BuchhaltungsButler
Enable AI assistants to manage BuchhaltungsButler bookkeeping through a comprehensive Model Context Protocol server.
This Model Context Protocol (MCP) server provides a secure and robust interface for AI assistants like Claude and Cursor to interact with the BuchhaltungsButler API. It exposes all 48 API v1 endpoints as safety-categorized MCP tools, distinguishing between read-only, write, and destructive actions. Designed for ease of use and security, it features Docker deployment with health checks, built-in rate limiting, and server-side credential injection, ensuring that sensitive data never reaches the AI model. The project aims to provide a reliable and safe way to automate accounting tasks using AI.
Key Features
01Full coverage of all 48 BuchhaltungsButler API endpoints as auto-generated MCP tools.
02Robust Docker and docker-compose setup with health checks, auto-restart, and optional bearer-token authentication.
03Supports both stdio and Streamable HTTP transports for local or hosted deployment.
041 GitHub stars
05Built-in rate limiting to prevent exceeding BuchhaltungsButler's API request caps and server-side credential injection for enhanced security.
06Every tool is safety-categorized (read-only, write, destructive) with clear banners and machine-readable annotations.
Use Cases
01Automate listing and filtering of inbound receipts, transactions, and postings based on specific criteria.
02Create and manage draft invoices, upload PDF receipts, and assign them to matching transactions.
03Book bank transactions against predefined posting accounts or add new creditors and debtors via AI commands.