Build, Automate and Observe: What You Can Do with Telerik, Kendo UI and AI Engineering This August

See what you can do with this August’s updates across Progress Telerik, Kendo UI, Telerik Reporting, Fiddler and AI Observability.

AI stopped being “the thing that writes your code” a while ago. At this point, it’s showing up at nearly every stage of the job: helping you build the app, automate the parts nobody wants to do by hand and keep an eye on how AI-powered features actually behave once real users touch them.

This month’s updates across Progress Telerik, Kendo UI, Telerik Reporting, Fiddler and Progress AI Observability are a good snapshot of that shift. Here’s what’s new and, more importantly, what it actually lets you do.

But before that … We’re walking through everything below on Progress AI Monthly: Telerik, Kendo UI & More livestream on August 28, 9 a.m. ET. No sign-up, just show up.

YouTube · LinkedIn · Twitch

Build

More AI, Fewer Tokens

Before getting to what’s new, a word on what’s gotten leaner: token consumption is down by up to 45% across our .NET products, with better output quality on top of that, so the same prompt now costs you less and gets you more. Alongside that, the Agentic UI Generator for Telerik UI for Blazor, KendoReact and Kendo UI for Angular is now available as an agent-native skill instead of an externally configured tool, which makes it noticeably easier to deploy in air-gapped and regulated environments.

See the MCP servers page.

Build UI with a Prompt

The Agentic UI Generator now supports Telerik UI for ASP.NET Core. Describe what you need in plain language, and it generates real, framework-aware code that follows the patterns you’d already be using: grids, forms, navigation, layout, the works. That means less time assembling scaffolding and more time on the parts of the app that actually need your judgment.

A few things people are already using it for: business applications, admin portals, customer-facing dashboards, analytics views and data-driven workflows.

See how the ASP.NET Core UI Generator works.

Build Reports with Natural Language

The Telerik Reporting AI Report Generator has already been generating charts and gauges from a plain description of what you want to see. Now that same idea extends to structured report content: tables, lists, crosstabs, summary rows, headers and footers, and grouped structures.

Describe the table you need. For instance, grouped by region, with subtotals. And the Generator builds it, then hands you a preview to check before anything actually lands in the report.

Explore AI-Generated Tables, Lists and Crosstabs.

Build Faster with New Components

Not every update this month is AI-flavored. Some of it is just useful, new UI:

• Telerik UI for Blazor: Expansion Panel
• KendoReact: OTP Input
• Kendo UI for Vue: Barcode, PDF Viewer
• Telerik UI for .NET MAUI: Charts (Preview), Circular Slider
• Telerik UI for WPF: Smart Paste Button, Inline AI Assistant
• Telerik UI for WinForms: Smart Paste Button

Build AI-ready Experiences with the LLM Kit

The LLM Kit isn’t just something you bolt on after the fact to keep tabs on your agents. It’s a set of components for building the AI-powered experience itself. Chat-style interfaces, multi-step agent workflows, tool-call displays, approval steps, citations, checkpoints: the pieces you’d otherwise have to design and build from scratch are already there, ready to wire into your app.

It gives you a real head start on shipping AI features that feel considered rather than bolted on, and, as a side benefit, your users and your team can actually see what the AI is doing along the way.

It’s available now for Telerik UI for Blazor, UI for ASP.NET Core, UI for MVC, Kendo UI for Angular, UI for jQuery and KendoReact.

Automate

Automate Application Upgrades

Nobody enjoys upgrading a codebase, and it’s rarely the highest-value use of anyone’s afternoon. AI-assisted migration is now available for Telerik UI for ASP.NET Core, UI for MVC, UI for WPF, KendoReact and Kendo UI for jQuery through the Upgrade Assistant (part of the MCP engine). It handles the changes it already knows how to make on its own and walks you through the ones that need a human decision, so version bumps stop eating whole sprints.

Automate Reporting Operations with AI Agents

Telerik Report Server now ships with a native MCP Server, so assistants like Copilot, Claude, Cursor and Codex can work directly inside your reporting environment instead of going through a custom integration you’d have to build and maintain. Once you’ve set up a Personal Access Token, agents can:

• Generate reports
• Manage data sources
• Schedule reports
• Handle general reporting-environment admin

See Report Server MCP docs.

Automate Debugging with Fiddler MCP

Debugging gets a lot more useful when your AI assistant can see what actually happened over the wire, not just what the code says should happen. Fiddler MCP connects your coding assistant to real traffic captured in Fiddler Everywhere, actual requests, responses, headers, status codes, so it’s reasoning from evidence instead of guessing. Set it up with one click inside Fiddler Everywhere, then go from “something’s broken” to a fix without leaving your IDE.

Explore Fiddler Everywhere MCP.

Automate Licensing for CI/CD

Automation isn’t only about what your agents do. It’s also about what shouldn’t need a human to manage. Deployment Keys give you a dedicated way to license automated builds and deployments: each key ties to a registered application and the exact Telerik or Kendo products it uses, so tracking, monitoring and reporting on product usage across your CI/CD environments gets a lot less manual.

Observe

Building faster and automating more only gets you so far if you can’t tell what your AI systems are actually doing once they’re live. That’s the part teams tend to underinvest in until something goes wrong.

Learn Why AI Observability Matters

If you want to go deeper on this, join Jeff Fritz and Progress Product Manager Lyubomir Atanasov on August 26 for “AI Observability: When AI Goes Off the Rails.” They’ll walk through some genuinely memorable public AI failures, then run a live Progress AI Observability demo covering agent tracing, drift and hallucination detection, and performance and cost monitoring.

Register for the webinar.

Help Shape the Future of Agentic Development

This is also the thinking behind the Progress Agent Harness Early Access Program: structured, auditable AI workflows, with early adopters getting a direct line to the roadmap. As agents take on more, being able to manage them predictably matters just as much as what they can build.

Join the Agent Harness EAP.

What’s New and Release History

To see everything that’ is new in August 2026 release edition, visit the What’s New in Telerik and Kendo UI page. For a deeper dive into each product, follow the links below.