About
Status and self-checks
New This page is new since 1.2.0-rc.1 (2026-11-06). Everything that changed
What this portal checks about itself on every build, and what it found. Automation reports; a person decides and acts.
The five routines
Self-verification
In order
26 site checks, 19 browser checks with deliberate breaks, 127 unit tests and the LinkML check run before every publish. A check is trusted only once its deliberate break is caught.
Self-assessment
In order
12 requirements met, 12 partly; 2 of 4 release gates met; 19 of 19 repository files in place.
Staleness and self-updating
In order
Review date, organisation list, change-flag baseline and watched sources, each against its own limit.
Self-reporting
Needs attention
Latest health run: 3 problem(s) for a person, 2026-11-18T13:50:53+00:00.
Self-repair
In order
40 defects found and fixed, each with a check that stops it coming back. Automation reports and rebuilds; only a person changes content, settings or the open register (20 items).
Self-verification
Every published build passes all of these first. Each was written in the same change as the defect it prevents, and each has a deliberate break that proves it can fail.
| Check | What it makes sure of |
|---|---|
| C-01 | every page sets a Content Security Policy with no unsafe sources. |
| C-02 | no script, stylesheet, image or font loaded from another host (E2: self-contained pages). |
| C-03 | no email address anywhere in the published files (COL-02), metadata and data included. |
| C-04 | ids are unique on every page. |
| C-05 | every internal link resolves, to a file and, if it has one, an anchor on that page. |
| C-06 | every table is captioned, has scoped headers and sits in a labelled, focusable scroll region. |
| C-07 | every page describes itself: artefact id, version, class, as-of, review date, and JSON-LD. |
| C-08 | every page carries the colophon: disclaimers, licence, version, QR code, barcode and the contact form. |
| C-09 | the first focusable element on every page is the skip link to main. |
| C-10 | no deck or film files in the repository or the site, unless the owner says so for that piece. |
| C-11 | no inline executable scripts and no inline event handlers, so the Content Security Policy can forbid them. |
| C-12 | every page shows the status banner: proposed, not endorsed, not an official service. |
| C-13 | every page has lang="en-GB" and a unique title. |
| C-14 | no page over 1,500 KB with what it loads from this site (envelope: median page 1,500 KB). |
| C-15 | every meaningful image has an accessible name; decorative icons are hidden. |
| C-16 | no unrendered template placeholders, such as [layers] or [prov("K")], in page text (D-007); inside <pre>, the |
| C-18 | llms.txt and llms-full.txt exist, every page is listed in llms.txt, and every link in it resolves to a built file. |
| C-19 | no style attributes in any page: the CSP (style-src 'self') blocks them, so they silently do nothing (D-018). |
| C-20 | every guided tour stop points at something on its page, so no tour silently skips a step. |
| C-21 | every page offers "Suggest a change to this page", pointing at an issue form that exists and naming that page. |
| C-22 | no protective markings or handling caveats anywhere public: pages, downloads and the requirements file (ADR 0015, RQ-18). |
| C-23 | no old directorate name and no personal or team details anywhere: pages, downloads or repository files. Contact is by form only. |
| C-24 | change flags agree with downloads/changes-since.json: each new or updated page says so under its title and in the global bar, and each changed section's heading is flagged (the Data Quality Commons pattern). |
| C-17 | no email address in the repository's own files either; test fixtures may use reserved example domains only. |
| C-25 | vendor connectors agree with downloads/connectors/vendors.json: each mapped product has its section, crosswalk CSV and add-on kit, every kit names each property the profile file lists for a dataset, and no repository is cited without a commit (ADR 0021). |
| C-26 | the glossary A to Z links exactly the letters that have a term, the namespace register lists every prefix in the profile as plain code (never a link, D-037), and the weaver offers open-a-file and swap. |
Browser checks run every page in light and dark at six widths, plus the tools and the no-script fallback. Their deliberate breaks: contrast, overflow, label, console, skip, validator, csp, bulk, menus, menuedge, current, fold, moreview, submit, meter, nested, suggest, map, extension.
