May 5, 2026 Oak Brook, IL
mrc (michaels, ross & cole, ltd.), creator of the m-Power Development Platform, today announced a major update that lets businesses build autonomous AI agents over their own data, governed by their own IT teams. The release introduces a new Visual AI Agent Builder and the ability to deploy agents that handle complete, multi-step tasks without human intervention.
The update extends the AI capabilities mrc introduced in late 2025, which gave m-Power customers the ability to build AI assistants, retrieval tools, and AI-driven workflows. With this release, customers can move beyond assistants that answer questions and into agents that take action.
“Every IT leader we talk to is being asked the same question by their leadership: What's our AI strategy?” said Sal Stangarone, President of mrc. “Most of them don't have a good answer yet, because the options on the market force a tradeoff. Either you hand your data to a SaaS platform, or you let employees create agents in public AI tools with no oversight. We think there's a third option, and that's what this release is about. Build the agents your business needs, run them on your data, in your environment, and under your rules.”
Agents that do the work, not just answer questions
The Visual AI Agent Builder lets developers and IT teams design AI agents that complete full tasks end-to-end. Where an AI assistant might answer a question or summarize a record, an agent can process an inbound request, look up information across multiple systems, take an action, and report the results...all on its own.
Agents tie directly into the business logic customers have already built in m-Power. That means an agent can retrieve data, write data, trigger existing workflows, or call existing applications. They can go beyond simply reading from a database and actually use the logic the business already runs on. Since these agents can break a complex task into smaller pieces and delegate each piece to specialized sub-agents, businesses can build focused agents that do one thing well and then orchestrate them together to handle larger processes.
Agents work with live business data because m-Power sits directly on the customer's existing databases. Nothing gets copied to a vendor's cloud, and there's no integration project to deal with before the first agent goes live.
Built for IT control
The Visual AI Agent Builder is designed around a principle mrc has heard from IT leaders for the past two years: The governed path has to be the easiest path. If it isn't, employees route around it, and shadow AI gets worse.
Inside the builder, IT defines what each agent can do. They define what data it can access, which actions it can take, who can invoke it, and which LLM powers it. Every agent inherits the same security model as the rest of the m-Power platform, so access rules, authentication, and audit trails are consistent across the environment.
“You can't govern what you can't see, and you can't enforce a policy that lives in a PDF,” Stangarone said. “The point of the agent builder is to make IT the place where AI gets built, not the place that finds out about it after the fact.”
Model-agnostic by design
m-Power remains model-agnostic. Customers choose which large language model powers each agent, and they can swap models as the AI landscape evolves without rebuilding their applications. Agents can run on commercial models, open-source models, or models hosted inside the customer's own environment.
This matters to IT leaders who do not want to bet their AI strategy on a single vendor, and who need the flexibility to match the right model to the right task based on cost, performance, and data sensitivity.
A visual build process for AI agents
Building an agent in m-Power is now a visual process. Developers and IT teams lay out the agent's job, its data access, and its available actions inside a single interface, without stitching together the underlying logic by hand. The result is something you can actually read and review, and even update when requirements change.
One platform, one license
Every capability in this release is included in the standard m-Power license. There's no extra charge for agents, AI seats, or usage metering. Build and deploy as many agents as you need, for as many users as you have, across as many applications as you want. Same as everything else in m-Power.
m-Power is sold as a perpetual license. Customers own everything they build, agents included.
Backed by the mrc team
Customers building agents with m-Power are not on their own. mrc's support team is staffed with experienced m-Power consultants, and the services team is available to help design, build, or fully execute agent projects when a customer needs additional capacity.
“The software matters, but the team behind it matters more,” Stangarone said. “AI is moving fast, and most IT teams are stretched thin. We want our customers to know that when they decide to build something with m-Power, there are real people on the other end who care whether it works.”
Availability
The Visual AI Agent Builder and the autonomous agent capabilities will be available in the coming weeks to all m-Power customers on current maintenance. New customers can request a demo at mrc-productivity.com/aw/trial.html.
###
Media Contact:
Steve Hansen
Marketing Director, Michaels, Ross & Cole, Ltd.
Phone: 630.916.0662 x131
Email: mrc@mrc-productivity.com
About mrc
michaels, ross & cole, ltd. (mrc) has been helping businesses build custom web applications since 1981. The company's m-Power Development Platform is used by more than 1,500 customers worldwide to build custom applications, reports, dashboards, portals, workflow tools, and AI-powered tools over their existing databases. m-Power runs on-premise or in the customer's own cloud, supports any major database, and is licensed perpetually with no per-user fees. Learn more at www.mrc-productivity.com.