Product Releases
and Updates

When you only have a few agents, managing them one by one works. But as teams ship more agentic apps with Agent Workbench, it becomes hard to answer basic questions at scale: Which agents are using which models? What actions are they allowed to run? Where do we need to tighten governance or troubleshoot? That information was scattered across individual agentic applications, not available in one place.
To better support governance, management, and operational efficiency at scale, we saw a clear need for a new console that provides this portfolio-wide visibility, making it much simpler to manage a growing ecosystem of AI Agents.

Introducing the Agents Console

The new Agents Console is a centralized hub for your AI Agents. In a single view, you get a complete list of agents across your environments and can immediately inspect each agent’s core configuration, including the specific AI Model it utilizes and the complete list of Actions it is authorized to execute. This new view streamlines governance, simplifies debugging, and provides the essential oversight needed to confidently manage and scale your AI applications.

To empower you with greater transparency and control over your external logic, we're announcing the ability to log and monitor your custom code.

This update introduces:

• Exposed Stack Traces: Whenever an error occurs within your custom code, you will now have access to the full stack trace. This provides details on where and why an error happened, accelerating diagnostics.
• Comprehensive Logging and Tracing Capabilities: You can now create detailed logs and traces for your External Library actions. These logs and traces will capture critical information, helping you monitor execution, track data flow, and identify potential issues proactively.

These new capabilities will streamline your development and maintenance work, making your experience with custom code better and more productive.

We are excited to announce the General Availability of the O11 to ODC Conversion Assessment Tool, the first component of the new O11 to ODC App Conversion Kit. This launch is a direct reflection of our commitment to supporting your modernization journey and empowering the transition to ODC at your own pace. This tool puts you in the driver's seat to decide if, when, and at what pace to convert your individual applications from O11 to ODC. It provides a clear, automated analysis of your entire O11 application portfolio, empowering you to properly plan your conversion journey accurately and build a predictable roadmap.

The Conversion Assessment Tool, installed from Forge onto your O11 infrastructure, analyzes your application portfolio to deliver crucial guidance for efficient ODC conversion planning. The tool assesses your O11 applications for conversion readiness by identifying incompatible code, data, and infrastructure patterns that need to be addressed for a successful transition to ODC.

It then delivers a detailed report with actionable insights, including:

• A breakdown of code, data, and infrastructure patterns found.
• Clear guidance on how and where to fix incompatibilities.

We are pleased to announce the release of MABS 11.2. This release delivers important improvements in scalability, stability, and containerization for the mobile build infrastructure, along with updates to several supported plugins.

If your production applications use any of the affected plugins, we recommend updating to MABS 11.2 as soon as possible to take advantage of the latest enhancements and ensure ongoing compatibility.

In accordance with the MABS versions and lifecycle guidelines, MABS 11.1 is now deprecated and will be made obsolete later this year.

We’re excited to announce that ODC allows configuring SMTP OAuth (Simple Mail Transfer Protocol Authentication). The new type of authentication helps ensure that only authorized users or applications can send messages, protecting systems from unauthorized use, spam, and spoofing.

SMTP OAuth on ODC enables seamless integration with modern email infrastructures, eliminates the need to store or transmit passwords, and strengthens the overall security posture of outbound communications.

We’re excited to introduce a major evolution in OutSystems Workflows — Internal Reusability.

This enhancement makes it easier and faster to design, maintain, and scale your process workflows. You can now reuse an activity (or set of activities) within the same workflow, reducing duplication and simplifying maintenance.

From a business perspective, this means teams can move faster and deliver more value — reusing existing workflow logic across processes instead of duplication of activities for the same operation. The result? Greater agility, simplicity, faster time-to-market, and more consistent process execution.

On the technical side, this update introduces a sideways ‘Go to flow step’ expanding beyond the previous limitation, which only allowed connections to previous nodes. Now, a ‘Go to flow step’ can target a wider range of activities within the same workflow, enabling the creation of reusable node patterns and more flexible, efficient process designs — all while maintaining flow integrity.

We're thrilled to announce the new OutSystems Personal Edition, a free, single access point for developers to explore the full power of the platform.

With this update, we're eliminating time-bound experiences and the need to move your apps. A Personal Edition is a "forever free", dedicated and permanent space for your personal projects and learning.

OutSystems Personal Edition includes:

• Dedicated space: Your personal space for a better, more focused experience.
• Agent Workbench: Free access to OutSystems Agent Workbench and other powerful AI capabilities to build, test, and innovate.
• Forge contribution: Developers can now submit assets to Forge under their individual ownership.
• Asset collaboration: Easily collaborate with other developers from the community on your projects.
• Dedicated storage: 1GB of dedicated application data storage—no more "database full" errors.

As businesses look to integrate agentic AI, they face a new challenge: "AI agent sprawl." This happens when different teams develop fragmented AI solutions without a central strategy, leading to a complex mess of tools, inconsistent data access, and a lack of governance. This siloed approach makes it difficult to manage agents at scale, introduces security risks, and prevents businesses from truly delivering enterprise-wide agentic transformation. The result is stalled pilots that can't deliver real business value.

Introducing Agent Workbench

The new Agent Workbench is now generally available, empowering you to build, manage, and scale intelligent AI agents on a unified platform for building agents and apps. Designed to help you build custom agents to streamline operations, elevate customer experiences, and grow revenue, Agent Workbench includes three sets of critical capabilities:

• AI model and data integrations: Set up a foundational layer of LLMs (custom or pretrained) and enterprise data ( structured or unstructured) to fuel agent action and decision-making and drive impactful results.
• AI agent lifecycle management: Build, test, deploy, and monitor agents with low-code agility, end-to-end security, and the full context of your entire app portfolio.
• AI agent orchestration: Build and orchestrate multi-agent workflows and patterns to deliver next-level automation and better outcomes.

Agent Workbench simplifies the transformation of existing business applications, workflows, and tools into intelligent, agentic systems that can reason, plan, and act. With one click, you can move the entire system—agents, apps, and all supporting layers—through development, testing, and production as an integrated whole, with all lifecycle management handled in a single toolchain.

Please review the Agent Execution limits fair usage policy for AI Agents in OutSystems Developer Cloud, as well as our promotional offer for Agent Workbench.