Self-assessment
| Status | Requirements |
|---|---|
| Met | 12 |
| Partly | 12 |
| Planned | 3 |
| Controlled | 2 |
| Gate | Status | Evidence |
|---|---|---|
| Reuse assessed for every new term | partly | Each property records whether it is reused, changed or new against the xGov MEM and DCAT; not yet a blocking check. |
| Every rule traced to a test | met | tests/test_profile.py fails the build if a rule has no deliberate break. |
| No open serious semantic-loss or disclosure defect | partly | The defect register is complete and C-22 blocks protective markings; semantic-loss defects are not yet graded by severity. |
| Compatibility claimed only with evidence | met | Conformance levels come from the kit; no allied or controlled compatibility is claimed (RQ-15, RQ-29). |
| File | In place |
|---|---|
README.md | Yes |
LICENSE | Yes |
SECURITY.md | Yes |
CONTRIBUTING.md | Yes |
CODE_OF_CONDUCT.md | Yes |
SUPPORT.md | Yes |
CITATION.cff | Yes |
CHANGELOG.md | Yes |
THIRD-PARTY-NOTICES.md | Yes |
.github/CODEOWNERS | Yes |
.github/PULL_REQUEST_TEMPLATE.md | Yes |
.github/dependabot.yml | Yes |
.github/ISSUE_TEMPLATE/contact.yml | Yes |
.github/ISSUE_TEMPLATE/defect.yml | Yes |
.github/ISSUE_TEMPLATE/change-request.yml | Yes |
.github/workflows/pages.yml | Yes |
.github/workflows/codeql.yml | Yes |
docs/registers.md | Yes |
content/section-baseline.json | Yes |
Staleness and self-updating
Each source of change has its own limit. Reaching it is reported, not fixed: refreshing needs a person (for example tools/import_orgs.py --govuk, tools/health.py --record or tools/section_baseline.py).
| What | As of | Days | Meaning | State |
|---|---|---|---|---|
| Review date | 2027-02-06 | 92 | days until the content must be reviewed | In date |
| Organisation list (GOV.UK) | 2026-11-09 | -3 | days since it was read (limit 120) | In date |
| Change-flag baseline | 2026-11-06 | 0 | days since 1.2.0-rc.1 baseline (reset at each release) | In date |
| Verified sources watched | 14 of 14 | 0 | not yet fingerprinted for the weekly run | In date |
Health run
Run 2026-11-18T13:50:53+00:00. It checks that verified sources are reachable and unchanged, that the review date is not close and that pages stay inside their weight envelope.
Problems
- Source changed: National Data Library: progress update, January 2026 (https://www.gov.uk/government/publications/national-data-library-progress-update-january-2026). Re-verify what the profile relies on, then run health.py --record.
- Source changed: UK GEMINI (https://www.agi.org.uk/uk-gemini/). Re-verify what the profile relies on, then run health.py --record.
- Source changed: data.gov.uk topics (https://www.data.gov.uk/search). Re-verify what the profile relies on, then run health.py --record.
Self-repair: defects and their checks
Automation never pushes, deletes or changes settings. It rebuilds what is derived and reports the rest. Every defect found has a root cause and a check, listed in the defect register.
| Defect | Found | What | Check |
|---|---|---|---|
| D-040 | 2026-11-15 | Code scanning found five weaknesses. The reference HTTP service returned internal error text to callers (posting null to /weave answered with a JavaScript TypeError message). The Copy buttons put back their label by re-parsing HTML kept in an attribute. A test treated any address containing github.com as GitHub. Two of the build's own HTML-stripping patterns missed upper-case or spaced closing tags. | tests/test_integrations.py ReferenceAPI asserts a known message is returned and an internal failure is generic; it fails against the previous service |
| D-039 | 2026-11-15 | Two defects were numbered D-037, and the build on main failed: two pull requests prepared in parallel each took the next free number, and merging them kept two copies of the knowledge page's Taxonomy heading and tab strip (one the linter had changed), so the page had duplicate ids (C-04) and the linter failed. | tests/test_check_site.py checks that every defect, open item and closed item in content/portal.yaml has a unique id |
| D-038 | 2026-11-15 | The Adopt page's Words for contracts box showed the build's unfilled template placeholder (the E of pc expression in braces) instead of the procurement clauses, on the live site, and Copy the clauses copied that text. Found when a Python linter flagged the clauses as built and never used. The same lint pass found 66 other findings: dead code (an old Display options form, unused counts), loop variables not bound in a closure, and test comparisons that would accept a validator returning fewer results than expected. | tests/test_check_site.py plants that placeholder inside pre and fails against the old C-16; tests/test_lint.py keeps both linters in the workflow |
