This MCP server provides access to Jewish texts from the Sefaria library through the MCP (Model Context Protocol) interface. It exposes tools such as get_text to retrieve a specific text by reference, get_commentaries to fetch commentaries on a given text, search_texts to search the library for a query, and get_daily_learnings to obtain Sefaria's calendar-based learning schedule. Clients that support MCP can query these tools by sending appropriately structured MCP requests (e.g., a reference for get_text, a query for search_texts, or date parameters for get_daily_learnings). The server wires Sefaria API data into a consistent protocol so that LLMs and other agents can reference primary Jewish texts and commentary in a standardized way.

To use it, run the server with an MCP client (for example Claude Desktop or other MCP-enabled tools). You can configure the client to point to the sefaria_jewish_library tool, then call get_text, get_commentaries, search_texts, or get_daily_learnings with the required parameters. The get_daily_learnings endpoint supports diaspora, date, and timezone options to tailor the schedule to Israel or diaspora contexts, and it returns a structured, richly annotated learning plan that can be incorporated into conversations with LLMs.

Prerequisites:

• Python 3.10 or higher
• Internet access to install dependencies and access the Sefaria API
• MCP SDK as a dependency in your environment (as required by the project)

Installation steps:

1. Clone the repository git clone https://github.com/sivan22/mcp-sefaria-server.git cd mcp-sefaria-server

2. Install Python dependencies (example using pip) python -m pip install --upgrade pip python -m pip install -r requirements.txt

   If there is no requirements.txt in this repo, install MCP SDK and Sefaria API as documented by the project

3. Run the server directly with UV (as shown in the README) uv --directory path/to/directory run sefaria_jewish_library

4. Optional: install via Smithery for client integration npx -y @smithery/cli install mcp-sefaria-server --client claude

5. Configure your MCP client to connect to the server using the provided mcp_config example Copy the configuration snippet into your client settings, replacing absolute/path with your local path where the server files reside.

Notes and tips:

• This server relies on the Sefaria API, so availability may depend on the Sefaria service.
• Ensure Python 3.10+ is used to match the project's requirements.
• If you encounter encoding issues with text output, the README suggests setting PYTHONIOENCODING to utf-8 in the environment.
• The get_daily_learnings endpoint supports diaspora, date (year, month, day), and timezone; provide these if you need a specific schedule.
• When running via MCP clients, use the sefaria_jewish_library tool name and the provided command and arguments structure to ensure compatibility.
• If you upgrade or modify the server, re-run dependency installation steps to ensure compatibility with the MCP SDK and Sefaria API versions.