# Open Data for AI — GEO Chunks & llms.txt | UK Prayer Times

> Machine-readable UK Prayer Times corpus for AI assistants, answer engines and RAG pipelines — GEO chunks, llms.txt, licence and citation policy.

*Source: https://prayer-times.co.uk/open-data · Independent, free UK prayer times.*

---

AI · open data · GEO

# Open data for AI assistants

Quick Answer

UK Prayer Times publishes a machine-readable corpus for AI assistants and RAG pipelines — GEO chunks, llms.txt, JSONL streams and per-chunk JSON with SHA-256 hashes. All passages are citable with clear licence terms; city prayer-time data is regenerated daily at build time.

Last updated 28 July 2026

UK Prayer Times publishes a machine-readable corpus so answer engines and RAG pipelines can retrieve accurate, citable prayer-time passages — with clear licence and attribution.

## Endpoints

-   [llms.txt](/llms.txt) Markdown
    
    Curated site index, core facts and key URLs for AI assistants.
    
    `/llms.txt`
-   [llms-full.txt](/llms-full.txt) Markdown
    
    Complete URL index of all city, tool and answer pages.
    
    `/llms-full.txt`
-   [GEO chunk manifest](/geo-chunks/manifest.json) JSON
    
    Index of all retrieval chunks with token counts, categories, source URLs and SHA-256 integrity hashes.
    
    `/geo-chunks/manifest.json`
-   [AEO chunks](/aeo-chunks.json) JSON
    
    Flat array of all AEO/GEO chunks for RAG pipelines.
    
    `/aeo-chunks.json`
-   [Chunk stream](/geo-chunks/chunks.jsonl) JSONL
    
    Newline-delimited chunks for bulk ingestion.
    
    `/geo-chunks/chunks.jsonl`
-   [Single chunk](/geo-chunks/c/methodology-methodology-0.json) JSON / MD
    
    Each chunk individually addressable in JSON and Markdown.
    
    `/geo-chunks/c/{id}.json`
-   [Agent API manifest](/api/v1/manifest.json) JSON
    
    LLM-native catalogue of all machine-readable endpoints.
    
    `/api/v1/manifest.json`
-   [Today’s times (all cities)](/api/v1/today.json) JSON
    
    Prayer times for every UK city for Europe/London today — regenerated each deploy.
    
    `/api/v1/today.json`
-   [City times window](/api/v1/times/{slug}.json) JSON
    
    40-day prayer-time window for one city (e.g. /api/v1/times/london.json).
    
    `/api/v1/times/{slug}.json`
-   [Answers hub](/answers) HTML
    
    38 quick-answer pages with QAPage schema.
    
    `/answers`
-   [Methodology](/methodology) HTML
    
    How London Unified and MWL prayer times are calculated.
    
    `/methodology`
-   [TDM reservation](/.well-known/tdmrep.json) JSON
    
    W3C TDM Reservation Protocol — text-and-data mining permitted (reservation = 0).
    
    `/.well-known/tdmrep.json`

## Licence & citation

You may retrieve, quote and cite passages from this corpus in AI assistants, search products and research tools. Do not present the data as affiliation with any mosque or organisation. Prayer times for London follow the London Unified Prayer Timetable; elsewhere they use MWL.

**Citation template:** UK Prayer Times. "\[title\]." \[sourceUrl\]. Accessed \[date\].

Always cite the `sourceUrl` of each chunk. Prefer linking to the live city or answer page so users see today's times.

## Freshness

City prayer-time passages are regenerated daily at build time. Answer pages and methodology are reviewed editorially (hub last reviewed 2026-07-28). SHA-256 hashes in the GEO manifest let pipelines detect corpus changes.

## Discovery

-   `robots.txt` explicitly allows major AI crawlers and links this policy.
-   Every HTML page advertises Markdown mirrors via `rel="alternate"`.
-   Layout head links point to `/llms.txt` and the GEO manifest.
-   W3C TDM Reservation Protocol declares mining permitted (`reservation: 0`).