| D-037 | 2026-11-15 | The first draft of the namespace register linked every namespace, though several do not resolve: the StatDCAT-AP namespace contains a literal "(xyz)" and w3id.org/dcat-ap-uk is not registered yet (O-17). | tools/check_site.py C-26 (a namespace cell must not be a link), with a deliberate break |
| D-036 | 2026-11-14 | Where each product runs was written from background knowledge for all 31 products, and Secoda was shown as SaaS only, though Secoda documents a self-hosted, on-premises install. | Site.vendor_problems (deploy list needs provenance; V must cite a pinned repository; open source needs a repository), and tests/test_vendors.py Register with five deliberate breaks plus a Secoda regression test |
| D-035 | 2026-11-07 | The OpenLineage documentation facet written by the weaver named schema version 1-0-1; the specification at the cited commit is 1-1-0. | tools/check_vendor_shapes.py in CI, and tests/test_vendors.py VendorSchemas with ten deliberate breaks |
| D-034 | 2026-11-07 | Commit IDs cited for vendor repositories were padded from seven characters to twelve with invented characters, and a bulk edit then emptied four citations. | tools/check_vendor_repos.py in CI (a padded ID fails it), the 40-character rule in the vendor register checks and check_site C-25, and tests/test_vendors.py |
| D-033 | 2026-11-07 | Invented example data was shown under real names: the synthetic cross-border example gave the DVLA as its publisher, the synthetic base record for the sector extensions gave GDS, and the identifier guidance used company and charity numbers that could belong to real bodies. | tests/test_provenance.py (no synthetic record under a real publisher, no unexplained synthetic content, every sample declared), shown failing with the DVLA put back |
| D-032 | 2026-11-07 | Four of the thirteen verified sources were never fingerprinted, so the weekly health run could not tell if they changed. The new Status page showed it. | tests/test_integrations.py VerifiedSourcesWatched, and the staleness routine on the Status page |
| D-031 | 2026-11-06 | The organisation list still showed the Department for Science, Innovation and Technology as live after the July 2026 machinery of government change, and the Information Commissioner's Office under its old name. | tests/test_integrations.py OrganisationsFresh: read no more than 120 days before the release, and DSIT closed |
| D-030 | 2026-11-06 | A person's name, their team and the old name of their directorate were in the citation file, the profile attribution and the governance text, against the rule of roles, not names, and contact by form only. | tools/check_site.py C-23 (personal details matched by hash, so the check publishes nothing), with its deliberate break |
| D-029 | 2026-11-06 | LinkML's own validator rejected every worked example, so the LinkML file disagreed with the SHACL, JSON Schema and portal validators. | tools/check_linkml.py, with its --break, and linkml-lint in CI |
| D-028 | 2026-11-06 | The standards map could not draw any standard in the products and connectors group: adding the first one, Apache Atlas, broke the build. | tests/test_integrations.py Standards: every standard is drawn on the map |
| D-027 | 2026-11-06 | On a phone with the largest text, opening About inside More showed nothing: its links were below the part of More in view. | tools/check_browser.py menus at six widths, standard and largest text (breaks: moreview, menuedge) |
| D-026 | 2026-11-06 | On About pages, and others whose section was folded into More, the bar did not show which section you were in. | tools/check_browser.py current section at six widths (break: current) |
| D-025 | 2026-11-06 | Pressing Enter in a form with one text field, such as Describe an API from OpenAPI, froze the page. | tools/check_browser.py submit: Enter in three forms, no error and no hang (break: submit) |
| D-024 | 2026-11-06 | Menus in the global bar did not open once the bar had folded them into More, which hid About, Help and others on most laptop and phone screens. On phones, and for About on wide screens, open menus also ran off the edge of the screen. | tools/check_browser.py menus at six widths: every group opens, its first item can be clicked and it stays on screen (breaks: menus, menuedge) |
