This MCP server exposes ChartMogul-related operations through a Python-based MCP server that you run locally via uv. The server expects a CHARTMOGUL_TOKEN for authentication with ChartMogul and provides access to a wide range of account, customer, contact, sales, plans, analytics, and data operations as listed in the README. After starting the server, you can use Claude Desktop (or equivalent MCP client) to connect to the MCP endpoint at the configured local address (default 127.0.0.1:6274) and invoke the available methods such as retrieve_account, list_customers, create_customer, list_invoices, all_metrics, and more.

To begin, ensure your token is configured in the environment variable CHARTMOGUL_TOKEN and that uv can execute the directory containing your chartmogul-mcp-server project. The example configuration maps the MCP server name mcp-chartmogul to uv with the target directory and runs the main.py script. Once running, you can manage data sources, customers, subscriptions, invoices, events, and analytics through the exposed tools and endpoints by name. The server is designed to be used with Claude Desktop configuration edits to point to this MCP instance, enabling seamless access to all supported ChartMogul operations.

Prerequisites:

• Python 3.8+ installed on your system
• uv (the Python package runner) installed and available in your PATH
• Access token for ChartMogul (CHARTMOGUL_TOKEN)
• A local clone of the chartmogul-mcp-server repository (or your working copy)

Step-by-step installation:

1. Clone the MCP server repository (or navigate to your existing copy): git clone https://github.com/chartmogul/chartmogul-mcp-server.git cd chartmogul-mcp-server

2. Create a local environment file and set your token: cp example.env .env

   Edit .env to include your ChartMogul token if required, or set CHARTMOGUL_TOKEN in your shell

3. Install uv and project dependencies as per the project instructions. This typically involves:

   Using pip to install uv if not already installed

   pip install uv

   Install any project dependencies (e.g., via a requirements file or setup script)

   uv sync # if the project uses uv's dependency syncing step

4. Ensure you can run the MCP server in development mode (as described in the README): mcp dev main.py:cm_mcp

5. Start the MCP server via uv using the configuration you set up in the mcp_config:

   Example command from the README configuration

   uv --directory /path/to/your/local/chartmogul-mcp-server run main.py

6. Verify the server is listening on http://127.0.0.1:6274 (as noted in the README) and that you can reach the supported endpoints.

Notes:

• If you modify any environment variables, restart the MCP server for changes to take effect.
• Ensure network access to the token resource (ChartMogul) is permitted from your environment.

Tips and common issues:

• Make sure CHARTMOGUL_TOKEN is set and valid; an invalid token will cause authentication failures when calling ChartMogul APIs.
• The server expects a local directory path to your chartmogul-mcp-server; ensure the path in the command matches your actual project location.
• If uv cannot be found, install it via pip install uv and ensure it is on your PATH.
• When developing, you may run mcp dev main.py:cm_mcp as described to start a development MCP instance; this may require Node.js if additional tooling is involved in your environment.
• The server exposes a large surface of APIs; refer to the README’s API sections (Account & Data Sources, Customer Management, Contacts, Customer Notes, Sales & CRM, Plans, Analytics & Metrics, Data Operations) to discover available methods and their arguments.
• To connect Claude Desktop (or another MCP client), add or edit your mcpServers configuration to point to the running uv process and include the required environment variables like CHARTMOGUL_TOKEN.