MCP server for GOV.UK — search, content, organisations and postcode lookup
Strong foundation with clear naming conventions (all tools start with action verbs: govuk_search, govuk_get_*, govuk_list_*, govuk_lookup_*). All 7 tools have descriptions (avg ~150 chars, within baseline 34-392 range). Input schemas present for all tools with proper JSON Schema types and constraints. However, output schemas are not documented, LLMs cannot predict response structure. Parameter descriptions are solid but some lack explicit format/range guidance. Error handling is minimal; no recovery guidance or actionable error messages visible in code.
Fetch a GOV.UK content item by base_path and return its metadata header, section index, and body sections.
Fetch a UK government organisation by slug. Returns the full organisation profile including title, acronym, type, state, web URL, parent/child relationships, and contact details.
Fetch a single section from a GOV.UK content item by base_path and anchor.
Search within a GOV.UK content item's body using regex or substring matching. Returns up to 25 matching sections with anchor, heading, snippet context, and the matched text. Useful for finding relevant sections without reading the entire page.
List UK government organisations with pagination. Returns a page of organisations with title, acronym, slug, type, state, web URL, and parent/child relationships. Use the slug field with govuk_search's filter_organisations parameter or govuk_get_organisation.
Resolve a UK postcode to administrative geography. Returns the postcode, local authority name and GSS code, region, and other geographic metadata. Uses the public postcodes.io API (no key required).
Output schemas not documented. LLMs cannot predict response structure (fields, types, nesting). Breaks downstream tool chaining and forces agents to guess field names.
Error handling lacks recovery guidance. No visible error messages that tell LLMs what to do next (retry, ask user, try alternative tool). Raw HTTP errors or exceptions provide no actionable path.
govuk_get_content and govuk_get_section descriptions are vague. 'Fetch a GOV.UK content item' does not explain WHEN to use get_content vs get_section vs grep_content. Missing dependency hints.
Inferred effective spec: 2025-06-18+.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 66 | 2025-06-18+ | v2 |
Search GOV.UK's 700k+ content items using the official Search API. Returns a list of matching content items with title, description, link, format, owning organisation(s), and last updated timestamp. Use filter_format to narrow to specific content types (e.g. 'transaction' for citizen-facing services, 'guide' for guidance, 'publication' for official documents). Use filter_organisations to restrict to a de
govuk_search filter_format parameter lacks enum constraint. Description lists examples ('guide', 'answer', 'transaction') but does not declare them as the only valid values. LLMs may hallucinate invalid formats.
govuk_grep_content 'pattern' parameter description mentions 'regex or substring' but does not specify error behavior if regex is invalid. Does fallback to literal matching happen silently? Needs explicit documentation.