Companies are increasingly using AI assistants for their daily workflows. At the same time, the question arises of how to ensure consistency of corporate language. With the Lexeri MCP-Server, AI assistants can directly access terminology databases while considering the established terminology standards of the company.
What is the Model Context Protocol (MCP)?
The Model Context Protocol is an open standard that functions like an intelligent bridge between AI assistants and external data sources. Imagine your AI assistant could not only access its trained knowledge but also communicate directly with your company's proprietary databases and tools – that's exactly what MCP enables.
Lexeri and AI Integration: Practical Application
Our MCP-Server enables AI assistants like Claude or ChatGPT to communicate directly with Lexeri terminology databases. This opens up various application possibilities for everyday work:
- Term checks: AI assistants can automatically check texts for correct terminology.
- Term extraction: New terms from documents can be identified and suggested as terminology candidates in term requests.
- Workflow integration: AI assistants can update terms or add synonyms, additional attributes, or translations.
Available Functions
The current version of the MCP-Server offers the following functions:
- Searching and editing term entries directly from the AI assistant
- Creating new term suggestions e.g., as part of a term request or translation project
- Organizing subject areas and assigning entries
- Term checks for any texts
- Creating usage examples for better understanding
Technical Implementation
The MCP-Server can currently be integrated into Claude Desktop and the ChatGPT Responses API, for example. Configuration is done through standardized JSON files and requires a valid Lexeri API token.
Beta Version: Development Status and Participation
The Lexeri MCP-Server is currently in the beta phase:
- Availability: The functions can already be tested and tried out in workflows.
- Feedback integration: Feedback from practical application flows into further development.
- Feature expansion: The range of functions is continuously expanded based on user feedback.
Further Information
How you can try out the beta version of our MCP-Server yourself is explained in our Help Center: https://support.lexeri.com/hc/en-us/articles/28206830506525-MCP-server-Integrating-Lexeri-in-LLMs
If you have ideas, feature requests, or questions, please feel free to contact us. We look forward to hearing from you: