magisterium_mcp_server
A node MCP Server for interacting with the Catholic LLM, Magisterium AI
claude mcp add --transport stdio jacobstephens2-magisterium_mcp_server npm run mcp \ --env MAGISTERIUM_API_KEY=""
How to use
Magisterium API Client provides an MCP server that exposes a dedicated tool called magisterium_query for querying the Magisterium API directly. After starting the MCP server, clients can send requests to the magisterium_query tool to obtain information about Catholic Church teaching, with support for optional model selection and related-question generation. This setup mirrors typical MCP ecosystems where a local or remote server exposes a tool that AI assistants can invoke to retrieve authoritative content from the Magisterium API.
To use it, ensure you have a valid MAGISTERIUM_API_KEY in your environment. Start the MCP server using the recommended npm run mcp command, and then invoke magisterium_query with a structured payload specifying the query, model (defaults to magisterium-1), and whether to return related questions. This enables integration with compatible AI assistants (e.g., Claude Desktop or other MCP clients) by pointing them to the server and providing the proper environment variable and configuration. The server will forward your request to the Magisterium API and return the formatted response for downstream consumption by your agent.
How to install
Prerequisites:
- Node.js 12.20.0 or newer installed on your system
- npm (comes with Node.js)
- A Magisterium API key to access the API
Setup steps:
- Clone the repository and navigate to the project root.
- Install dependencies: npm install
- Create a .env file (optional for local testing) and add your API key: MAGISTERIUM_API_KEY=your-actual-api-key-here
- Start the MCP server that exposes magisterium_query: npm run mcp
- Ensure the server is reachable by your MCP clients and test using the provided tooling or compatible clients.
Notes:
- If you deploy to a specific environment, you may adjust the working directory (cwd) in the MCP config accordingly.
- Keep your API key secure and do not commit it to version control.
Additional notes
Environment and usage tips:
- Ensure MAGISTERIUM_API_KEY is set in the environment where the MCP server runs.
- The MCP tool magisterium_query supports parameters such as query (required), model (default magisterium-1), and return_related_questions (default true).
- When integrating with clients like Claude Desktop, configure the mcpServers.magisterium object in claude_desktop_config.json with the appropriate command, args, and environment.
- Monitor for API rate limits or authenticated errors from the Magisterium API and handle retries as needed.
Common issues:
- Missing or invalid MAGISTERIUM_API_KEY leads to authentication errors when querying the API.
- If npm run mcp fails, verify that dependencies are installed (npm install) and that your Node.js version satisfies the requirement.
Related MCP Servers
time
⏰ Time MCP Server: Giving LLMs Time Awareness Capabilities
advanced-homeassistant
An advanced MCP server for Home Assistant. 🔋 Batteries included.
kanban
MCP Kanban is a specialized middleware designed to facilitate interaction between Large Language Models (LLMs) and Planka, a Kanban board application. It serves as an intermediary layer that provides LLMs with a simplified and enhanced API to interact with Planka's task management system.
rod
Model Context Protocol Server of Rod
mcp-human
Human Assistance for AI Assistants
memory
A MCP (Model Context Protocol) server providing long-term memory for LLMs