Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Magisterium MCP Server What does the Catholic Church teach about the Eucharist?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
[ARCHIVED] Built in 2025 before an official Magisterium MCP server existed, for personal use in Cursor. Archived now that I've moved tooling and the gap this filled has closed. Kept as a reference example of an MCP server integrating an external API — demonstrating early MCP adoption, end-to-end deployment, and AI-assistant integration.
An MCP server that provides AI assistants with access to the Magisterium API for authoritative Catholic Church teaching with citations.
Exposes one tool — magisterium_query — that any MCP-compatible client (Claude Desktop, Claude Code, Cursor, etc.) can call.
The server is hosted at https://magisterium.stephens.page/mcp — no local install required.
Add to ~/.claude/settings.json:
{
"mcpServers": {
"magisterium": {
"type": "url",
"url": "https://magisterium.stephens.page/mcp"
}
}
}Add to ~/.openclaw/openclaw.json. OpenClaw's URL-based transport is in the schema but not yet fully supported at runtime (#55087), so use the stdio transport with a local install for now:
{
mcp: {
servers: {
"magisterium": {
command: "/bin/bash",
args: ["/path/to/magisterium_mcp_server/run-magisterium.sh"],
},
},
},
}Once OpenClaw ships native HTTP MCP support, you can switch to the remote endpoint:
{
mcp: {
servers: {
"magisterium": {
url: "https://magisterium.stephens.page/mcp",
},
},
},
}See the Local Setup section below for install instructions.
Use the Streamable HTTP URL https://magisterium.stephens.page/mcp in your client's MCP configuration. Refer to your client's docs for the exact format.
Related MCP server: bible-mcp
If you prefer to run the server locally via stdio:
Node.js 18+ (check with node --version)
Magisterium API key from magisterium.com
git clone https://github.com/JacobStephens2/magisterium_mcp_server.git
cd magisterium_mcp_server
npm install
npm run buildCreate a .env file with your API key (see .env.example):
MAGISTERIUM_API_KEY=sk_your_key_hereVerify it works:
npm testAdd to your MCP config (claude_desktop_config.json for Desktop, ~/.claude/settings.json for Claude Code):
{
"mcpServers": {
"magisterium": {
"command": "/bin/bash",
"args": ["/path/to/magisterium_mcp_server/run-magisterium.sh"]
}
}
}magisterium_queryParameter | Type | Required | Default | Description |
| string | yes | — | Question about Catholic Church teaching |
| string | no |
| Model to use |
| boolean | no |
| Include related questions in response |
{
"name": "magisterium_query",
"arguments": {
"query": "What does the Catholic Church teach about the Eucharist?"
}
}The response includes the formatted teaching text, citations with document titles and authors, and optionally related follow-up questions.
The server runs as a single Express process serving three endpoints:
/ — Static frontend
/api/query — REST API for the frontend
/mcp — Streamable HTTP MCP endpoint for AI agents
Apache reverse-proxies /api and /mcp to the Node.js backend; TLS is terminated at the edge via Let's Encrypt.
npm run build # Compile TypeScript to dist/
npm run dev # Watch mode — auto-rebuild on changes
npm start # Run the stdio MCP server
npm run web # Run the web server (REST API + MCP endpoint)
npm test # Test server responds to MCP handshakeThe main source files are web-server.ts (Express server with REST + MCP endpoints) and mcp-magisterium.ts (stdio MCP server). After editing, run npm run build to recompile.
"Cannot find module" errors — Run npm install then npm run build.
"MAGISTERIUM_API_KEY environment variable is not set" — Ensure .env exists in the project root with your API key. The run-magisterium.sh wrapper script cds to the correct directory so dotenv can find it.
Server works manually but not in MCP client — Use the run-magisterium.sh wrapper script (not node directly). It ensures the working directory is correct regardless of how the client spawns the process.
"0 tools enabled" in Cursor — Fully quit and restart Cursor after adding the config. Check that the path in your config is correct and run-magisterium.sh is executable (chmod +x run-magisterium.sh).
Send a query to the Magisterium API to get authoritative Catholic Church teaching responses with citations
| Name | Required | Description | Default |
|---|---|---|---|
| model | No | The model to use (default: magisterium-1) | magisterium-1 |
| query | Yes | The question or topic to ask about Catholic Church teaching | |
| return_related_questions | No | Whether to return related questions (default: true) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, leaving the description to carry the transparency burden. It adds some behavioral context by indicating the responses are 'authoritative' and include 'citations', but it does not disclose potential side effects, permissions, rate limits, or return format. For a simple query tool, this is adequate but not rich.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, well-structured sentence. It is front-loaded with the main action and provides the key benefit ('with citations') without extraneous detail. Every word earns its place.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the lack of an output schema and sibling tools, the description provides sufficient context to understand the tool's purpose and likely return value (cited teaching responses). It is complete for a simple query tool, though it could optionally mention the response format or limitations.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, and all parameters already have descriptive text. The tool description itself adds no additional parameter semantics beyond restating that it sends a query. A baseline score of 3 is appropriate since the schema does the heavy lifting.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's verb ('Send a query'), resource ('Magisterium API'), and expected outcome ('authoritative Catholic Church teaching responses with citations'). It is specific and leaves no ambiguity about what the tool does.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
While no explicit alternatives or exclusions are given, the description clearly implies when to use this tool: whenever the user needs authoritative Catholic Church teachings. The context is sufficiently clear for an agent to determine appropriate usage, though it lacks explicit 'when not to use' guidance.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
v1.0.0magisterium_queryOnly one tool exists, so there is no potential for confusion between tools. The tool's purpose is clearly defined and distinct.
The tool name follows a clear verb_noun structure with 'magisterium_query' indicating the domain and action. With a single tool, there is no inconsistency to worry about.
The server has a single tool, which is slightly under the typical range but appropriate for its narrow scope of querying the Magisterium API. One tool feels justified given the focused purpose.
The tool covers the core need of querying for authoritative Church teaching responses with citations. Additional tools could explore related resources, but the current surface is functional for its intended use.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
AI Catholic formation — spiritual direction, saints, scripture, and semantic search.
Christian source evidence for AI agents, with provenance, tradition coverage, and citation checks.
Scripture-cited answers to any Bible question, plus verse text and study pages, for AI agents.
Web search and page-reading for AI agents. One-click OAuth connect, or a Caesar API key.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/JacobStephens2/magisterium-mcp-server'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
