.png)
Four updates landed in the last month. Two of them change how you build automations. One changes how you map processes. And one opens up a entirely new data source for process mining.
Earlier this year we launched the Tekst MCP with read tools. Today, you can also build with it. Create and update automations, models and integrations. Directly from Claude, ChatGPT or Copilot. Just describe what you want and your AI assistant handles the configuration. Testing and validation are included. The only thing the MCP cannot do is publish an automation, which always requires your manual confirmation.
There's a reason our own solutions team has stopped building automations manually and fully switched to the MCP. It's that much faster. And the more MCP connections you combine, the more context your AI assistant has, which means better automations.

You know that 6-month process mapping project? We turned it into a coffee break. Point Tekst at your integrations and within minutes you get a complete process model. Main processes, subprocesses and estimated volumes to help you prioritize. This fully replaces Topic Discovery.
Not happy with the first result? Taxonomy Rework lets you remove, add, rename or lock processes, and let the AI regenerate the rest until it's exactly right. And with Draft Models, nothing goes live until you say so. Every change is saved as a draft first, with full version history so you can always go back.

Every change to an automation is now saved as a subversion. With a description of what changed, a view of the automation at that point, and the ability to roll back instantly. Combined with the existing QA diagnostics, this gives you a much more robust environment to build in, especially when building through the MCP.

For most manufacturing enterprises, the inbox is just the tip of the iceberg. The real operations happen in SAP, Dynamics or Oracle. Until now, that data was out of reach for process mining.
Not anymore. Tekst can now connect to your ERP data through a data warehouse integration. Redshift, PostgreSQL and MS Fabric are supported today, with more on the way. That data can then be used for process mining, including universal tracing across multiple integrations.
This means you can map the full process, from the first email to the final ERP entry.

Discover the impact of AI on your enterprise. We're here to help you get started.