MCP data server in NET Genium
Every NET Genium instance includes an integrated MCP server (Model Context Protocol) that allows AI assistants and agents to work safely and directly with application data – to read it, analyse it, and, in dedicated environments, also modify it. It complements the AI models that describe the structure of an implementation: while the AI model explains how the application is built, the MCP server provides access to the contents of its records.
Typical clients are tools such as Claude, Claude Code, or Microsoft Copilot Studio, but any tool supporting the MCP protocol can connect. The client connects to the “api.aspx” page in the root of the NET Genium instance and authenticates with a personal MCP token.
What the server can do
The server tools are divided by permission into reading and writing:
- Reading data – listing available tables, reading records with optional filtering, and counting records. Change history tables and attachment metadata can be read as well.
- Writing data – creating, updating, and deleting records. Writing is intended primarily for development and test environments, not for production.
An assistant with read access can therefore answer questions that the AI model alone cannot – for example “how many records exist”, “find a specific document”, or “how did a record change over time”.
Personal MCP token
Access to the server is controlled by a personal MCP token of a specific user. Each user generates their own token in the “Change password” dialog (the “Change password” link in the “My settings” section of the navigation area) on the “MCP token” tab:
- the token is displayed in readable form only once, when generated – only its fingerprint (hash) is stored on the server,
- the token is valid for 90 days and can then be extended with a single click,
- each user has at most one token and can delete it at any time.
The token permissions determine which tools are available to the client:
| Reading all data via the MCP server | reading tools | only users with administrator rights |
| Writing all data via the MCP server | writing tools | only users with administrator rights |
| Access via a custom API | authentication against a custom API of a customer project | any user |
The built-in Administrator account additionally manages the tokens of other users – it can delete them and toggle their permissions.
Security
- Every request through the MCP server logs in as the token owner – it is therefore subject to the allowed IP addresses in the NET Genium settings and to licences, just like a regular login.
- Permissions are verified on every access – if the token owner loses administrator rights, access to the data ends immediately.
- The server is protected against token guessing – repeated requests with an invalid token are slowed down.
- It is recommended to enable the write permission only in development and test environments – never on production.
Typical use cases
- natural-language questions about data – “find projects without a responsible person”, “how many invoices are added each month”,
- consistency checks and summary analyses of data across applications,
- reading the change history to find out when and how a record changed,
- authenticating users against a custom API in customer projects.
Working with data through the MCP server is naturally followed by agentic development – an agent that knows the structure of the implementation from the AI model can, thanks to the data MCP server, also work with the real contents of the applications.