| D-023 | 2026-11-06 | The labels workflow failed on its first run. | tests/test_integrations.py RepoAsCode, with a deliberate numeric colour |
| D-022 | 2026-11-06 | The contrast check could no longer see the text in page heading bands, so poor contrast there would pass. | tools/check_browser.py contrast (break: contrast, which targets heading band text and was not caught until the fix) |
| D-021 | 2026-11-06 | The quality score bars in the builder and checker were empty. | tools/check_browser.py CSP pass (break: meter) |
| D-020 | 2026-11-06 | The standards map hid its links from assistive technology, and some link names did not match their visible text. | tools/check_browser.py axe nested-interactive and label-content-name-mismatch (break: nested) |
| D-019 | 2026-11-06 | Every page scrolled sideways by 4 pixels on phones and tablets. | tools/check_browser.py overflow at 320, 390 and 768 pixels (break: overflow) |
| D-018 | 2026-11-06 | The barcode, the knowledge graph and several new pages used style attributes, which the content security policy blocks, so they lost their styling. | tools/check_site.py C-19, and a browser pass with the policy on (break: csp) |
| D-017 | 2026-11-06 | Validating inside the metadata weaver crashed on nested records. | tests/test_integrations.py Weaver: no two top-level functions in validator.js share a name |
| D-016 | 2026-11-06 | The adoption page had two elements with the id checklist. | tools/check_site.py C-04 caught it before release |
| D-015 | 2026-11-06 | The record builder’s Turtle dropped every property of a distribution that had its own address. | tests/test_toolkit.py test_turtle_matches_python: browser Turtle and build RDF must be isomorphic for every example |
| D-014 | 2026-11-06 | An unknown property name with a space crashed the Turtle export instead of being reported. | tests/test_toolkit.py test_unknown_key_with_space_serialises |
| D-013 | 2026-11-06 | The build stopped with a type error while checking codelists. | tests/test_toolkit.py test_codelist_code_must_be_text, and Norway must still be on the list |
| D-012 | 2026-11-06 | The overflow check failed on random pages after a browser update. | tools/check_browser.py overflow, with its "overflow" break still caught |
| D-011 | 2026-11-06 | The lineage said 20 xGov MEM issues were open; there were 22. | tests/test_profile.py ReadmeCounts: totals must equal the register, and the README counts must match it |
| D-010 | 2026-11-06 | Escape did not close search when text had been typed. | tools/check_browser.py palette checks |
| D-009 | 2026-11-06 | Loading the critical data asset example into the record builder lost its licence. | tools/check_browser.py: the critical data asset example must conform in the builder |
| D-008 | 2026-11-06 | Small download buttons failed contrast in dark mode. | tools/check_browser.py axe in dark mode, and its "contrast" break |
| D-007 | 2026-11-06 | The strategy and organisations sections showed raw template placeholders (a variable name in braces) instead of content. | tools/check_site.py C-16 (no unrendered placeholders) and its deliberate break |
| D-006 | 2026-11-06 | The caption check accepted a broken caption tag. | tools/check_site.py C-06 deliberate break |
| D-005 | 2026-11-06 | The table check reported header cells without scope on every page. | tools/check_site.py C-06 |
| D-004 | 2026-11-06 | The duplicate-title check missed a real duplicate. | tools/check_site.py C-13 and its deliberate break |
| D-003 | 2026-11-06 | The build library could not be imported in isolated mode and shadowed a standard Python module. | CI runs every script with python3 -I. |
| D-002 | 2026-11-06 | JSON Schema cannot check that an end date is not before a start date. | The fixture is flagged; the test fails if JSON Schema starts catching it, so the flag is removed. |
| D-001 | 2026-11-06 | SHACL accepted badly formed values in recommended and optional properties. | tests/test_profile.py DeliberateBreaks (bad-date, summary-too-long, period-end-before-start) |
Routines and who is in control
Day-to-day upkeep is done by routines; a person keeps control. Each routine has a tier: 0 Observe reads and reports; 1 Maintain changes derived or reference data only after every check passes, in one commit that names the routine, so it is one revert to undo; 2 Propose prepares a pull request or an issue for a person; 3 Person only is never automated. Every Monday the digest routine opens one issue listing what each routine did and what needs a person. To stop every tier 1 routine changing anything, set the repository variable AUTOMATION to off.
