What is Berea?

Berea is a self-contained Bible service providing instant access to 59 Bible translations, Strong's Hebrew and Greek concordance with full scholarly lexicons, word-by-word morphological analysis, 446,544 cross-references, and 13 comprehensive Bible dictionaries. All data is embedded in a single optimized package for lightning-fast response times.

Features

• 59 Bible Translations — Including KJV, ASV, WEB, and translations in 38 languages
• Strong's Concordance — 8,674 Hebrew and 5,523 Greek entries with Abbott-Smith, Liddell-Scott-Jones, and Thayer's scholarly lexicons
• Word Study — Complete word analysis in one call: definitions, etymology trees, frequency data, scholarly lexicon entries, and sample verses
• Morphological Analysis — Word-by-word grammatical parsing for every verse: tense, voice, mood, person, number
• Cross-References — 446,544 connections between related passages
• 13 Bible Dictionaries — Thayer's Greek-English Lexicon, Bullinger's Figures of Speech, Wilson's Bible Types, ISBE, Easton's, Smith's, Fausset's, and more
• Verse Comparison — Compare any verse across multiple translations side by side, with Strong's-tagged text
• Semantic Search — AI-powered search that finds passages by meaning, not just keywords
• Projector Mode — Large fonts, fullscreen, configurable display for presentations
• Self-Hosted — Run on your own server or intranet for complete data control

Available Languages

Major Languages: English (KJV, ASV, WEB), Telugu, Hindi, Tamil, Spanish, French, German, Italian, Portuguese, Russian, Chinese, Korean, Japanese

Ancient Languages: Latin (Vulgate), Hebrew (Leningrad Codex), Greek

And 20+ more including Norwegian, Swedish, Danish, Finnish, Dutch, Polish, Czech, Hungarian, Thai, Vietnamese, and others.

ELS Torah Code Discovery (Berea MCP only)

The Berea MCP binary includes the only keyless Equidistant Letter Spacing discovery engine in existence. Every other Torah code tool ever built requires a search term — you feed it a word and it tells you whether that word is encoded. Berea turns the method around: it scans the Torah with no vocabulary input and returns whatever is actually there. No hypothesis. No selection bias. The Torah speaks and the tool transcribes.

The method itself goes back to Witztum, Rips, and Rosenberg's 1994 Statistical Science paper. Berea's contribution is industrialising it — scanning every letter at every skip interval, annotating every result against an empirical full-Torah noise baseline, and surfacing what rises above noise.

What makes Berea's approach unique:

• Keyless discovery — scans all 19,321 Hebrew words at every skip 2–49, in both linear and cylindrical modes (8 grid directions with column wrapping). No word list. No prompt. The tool asks the Torah what's encoded and returns the answer.
• Empirical noise baseline — 108,310 Hebrew words measured across all 304,805 Torah letters. Every finding is annotated with how many times that exact word appears across the entire Torah at that skip. Words with zero occurrences elsewhere are flagged as hapax-at-skip. Rarity is measured, not estimated.
• Statistical rigour — permutation-based p-values via els_pvalue, run in parallel against ten independently shuffled Torahs (same alphabet, same letter frequencies, same length, only the letter order randomised). If the signal were a property of random Hebrew, the shuffles would reproduce it. They don't.
• Deep scanner — ScanGridDeep (v0.6.4+) exhaustively searches a 2D cylindrical grid across skips 2–150, eight directions, with on-the-fly rarity annotation. Pre-committed falsification tests are published even when they fail.
• AI-native API — two-step discovery (summary scan then detail drill), cached, paginated. Works natively with Claude Desktop, Claude Code, Codex, and every MCP-compatible assistant.

Selected findings — every one produced without vocabulary input, each independently reproducible by running the same command:

• Deuteronomy 21:23 — "Cursed is every one that hangeth on a tree" (quoted in Galatians 3:13 about the crucifixion). Three words overlap the verse with zero occurrences elsewhere at their signed skip: תאלה (curse/imprecation), תלא (to hang/suspend), תקע (to drive a nail). The subject of the verse, the act it describes, and the execution method of its NT fulfilment — all encoded.
• Leviticus 16:21 — the Yom Kippur scapegoat. נרא overlaps the verse as a cross-reference to Isaiah 53:2 — "he hath no form nor comeliness." The Day of Atonement verse encodes a pointer to the suffering-servant chapter seven centuries before Isaiah wrote it.
• Genesis 22:2 — "Take now thy son, thine only son Isaac." יחידת cross-references Psalm 22 (the crucifixion psalm); הרר (a mountain) — they go to a mountain, and the mountain is encoded in the going-command.
• Exodus 12:13 — the Passover blood verse. Three forms of "pierced" plus "tears" overlap the verse — the Isaiah 53:5 vocabulary compressed into the verse that commands the first Passover.

The same rare words appear independently across multiple Torah verses, forming a cross-verse network: Deuteronomy 21:23 and Exodus 12:13 both encode crucifixion-vocabulary, Leviticus 16:21 encodes a pointer to Isaiah 53:2. These connections were produced by a tool that received no vocabulary input — each verse finds its own theological companions.

The seal rendered on every Berea page is itself a finding: ישוע שמי ("Yeshua Shmi" — "My name is Yeshua") is encoded at Gen 32:11 skip 1367 in the Koren Torah.

Explore the ELS Visualizer · 40 findings on berea-mcp.publifye.pro/els · or from the command line: berea-service call els_verse_signal ref="Deu 21:23"