This project integrates the Planning Center Online (PCO) API with an MCP server to enable seamless interaction with a Large Language Model (LLM). The goal is to allow users to ask questions and retrieve data from Planning Center in a conversational manner.
- PCO API Integration: Connects to Planning Center Online to access and manage data.
- FASTMCP Server: Acts as a middleware to handle requests and responses between the LLM and PCO API.
- LLM Query Support: Enables natural language queries to fetch and manipulate data from Planning Center.
- Retrieve information about services in Planning Center.
- Automate workflows by querying and updating data using natural language.
- Provide insights and analytics through conversational queries.
- Access to the Planning Center API.
- Python environment
- MCP Client (i.e. Claude Desktop)
- API keys for authentication.
-
Clone this repository:
git clone https://github.com/your-repo/pco-mcp-integration.git
-
Install dependencies:
uv pip install -r requirements.txt
-
Configure environment variables:
PCO_SECRET_KEY: Your Planning Center API key.PCO_APPLICATION_ID: URL of the MCP server.
-
Test the server:
fastmcp dev services.py
- Send a natural language query to the MCP server.
- The server processes the query and interacts with the PCO API.
- Receive a structured response or perform the requested action.
Add MCP server config
{
"mcpServers": {
"pco-services": {
"command": "/path/to/fastmcp",
"args": ["run", "/path/to/services.py"],
"env": {
"PCO_APPLICATION_ID": "PCO_CLIENT_ID",
"PCO_SECRET_KEY": "PCO_CLIENT_SECRET_KEY"
}
}
}It is intended to continue work on other areas of planning center.
Contributions are welcome! Please submit a pull request or open an issue for any suggestions or improvements.
This project is licensed under the MIT License.