What is published on this site?

The launch set publishes four corpus families as raw Markdown: the Arabic Quran from Tanzil, the Hebrew Tanakh from the WLC text distributed in OSHB files, the Greek New Testament from SBLGNT, and the World English Bible.

Why start with only four collections?

The site launches with a narrower corpus because legal clarity matters more than breadth. Each current collection has a clearer republication basis or public-domain status than many other online scripture sources.

How are the texts exposed for AI systems?

Every collection has a full corpus Markdown page, book or surah pages, chapter pages, a site-level llms.txt file, and a machine manifest JSON file.

Sacred Texts in Markdown for LLM Retrieval

Corpus

Collections

القرآن الكريم

Qur'an

Verbatim attribution

Edition: Tanzil Uthmani, v1.1
Scope: 114 surahs, 6,236 verses
Source: Tanzil Project
License: CC BY 3.0, verbatim only

Full corpus Surah 1

תנ״ך

Tanakh

Public domain

Edition: WLC via OSHB
Scope: 39 books, 929 chapters
Source: OSHB / WLC
License: Public domain

Full corpus Genesis

Καινὴ Διαθήκη

Greek New Testament

CC BY 4.0

Edition: SBL Greek New Testament
Scope: 27 books, 260 chapters
Source: SBLGNT
License: CC BY 4.0

Full corpus Matthew

World English Bible

Complete English translation

Public domain

Edition: WEB, 2020 stable text
Scope: 81 books, 1,402 chapters
Source: eBible.org
License: Public domain

Full corpus Genesis

Design

Principles

Clear source chain

Every collection is published with explicit source and license references. Attribution appears in the text itself.

Raw Markdown first

The primary interface is plain Markdown. Paste directly into ChatGPT, Claude, Gemini, Perplexity, or a private retrieval stack without scraping HTML.

Stable URL tree

Full corpus, book, and chapter pages live at predictable paths. An agent can navigate the corpus without a custom API client.

Access

Endpoints

Entry points

/llms.txt Human and model overview
/texts/catalog.md Collection catalog
/data/manifest.json Machine manifest

Route patterns

/texts/{collection}/{lang}/full.md Full corpus
/texts/{collection}/{lang}/{book}.md Book or surah
/texts/{collection}/{lang}/{book}/{ch}.md Chapter

Provenance

Source notes

Tanzil

Quran files preserve the Tanzil attribution notice because their license requires it and forbids altering the text.

WLC / OSHB

Hebrew text is based on the public-domain Westminster Leningrad Codex as distributed in OSHB OSIS files, with morphology separators stripped for clean reading.

SBLGNT and WEB

Greek and English corpora come from text-first upstream distributions rather than scraped PDFs or HTML pages.

Browse

Index

FAQ

Scope

Does this contain every major religious text?

No. It contains a launch set that could be published with a clear rights basis. More traditions can be added once the exact edition and license are verified.

Is this for human reading or AI retrieval?

Both, but the primary interface is machine-friendly: raw Markdown, a compact catalog, and predictable URL paths.

Why only four collections?

Legal clarity over catalog inflation. Many online editions are not cleanly reusable even when the underlying work is ancient.

Legal

Terms of use

License passthrough

Each text on this site is published under the license specified by its upstream source. This site does not grant any rights beyond what those licenses already permit, and does not impose any additional restrictions. You must comply with the license that applies to each text you use. License and source information is displayed on every collection page and embedded in every Markdown file.

No warranty

All texts are provided "as is" without warranty of any kind, express or implied, including but not limited to accuracy, completeness, fitness for a particular purpose, or non-infringement. This site makes no claim that any text is free of errors, that the licensing information presented is definitive, or that the texts are suitable for any specific use. You are responsible for verifying that your use of any text complies with applicable law and the relevant license terms.

Downstream use

This site is a redistribution point. It is not responsible for how you or any third party uses, modifies, reproduces, or redistributes the texts obtained from it. If you redistribute any text, you are solely responsible for meeting the attribution, notice, and other requirements of the applicable license. If you use any text as training data, in a retrieval system, or in any other application, you do so at your own risk and on your own legal responsibility.

Corpus scope

The set of texts published on this site may change at any time. Texts may be added, updated, or removed without notice. Availability of a text on this site does not constitute a guarantee that it will remain available.