Skip to main content

​
Overview

The Chift Model Context Protocol (MCP) server exposes the Chift Unified API as a set of tools that any LLM or agent framework supporting the MCP protocol (Claude, Cursor, VS Code, and others) can call directly. Instead of wiring up each endpoint by hand, your agent discovers the operations it is allowed to use and calls them like native functions. The tools mirror the Chift Unified API one-to-one and cover every vertical: accounting, banking, ecommerce, invoicing, payment, PMS, and POS. Because the tools map directly to the API, an agent can both read and write across your integrations. Asking an assistant to create an invoice, for example, walks through the same steps you would: it locates the customer, checks the VAT and revenue coding already used in the ledger, and posts a consistent invoice.
An AI assistant creating an invoice through the Chift MCP server, locating the customer and posting it to the accounting system

​
Two ways to run the MCP server

Chift offers two MCP servers. They share the same tool model but are built for different jobs, so pick the one that matches what you are doing.

​
Remote MCP server

The remote server is hosted by Chift at https://mcp.chift.eu/mcp. You point an MCP client at that URL, authorize through Chift, and you are connected, with nothing to install or run. It is designed to be productized: authorization is tied to a single consumer, and the tools an agent sees are limited to the scopes granted during that authorization. That makes it the right choice when you embed Chift in your own product or build an agent that acts on behalf of one of your customers, with a deliberately bounded set of tools. When the authorized consumer is connected to the DataLayer, reads are served from Chift’s synced data store, so your agent’s queries stay fast and are not throttled by the source system’s rate limits.

​
Local MCP server

The local server is an open-source Python package you run on your own machine over stdio. You authenticate once with your Chift API credentials, and it can work across all of your consumers rather than a single one. It is designed for builders: if you have a Chift account with several consumers and you want an assistant that helps you configure Chift and explore the Unified API across them, this is the one to use. It also ships an optional documentation search tool that brings the entire Chift documentation into your coding assistant’s context.

​
Which one should I use?

  • Building an agent or product feature that operates on one consumer with a fixed, scoped set of tools → use the remote server.
  • Working as a builder who needs an assistant across many consumers to set up and explore Chift, ideally with documentation search → use the local server.