📖 Multilingual Holy Bible MCP Server

A high-performance, 100% offline, zero-latency Model Context Protocol (MCP) Server designed to connect Large Language Models (LLMs) to 11,907,047 Biblical verses across 1,081 translations in 800+ languages.

Equipped with strict theological guardrails (bible_scholar system prompt), Greek/Hebrew Strong's lexicon verification, and SHA-256 Merkle Root cryptographic integrity.

---

🌟 Key Features

• ⚡️ Zero-ML High-Speed Engine: Built on SQLite FTS5 with Memory-Mapped I/O (PRAGMA mmap_size), secondary SQL indexes, and an in-memory LRU cache (< 1.25 ms query response).
• 🔒 Cryptographic Proof (Merkle Root): Every verse is SHA-256 hashed. Global Merkle Root: e221d09e3870ddc23d3e1f62858a12b4152792847b911728371d39fa85279bb3.
• 🛡 Strict Theological Guardrails: Built-in bible_scholar system prompt prevents LLM hallucinations by forcing AIs (Claude, GPT, Gemini) to rely exclusively on scripture and verify Greek/Hebrew roots.
• 💻 100% Cross-Platform & Offline: Runs natively on macOS (M-Series ARM / Intel), Windows (x64 / ARM64), and Linux.

---

🚀 Quick Start (IDE Configuration)

1. Antigravity IDE Configuration

Add the following block to your mcp_config.json:

{
  "mcpServers": {
    "bible-mcp": {
      "command": "node",
      "args": [
        "/path/to/holy/mcp-server/build/index.js"
      ]
    }
  }
}

2. Claude Desktop Setup

Config location:

• 🍏 macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• 💻 Windows: %APPDATA%\Claude\claude_desktop_config.json
• 🐧 Linux: ~/.config/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "bible-mcp": {
      "command": "node",
      "args": [
        "/path/to/holy/mcp-server/build/index.js"
      ]
    }
  }
}

3. Cursor IDE Setup

1. Open Cursor Settings -> Features -> MCP.
2. Click + Add New MCP Server.
3. Set Name: bible-mcp | Type: command
4. Set Command: node /path/to/holy/mcp-server/build/index.js

---

🛠 Available MCP Tools

Tool Name	Parameters	Description
search_keyword	query (string), language (opt string)	Full-Text Search using FTS5 operators (AND, OR, wildcards *).
get_verse	book (string), chapter (num), verse (num), language (opt string)	Retrieve specific verse by canonical reference.
get_chapter_context	book (string), chapter (num), language (opt string)	Retrieve an entire chapter formatted for LLM context windows.
get_strongs_definition	word_id (string e.g. G26, H456)	Lookup Greek/Hebrew lexicon root meanings.
get_related_verses	book, chapter, verse	Retrieve cross-references and parallel passages.

---

💻 Cross-Platform Building & Running

Prerequisites

• Node.js >= 18
• Python >= 3.10 (for dataset building)

Local Compilation

# macOS / Linux
cd mcp-server
npm install
npm run build

# Windows (PowerShell)
cd mcp-server
npm install
npm run build

---

📄 License

Licensed under the MIT License. Free for open-source, commercial, and personal AI integration.