In this newsletter, we’ll cover all of the features included in the August 2026 release candidate. We can break them up into two parts: First, there are a few enhancements to the core m-Power platform. Second, you’ll find some exciting new AI tools and features!
Webinar Replay
We covered these features during our webinar on August 20th. If you missed it, or would like to watch it again, you can find the recording below.
m-Power AI Studio
The m-Power AI Studio puts AI Assistant and Agent development in one place: A visual workspace where you build, test, and manage AI inside m-Power. From there you can build:
- AI Assistants. Build AI Assistants that are connected to live database data and/or internal reference documents.
- AI Agents. An agent understands what the user is asking for and routes the request to the right assistant. Use this for multi-step logic, instead of one giant prompt.
- Knowledge Bases. Upload the documents your business uses, like handbooks, policies, procedures, PDFs, etc….
The AI Studio also includes a Trace Viewer. It shows every user message and every assistant response, along with the details behind each one. When an assistant gives a strange answer, the trace viewer lets you see why that happened.
Learn more: AI Studio documentation | Watch the walkthrough
Human in the Loop
AI can be used for more than viewing data. A lesser known, but still incredibly powerful, feature of AI is the ability to write data back to your database.
However, due to data safety, m-Power utilizes a “Human in the Loop” approach when writing to your database. This means that instead of letting AI create SQL statements that autonomously communicate with the database, m-Power includes an interim step that lets you review what is being committed. m-Power also uses traditional Maintainer templates to handle the writing functions. These maintainers have additional functionality built in such as CSRF protection, input sanitization, etc.
Learn more: Human in the Loop documentation
AI Data Analytics
Data Analytics lets your end users ask questions about their data through a chatbot, instead of asking you to build a report for each one. Questions like “What are my top 10 items sold this year,” “Compare profit margin by customer for this year to date versus last year,” or “What departments have the highest callout rate.” It will provide answers, data, and even charts/graphs.
Learn more: Data Analytics documentation | Watch the walkthrough
Tabbed Excel Documents
m-Power can now split a single Excel export across multiple worksheet tabs.
You set it up on the Dimensions screen of your Data Model. Pick the field you want to break on, set “Page Break?” to Yes, and export. m-Power starts a new tab every time that field’s value changes. Break on State, and each state’s salespeople get their own tab. It works with both Normal Excel and Formatted Excel output.
Learn more: Excel outputs documentation
Field Validation
Field Validation lets you set data entry rules on maintainer web forms, catching bad input before it reaches the database. Rules can be set at the dictionary level or the application level.
Learn more: Field Validations documentation
Multi-File Upload
m-Power maintainers can now accept more than one file at a time. At runtime the filename field becomes an upload box: Users drag files in or click to browse, review what they picked, and remove any wrong ones before submitting.
Learn more: Multi-file upload documentation
One More Thing
If there’s something you’d like m-Power to do, tell us. Customer requests drive a lot of what we build. And if you want help putting any of this to work in your own applications, get in touch with our services team and we’ll walk through it with you.