Use Besorah in your AI assistant

Connect in a minute

The server’s address is https://besorah.bible/mcp. Add it as a remote MCP server (“connector”) in your assistant:

Claude

In the Claude apps, add a custom connector (Settings → Connectors) with the address above. In Claude Code:

claude mcp add --transport http besorah https://besorah.bible/mcp

ChatGPT

Add a custom connector (Settings → Connectors; your plan may require developer mode) and give it https://besorah.bible/mcp. No sign-in is needed.

Cursor

// .cursor/mcp.json
{ "mcpServers": { "besorah": { "url": "https://besorah.bible/mcp" } } }

VS Code

// .vscode/mcp.json
{ "servers": { "besorah": { "type": "http", "url": "https://besorah.bible/mcp" } } }

Anything else that speaks MCP over HTTP works the same way. A machine-readable card is at /.well-known/mcp.json, and everything in one file for an assistant to read is /llms-full.txt.

What to ask

Ask in plain language; the assistant picks the tools. These all work today:

“Find every Hiphil imperative in Isaiah.”
query_words — grammar and Strong’s search within a book

“Which words most often keep company with the Hebrew word for “shepherd”?”
get_collocations

“Compare Isaiah 7:14 in the Hebrew, the Septuagint and Matthew 1:23, and say what differs.”
compare_witnesses

“Which Old Testament passages does Hebrews 1 quote?”
get_quotations

“Trace the Passover theme from Exodus 12 through the New Testament.”
trace_theme

“Do a word study of hesed: its range of meaning and how the Septuagint renders it.”
search_words, get_word, get_word_occurrences

“What vocabulary do Isaiah and Micah share that few other books use?”
compare_books

“Which words occur only once in the Hebrew Bible and are found in Ruth?”
get_vocabulary (hapax legomena)

“Read Psalm 23 in Hebrew with each word’s lemma and parsing.”
get_passage with words

There are also ready-made prompts for a word study, a passage study, a grammar hunt, a theme survey, a book’s vocabulary and a quotation study — your assistant may offer them from its prompt menu.

What it knows

The Tanakh in Hebrew (Westminster Leningrad Codex), the Septuagint and Apocrypha, the New Testament in Greek and Hebrew, and the Targumim in Aramaic — each with English, in the traditional Jewish order with Hebrew numbering — plus Strong’s and Brown–Driver–Briggs, grammatical parsing for every tagged word, cross-references, quotations and parallels, songs, and the weekly portions. See the texts and sources & credits.

The 23 tools

Tool	What it does
Read a passage get_passage	Read a passage (verse, range or chapter) in its original language (Hebrew, Greek or Aramaic) and English, with a link.
Study a verse get_verse_study	Everything Besorah knows about one verse (or a range of up to 5): text, word-by-word Strong's/lemma/parsing, cross-references, quotations between books, curated parallels, similar wording, questions it asks or answers, songs, scribal marks, Ketiv/Qere, Targum connections.
Search the text search_text	Search the text itself.
Find a Hebrew or Greek word search_words	Find Hebrew or Greek words (Strong's entries) from an English word, a transliteration ('shalom', 'agape') or the original word.
Look up a word get_word	A Strong's lexicon entry (Hebrew H… or Greek G…): lemma, definition, KJV renderings, occurrences, for Hebrew the Brown–Driver–Briggs entry, how names are built, verses per book, and the Septuagint renderings of a Hebrew word (or the Hebrew words a Greek word renders).
Every verse a word is in get_word_occurrences	Every verse a Strong's number is in (Hebrew Tanakh, Greek New Testament and Septuagint), with the text, optionally within one book.
Related passages get_related_passages	For one verse: voted cross-references, quotations in either direction between books, curated parallel passages, and similar wording.
Cross-references get_cross_references	Cross-references for one verse (voted, OpenBible.info); the first few include their English text.
Connections between books get_book_connections	The chapters a book (or one chapter) is most joined to by cross-references across the whole Bible.
List the books list_books	Every book with its code, chapter count, the editions it exists in, and its URL; also the editions.
Songs and prayers in Scripture list_songs	Songs and prayers inside Scripture (the Song of the Sea, Hannah's prayer, the Magnificat…), by kind.
Questions and their answers list_questions	Questions asked in one place and answered in another (e.g.
Targum connections get_targum_connections	Where the Targumim and the New Testament share an idea or phrase.
Weekly Torah portion get_parashah	A weekly Torah portion with its Torah, Haftarah and Besorah (New Testament) readings.
Guide to the texts and tools get_guide	How Besorah's texts and tools fit together, and the conventions (numbering, ids, codes).
Find words by grammar query_words	Find every word by grammar and/or Strong's number, within a book or testament — e.g.
Words that keep company get_collocations	The words that keep company with a Hebrew or Greek word: which other words most often appear in the same verse, ranked by how much more than chance.
A book's vocabulary get_vocabulary	A book's vocabulary: size, most frequent and most characteristic words (summary); hapax legomena (words used once in the whole text); words used only in that book; or every word by frequency.
Compare two books' vocabulary compare_books	Compare two books' vocabularies: how many words they share, the rare words they share (most of whose uses are in these two books), and the words only one uses.
A verse in its setting get_context	A verse in its setting in one call: the verses before and after, the verse (or up to 5) word by word, and its cross-references, quotations and parallels.
Trace a theme trace_theme	Follow a passage's connections outward — voted cross-references plus quotations and parallels between books — one to three steps, and return the verses most strongly connected, with their text.
Compare the texts of a verse compare_witnesses	One verse across the texts: Hebrew, the Septuagint (matched by the app's verse tables, even where numbered differently), the Targumim, the Greek and Hebrew New Testament, and the quotations that join them — with the Greek words a New Testament quotation and the Septuagint do and do not share.
New Testament quotations get_quotations	Where the New Testament quotes the Old, in either direction: give a New Testament book/chapter/verse for the Old Testament passages it quotes, or an Old Testament one for the New Testament passages that quote it.

Good to know

• Read-only. Every tool is marked read-only; none writes or sends anything.
• Cited. Answers carry the address of the page they came from, so you can check the text.
• Honest about gaps. Verses whose words don’t line up with the word tagging are skipped, not guessed; approximate data is labelled.
• Open data. Texts and data are public domain or openly licensed — terms in the data licence notice.

Questions

Is it free, and do I need an account?

It is free and needs no account or key. Just add the address to your assistant.

Which assistants can use it?

Any that supports remote MCP servers over HTTP: Claude (the apps and Claude Code), ChatGPT’s connectors, Cursor, VS Code and others. If yours lists “Streamable HTTP”, it will work.

Does it change anything or collect my data?

No. Every tool is read-only and only reads texts that are public. Besorah keeps no account and no record of who asks what.

How accurate is it?

The texts and word data come from named open sources, credited in every answer. Where word tagging does not line up with the text, the tools skip the verse rather than guess, and approximate data (the Hebrew New Testament, Aramaic) is labelled as such.

Is there a limit?

A fair-use limit protects the server (about 60 requests every 10 seconds per address — far more than an assistant needs). Large answers can be shortened with max_chars.

For developers: the developer guide covers the REST API, Markdown/JSON for every page, feeds and calendars.