Organizze
@SamuelMoraesF
Organizze Finance Manager MCP Server. Create transactions, consult balances, and manage goals.
概要
What is Organizze?
Organizze is an MCP server that integrates with the Organizze financial manager, exposing its v2 API as AI tools. It is compatible with any MCP client such as Claude Desktop, allowing users to create transactions, check balances, manage goals, and more.
How to use Organizze?
You need two credentials: ORGANIZZE_EMAIL (your login email) and ORGANIZZE_API_KEY (your API key). You can run the server via uvx (recommended, no installation required), Docker (defaults to Streamable HTTP/SSE on port 8000, with a STDIO option), or local installation using pip or uv. For Claude Desktop integration, add the server to your mcpServers configuration file with the appropriate environment variables.
Key features of Organizze
- List, create, and detail bank accounts.
- Create expenses/revenues and list transactions.
- List and detail credit card invoices.
- Full management of categories and goals.
Use cases of Organizze
- Automate personal or small‑business expense logging via conversational AI.
- Query account balances and recent transactions without opening the Organizze app.
- Manage credit card invoices and track spending goals through natural language.
- Integrate Organizze data into AI‑powered financial dashboards.
- Create recurring or one‑off transactions programmatically using an MCP client.
FAQ from Organizze
What credentials are required to use Organizze?
You need your Organizze login email (ORGANIZZE_EMAIL) and your API key (ORGANIZZE_API_KEY).
What transport modes does Organizze support?
The server supports STDIO (default for Claude Desktop) and Streamable HTTP (SSE) on port 8000 when run via Docker.
How do I integrate Organizze with Claude Desktop?
Add an entry to your Claude Desktop configuration file under mcpServers with the command uvx, argument mcp-organizze, and the required environment variables for your email and API key.
Do I need a runtime like Docker or uv installed?
Yes. The recommended method uses uvx (from the uv package manager). Docker is also supported for containerized deployments, and a local pip install is available.
Where can I find the source code and contribute?
The project is hosted on GitHub under SamuelMoraesF/mcp-organizze, with CI/CD workflows and a Dockerfile for development.