| Routine | Tier | When | What it does | What it may change | Workflow |
|---|---|---|---|---|---|
| pages | 1 Maintain | every push, every pull request, Mondays 06:23 UTC, on request | Builds the site, runs every check and deliberate break, runs the health run for the Status page, and publishes main. | The published site only (gh-pages), after every check passes. | pages.yml |
| health-issue | 0 Observe | Mondays, on request | Opens or updates an issue labelled health when a source changed or is unreachable, the review date is near or a page is too heavy; closes it when healthy. | Issues only. | pages.yml |
| release | 3 Person only | a person pushes a version tag, or presses Run on the Release workflow | Builds and checks the tagged version and publishes a GitHub release with the profile files and the site. | A GitHub release. Only a person starts it. | release.yml |
| flags-reset | 1 Maintain | straight after each release | Records the section baseline from the released tag, so change flags show what changed since that release; checks the site, commits content/section-baseline.json and republishes. | content/section-baseline.json. | release.yml |
| upkeep | 1 Maintain | Mondays 05:47 UTC, on request | Refreshes the organisation list from GOV.UK, repairs a change-flag baseline that lags the latest release, reminds about the review date, and keeps one provenance issue listing facts to verify and samples to replace (ADR 0020). | profile/organisations.json and content/section-baseline.json, after every check passes. A large organisation change (more than 25 rows) becomes a pull request instead. Review reminders and the provenance list are issues. | upkeep.yml |
| triage | 0 Observe | weekdays 07:23 UTC, on request | Labels contact, defect, change request and conformance claim issues that have waited too long for a reply (needs-reply) or a decision (overdue). | Labels only. | triage.yml |
| digest | 0 Observe | Mondays 08:13 UTC, on request | Writes the weekly operations digest for the owner: the Status page, what each routine did, failed runs, issues waiting, pull requests open and decisions needed. | One issue a week (label digest); closes the previous one. | digest.yml |
| labels | 1 Maintain | when .github/labels.yml changes, on request | Keeps the repository labels in line with .github/labels.yml. | Labels. | labels.yml |
| codeql | 0 Observe | every push, every pull request, Mondays | Scans the code for security problems. | Security alerts only. | codeql.yml |
| scorecard | 0 Observe | Mondays, on request, when branch rules change | OpenSSF Scorecard: an independent check of how the repository is run. | Security alerts only. | scorecard.yml |
| dependency-review | 0 Observe | every pull request | Stops a pull request that adds a dependency with a known moderate or worse vulnerability. | Nothing; fails the pull request. | dependency-review.yml |
Never automated
- Change repository settings, rulesets, secrets or the kill switch.
- Delete a branch, tag, release, issue or file a person made.
- Send anything outside the repository (pull requests or issues elsewhere, email, posts).
- Change the profile, its codelists, the requirements, the registers or any page content written by a person.
- Claim conformance, endorsement or approval.
- Merge a pull request.
Trials
New ideas are tried as trials: each states what it is trying to learn, what it may touch, how success is measured, when to stop, and an end date at most 90 days away. Each has its own kill switch, a repository variable TRIAL_ followed by its name, and AUTOMATION stops them all. The build fails if a trial runs past its end date, so every trial is either made a routine by a person or stopped.
No trials are running.
Near misses
Anyone, person or routine, reports a near miss or a surprise with the occurrence issue form. No blame: the report asks what happened and what would have stopped it, never who. A routine that stops itself (kill switch, failed gate, abort rule) says so in the digest. Every occurrence closes with either a check that would have caught it (as for a defect) or a recorded reason why none is needed. Report one with the occurrence form.
The same pattern runs across the connected portals; see ADR 0017.
Where the content comes from
Every fact, example and sample is listed with where it comes from. 264 items: 100 verified at source, 17 synthetic. Synthetic content is used only where nothing genuine can be shown in public, says why, and never names a real organisation as its publisher. Every Monday the upkeep routine opens or updates one issue (label provenance) listing what needs a person: facts to verify, checks that are out of date, and samples still to replace.
