavaScript examples.

Pricing

Pay per event: $2 per 1,000 results ($0.002 each), plus Apify's minimal start fee. A result is one row in the dataset: one GOV.UK page fetched. The run stops at your spending limit.

Sources

1. GOV.UK Content API: about (anyone may use it for any purpose, no authentication, 10 requests a second per client, www.gov.uk pages only). Accessed 3 October 2026.
2. GOV.UK Content API: reference (GET /api/content/{path}, 404 and 410 responses, withdrawn_notice, public_updated_at). Accessed 3 October 2026.
3. Open Government Licence v3.0 (attribution statement). Accessed 3 October 2026.
4. Turndown (HTML to Markdown, MIT) with turndown-plugin-gfm for tables. Accessed 3 October 2026.