Sefaria Jewish Library MCP Server

The Sefaria Jewish Library MCP Server bridges large‑language models with the extensive corpus of Jewish texts housed on Sefaria. By exposing a uniform Model Context Protocol interface, it allows AI assistants to retrieve, search, and contextualize scripture, commentaries, and learning schedules without embedding the entire library locally. This solves a common pain point for developers building educational or religious applications: accessing authoritative, up‑to‑date content while keeping the client lightweight and privacy‑respectful.

At its core, the server offers four primary tool families. get_text pulls a specific passage by reference—whether in Hebrew, English, or traditional Jewish notation—returning the exact text block. get_commentaries lists scholarly commentaries for that reference, enabling an assistant to surface multiple perspectives. search_texts performs full‑text search across the library, supporting filters (e.g., Talmud Bavli) and proximity scoring. Finally, get_daily_learnings taps Sefaria’s calendar API to deliver the current Torah portion, Haftarah, Daf Yomi, and other learning cycles, with optional diaspora or custom selections. These tools give developers a rich set of data points to construct conversational flows that are both accurate and pedagogically sound.

Developers can integrate the server into any MCP‑compatible workflow—whether they run a local UV server, invoke it through Smithery for Claude Desktop, or embed it in a custom agent. The protocol’s request/response model means that an assistant can seamlessly ask for a verse, receive the text, and then request related commentaries or search results in the same interaction. This tight coupling reduces round‑trip latency and simplifies state management, allowing developers to focus on higher‑level dialogue design.

Real‑world scenarios abound: a study app could let users query “Genesis 1:1” and instantly receive the Hebrew text, an English translation, and a list of commentaries to read next. A virtual Torah study group might pull the weekly portion schedule for a given date and generate discussion prompts automatically. A chatbot teaching Jewish law could search for all references to “moshiach” and present relevant passages filtered by tractate. Because the server relies on Sefaria’s continuously updated API, content remains current without manual curation.

What sets this MCP server apart is its blend of depth and accessibility. It exposes a full, searchable corpus—something that would otherwise require significant storage or licensing effort—and does so through a standard protocol that any LLM can consume. The inclusion of calendar‑based learning schedules further enriches the experience, making it a one‑stop solution for developers building faith‑based educational tools that need authoritative, real‑time Jewish text access.