From ERP for humans
Customers increasingly expect ERP systems to support AI-enabled ways of working. They want to use AI to search, analyze, automate, and act across business processes. They want real-time context instead of generic answers. They want flexibility to use the AI tools they have already licensed. And they want control over what AI can access and what it can do.
By making RamBase operable by AI, the MCP server creates a bridge between modern AI agents and the operational core of the business. It gives customers a way to experiment, automate, and build new workflows on top of their existing ERP data and processes.
In short, RamBase is moving from being a system people use, to a system people and AI agents can work with.
RamBase MCP makes it possible for AI to:
-
Read real data from RamBase
-
Understand your business processes
-
Create and update records
-
Support workflows across your ERP
RamBase MCP Beta Program
- Get early access to the MCP server
-
Explore real use cases tailored to your business
-
Work closely with our team to identify value
-
Influence how this capability evolves
A few important points
AI tools
You will use your own AI tools (for example Microsoft 365 Copilot, ChatGPT, Claude, Codex, or other MCP-compatible solutions).
Roles and permissions
The AI operates within the same roles and permission model that your users already have in RamBase, and it is also possible to further restrict access, for example by starting with read-only scenarios.
Data sharing
When connecting AI tools to RamBase via MCP, relevant data involved in requests and tasks will be shared with the chosen AI tool. How this data is handled depends on the specific AI solution, as well as your own configuration and agreements with that provider.
We value your feedback
This is an early version, and we expect active feedback and dialogue.
RamBase MCP Beta Program
FAQ
MCP is short for Model Context Protocol. An MCP server gives AI agents a standardized way to discover and use system capabilities. In simple terms, it works like an API layer built for AI, making it possible for AI tools to understand what actions are available and how to use them.
It makes RamBase accessible to AI agents. This means an AI tool can read data, search for information, create records, update fields, and support workflows in RamBase through controlled access to existing system capabilities.
Customers can use their own AI environment, as long as it supports MCP. Examples include tools such as Microsoft 365 Copilot, GitHub Copilot, ChatGPT, Claude, Codex, and other MCP-compatible AI tools.
No. The MCP server makes RamBase available to AI tools, but customers bring their own AI licenses and environments. The cost and choice of AI model are handled by the customer.
An AI agent can support tasks such as searching sales orders, creating quality issues, preparing supplier invoices, creating tasks, registering customer data, generating products from files or screenshots, and supporting purchasing or finance workflows.
Yes, if the user has the right permissions. The MCP server can allow AI agents to create, update, and potentially delete data through the API, but only within the access rights and governance rules that apply.
The AI agent works within the user’s existing role and permission setup. If a user is not allowed to perform an action in RamBase, the AI should not be able to perform that action on the user’s behalf. Actions are also logged under the user’s identity.
The MCP server is planned to be made available first to selected customers and partners for testing and feedback. This early phase will help validate use cases, technical setup, customer value, and go-to-market readiness.
Control your business with industry specific software.
Let’s have a no-obligation conversation and find out if RamBase is the right fit to help you control your business and achieve your goals.