| Source | Label | Items |
|---|---|---|
| From a named source | S | 101 |
| Verified at source | V | 100 |
| Background knowledge, to verify | K | 38 |
| Synthetic, with the reason stated | 17 | |
| Built from a public record, not confirmed by the publisher | U | 6 |
| From a supplied pack | P | 2 |
To verify or replace (47)
| Item | Kind | Label | File | What to do |
|---|---|---|---|---|
| A roadmap for modern digital government 2025 to 2030 | fact | V | content/lineage.yaml | record the date it was checked |
| The UK Open Standards Principles ask government to choose open standards for software interoperability, data… | fact | K | content/lockin.yaml | verify against an authoritative source and relabel V with the date checked |
| The Technology Code of Practice asks teams to make things open, use open standards and avoid being locked in… | fact | K | content/lockin.yaml | verify against an authoritative source and relabel V with the date checked |
| The Service Standard asks services to use open standards and common components where they exist. | fact | K | content/lockin.yaml | verify against an authoritative source and relabel V with the date checked |
| xGov MEM v0.1.0 | fact | V | content/standards.yaml | record the date it was checked |
| data.europa.eu Metadata Quality Assessment | fact | K | content/toolkit.yaml | verify against an authoritative source and relabel V with the date checked |
| DCAT-AP.de | fact | K | content/toolkit.yaml | verify against an authoritative source and relabel V with the date checked |
| DCAT-AP-SE | fact | K | content/toolkit.yaml | verify against an authoritative source and relabel V with the date checked |
| DCAT-AP-ES (NTI-RISP) | fact | K | content/toolkit.yaml | verify against an authoritative source and relabel V with the date checked |
| DCAT-AP-CH (eCH-0200) | fact | K | content/toolkit.yaml | verify against an authoritative source and relabel V with the date checked |
| schema.org Dataset | fact | K | content/toolkit.yaml | verify against an authoritative source and relabel V with the date checked |
| (unnamed) | fact | K | content/vendors.yaml | verify against an authoritative source and relabel V with the date checked |
| (unnamed) | fact | K | content/vendors.yaml | verify against an authoritative source and relabel V with the date checked |
| (unnamed) | fact | K | content/vendors.yaml | verify against an authoritative source and relabel V with the date checked |
| (unnamed) | fact | K | content/vendors.yaml | verify against an authoritative source and relabel V with the date checked |
| (unnamed) | fact | K | content/vendors.yaml | verify against an authoritative source and relabel V with the date checked |
| (unnamed) | fact | K | content/vendors.yaml | verify against an authoritative source and relabel V with the date checked |
| (unnamed) | fact | K | content/vendors.yaml | verify against an authoritative source and relabel V with the date checked |
| (unnamed) | fact | K | content/vendors.yaml | verify against an authoritative source and relabel V with the date checked |
| (unnamed) | fact | K | content/vendors.yaml | verify against an authoritative source and relabel V with the date checked |
| (unnamed) | fact | K | content/vendors.yaml | verify against an authoritative source and relabel V with the date checked |
| (unnamed) | fact | K | content/vendors.yaml | verify against an authoritative source and relabel V with the date checked |
| (unnamed) | fact | K | content/vendors.yaml | verify against an authoritative source and relabel V with the date checked |
| (unnamed) | fact | K | content/vendors.yaml | verify against an authoritative source and relabel V with the date checked |
| Ab Initio Software | fact | K | content/vendors.yaml | verify against an authoritative source and relabel V with the date checked |
| Alex Solutions | fact | K | content/vendors.yaml | verify against an authoritative source and relabel V with the date checked |
| Anjana Data | fact | K | content/vendors.yaml | verify against an authoritative source and relabel V with the date checked |
| erwin by Quest | fact | K | content/vendors.yaml | verify against an authoritative source and relabel V with the date checked |
| Global Data Excellence | fact | K | content/vendors.yaml | verify against an authoritative source and relabel V with the date checked |
| Solidatus | fact | K | content/vendors.yaml | verify against an authoritative source and relabel V with the date checked |
| CKAN with ckanext-dcat | fact | K | profile/crosswalks.yaml | verify against an authoritative source and relabel V with the date checked |
| Dublin Core (15 elements) | fact | K | profile/crosswalks.yaml | verify against an authoritative source and relabel V with the date checked |
| UK GEMINI 2.3 | fact | K | profile/crosswalks.yaml | verify against an authoritative source and relabel V with the date checked |
| HDR UK Gateway schema | fact | K | profile/crosswalks.yaml | verify against an authoritative source and relabel V with the date checked |
| Croissant 1.1 (ML datasets) | fact | K | profile/crosswalks.yaml | verify against an authoritative source and relabel V with the date checked |
| DDI Codebook (study level) | fact | K | profile/crosswalks.yaml | verify against an authoritative source and relabel V with the date checked |
| StatDCAT-AP | fact | K | profile/extensions.yaml | verify against an authoritative source and relabel V with the date checked |
