home / mcp / planning center mcp server

Planning Center MCP Server

A Python MCP server exposing Planning Center APIs with read‑only tools for AI assistants across People, Services, Check-Ins, Giving, Groups, and Calendar.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "afristrup-planningcenter-python": {
      "command": "uv",
      "args": [
        "run",
        "python",
        "mcp_mock_server_comprehensive.py"
      ]
    }
  }
}

You run a Python MCP server that exposes Planning Center APIs to AI assistants and client applications. It provides read‑only MCP tools across People, Services, Check-Ins, Giving, Groups, Calendar, and more, enabling safe, structured access for automation, analytics, and conversational interfaces.

How to use

Install and run the MCP server in your environment, then connect your MCP client to the local server to access a wide set of Planning Center endpoints through ready‑to‑use tools. The server is designed for AI assistants and developers who want a stable, MCP‑compliant surface for Planning Center data. You can start a mock server for testing or run a real server with your API credentials, then use your MCP client to query data, filter results, paginate large datasets, and apply advanced analytics provided by the server.

Once the server is running, you interact with it through MCP‑compliant tooling or a client that supports the MCP protocol. You can perform read‑only operations to fetch lists of resources, retrieve individual items, and explore related data. For testing and development, use the mock server; for production, configure the real server with your Planning Center credentials and point your MCP client at the running process.

How to install

# Mock server (no credentials required)
cd planning-center-mcp-server
uv run python mcp_mock_server_comprehensive.py

# Real server (requires credentials)
# Set up your environment first, then start the real server
cd planning-center-mcp-server
uv run python mcp_server_fixed.py

Prerequisites you need installed before starting:
- Python 3.8+ and pip for dependency management.
- UV (the UV runner) to execute the MCP server commands.
- Optional: a .env file with your Planning Center API credentials if you run the real server.

Configuration and usage notes

MCP server configuration is provided through runtime commands. For local testing, you can start the mock server directly as shown above. For real usage, follow the steps to obtain your Planning Center API credentials, create or copy environment files as needed, and then start the real server so that it exposes the 80+ tools described in the MCP server surface.

Environment configuration examples include preparing an environment file or exporting variables, such as credentials or tokens, required by the real server. Ensure your client is configured to reach the running server endpoint and that your MCP client supports the required toolset and authentication method.

Troubleshooting and tips

If you encounter connection or authentication errors while using the real server, verify that your API credentials are correct and that the server process has started without errors. For rate limiting or transient server issues, rely on the MCP server’s built‑in retry handling and backoff policies.

For development and testing, prefer the mock server to avoid credential handling while validating tool availability and MCP workflow. If you need to simulate data, use the mock server’s realistic datasets and relationships to verify your client logic.

Examples of typical workflows

Fetch a list of people and iterate through them, including related emails and phone numbers when needed, to build a customer roster for analytics or outreach.

Retrieve calendar events for a given date range and filter by status to present upcoming engagements in a dashboard.