moneybird

Moneybird MCP Server exposes Moneybird data to AI assistants via the Model Context Protocol (MCP). It enables tools to list and retrieve contacts, view sales invoices, financial accounts, products, projects, and time entries, and it also supports making custom Moneybird API requests. To use it, start the server with the provided command and ensure your Moneybird API token and administration ID are configured. Once running, you can connect your MCP-enabled AI assistant to the server URL and invoke the available tools such as list_contacts, get_contact, list_sales_invoices, get_sales_invoice, list_financial_accounts, list_products, list_projects, list_time_entries, and moneybird_request. This setup allows the AI to perform real-time queries against Moneybird data and perform common business operations through the assistant’s interface.

Prerequisites:

• Node.js (v18+ recommended)
• npm (comes with Node.js)
• A Moneybird account with API access

1. Install the MCP server package globally (or adapt to your deployment method): npm install -g moneybird-mcp-server

2. Create a configuration file or set environment variables for Moneybird credentials: MONEYBIRD_API_TOKEN=your_api_token_here MONEYBIRD_ADMINISTRATION_ID=your_administration_id_here

3. Run the MCP server locally to verify: npx moneybird-mcp-server

4. (Optional) Persist credentials in a .env file and load them in your deployment environment. For Claude Desktop or similar integrations, ensure the config includes the required environment variables.

5. If deploying in production, consider containerization or a process manager and ensure secure access (HTTPS, authentication, and network restrictions).

Tips and common considerations:

• Keep your Moneybird API token secure and rotate it periodically.
• If you encounter rate limits, implement appropriate paging/filters when listing large datasets.
• Use the advanced filtering options supported by Moneybird for more precise queries (e.g., created_after, updated_after).
• When deploying behind a proxy or load balancer, ensure the MCP server is reachable only by trusted clients and enforce authentication.
• The available tools are designed to cover typical Moneybird usage; for custom needs, use the moneybird_request tool to issue direct API calls.
• Ensure the environment variables are loaded in the runtime environment where the MCP server executes (e.g., cloud host, container, or Claude Desktop configuration).
• Maintain proper error handling in the AI prompts to gracefully handle API errors and transient failures.