| Code of Practice for Statistics | fact | K | profile/extensions.yaml | verify against an authoritative source and relabel V with the date checked |
| mobilityDCAT-AP | fact | K | profile/extensions.yaml | verify against an authoritative source and relabel V with the date checked |
| org-id.guide identifier lists | fact | K | profile/extensions.yaml | verify against an authoritative source and relabel V with the date checked |
| An open dataset with two files | worked example | U | examples/examples.yaml | compare with the publisher's record each review |
| An open event API | worked example | U | examples/examples.yaml | compare with the publisher's record each review |
| An internal API that needs an agreement | worked example | U | examples/examples.yaml | compare with the publisher's record each review |
| A dataset series and one release in it | worked example | U | examples/examples.yaml | compare with the publisher's record each review |
| One release in a series | worked example | U | examples/examples.yaml | compare with the publisher's record each review |
| Geospatial example | extension example | U | profile/extensions.yaml | compare the added values with the publisher's own documentation |
| Weaver DCAT-US sample, an invented school census | tool sample | synthetic | tools/build.py | replace with a real public record |
Synthetic content and why (17)
- A critical data asset (synthetic): Departments report their critical data assets in returns that are not published, so no real record can be shown. Built to the April 2026 guidance, under an invented department.
- A catalogue (synthetic): It declares conformance to DCAT-AP-UK. No real catalogue has adopted this proposed profile, and showing a real one would claim a conformance it has not made.
- A harvested catalogue record (synthetic): It states which values a harvester added and which profile the record follows. Attributing that to a real central catalogue would describe behaviour it has not declared.
- Personal data shared across borders (synthetic): Details of personal data shared across borders are not published for individual datasets, and showing them for a real one would be wrong and could be harmful. Under an invented department.
- Values left empty on purpose (synthetic): It shows properties left empty on purpose. A real record with these gaps would misstate what that publisher knows or withholds.
- Synthetic example dataset: The extensions are candidates that no publisher uses yet. Attributing their values to a real dataset would misrepresent it, so the examples build on this invented record, under an invented organisation (ADR 0020).
- Synthetic GP appointments, Wales: The extensions are candidates that no publisher uses yet. Attributing their values to a real dataset would misrepresent it, so the examples build on this invented record, under an invented organisation (ADR 0020).
- Synthetic quarterly statistics: The extensions are candidates that no publisher uses yet. Attributing their values to a real dataset would misrepresent it, so the examples build on this invented record, under an invented organisation (ADR 0020).
- Synthetic bus timetables: The extensions are candidates that no publisher uses yet. Attributing their values to a real dataset would misrepresent it, so the examples build on this invented record, under an invented organisation (ADR 0020).
- Synthetic geospatial high-value dataset: The extensions are candidates that no publisher uses yet. Attributing their values to a real dataset would misrepresent it, so the examples build on this invented record, under an invented organisation (ADR 0020).
- Synthetic AI training dataset: The extensions are candidates that no publisher uses yet. Attributing their values to a real dataset would misrepresent it, so the examples build on this invented record, under an invented organisation (ADR 0020).
- Synthetic data space offer: The extensions are candidates that no publisher uses yet. Attributing their values to a real dataset would misrepresent it, so the examples build on this invented record, under an invented organisation (ADR 0020).
- Synthetic charity dataset: The extensions are candidates that no publisher uses yet. Attributing their values to a real dataset would misrepresent it, so the examples build on this invented record, under an invented organisation (ADR 0020).
- Synthetic network capacity data: The extensions are candidates that no publisher uses yet. Attributing their values to a real dataset would misrepresent it, so the examples build on this invented record, under an invented organisation (ADR 0020).
- Synthetic coordinate time series for a survey control network: The extensions are candidates that no publisher uses yet. Attributing their values to a real dataset would misrepresent it, so the examples build on this invented record, under an invented organisation (ADR 0020).
- Weaver DCAT-US sample, an invented school census: Still to replace with a real US federal record from catalog.data.gov; none was verified on 7 November 2026. Listed for the weekly provenance issue.
- Weaver Apache Atlas (Microsoft Purview) sample, an invented data map entity: Data map entities live inside an organisation's own Purview account and are not published, so no real one can be shown.
Machine-readable: provenance.json.
Machine-readable: status.json.