Product Releases
and Updates

We have reinforced LifeTime Analytics - our built-in monitoring tool - to deliver out of the box performance insights for Reactive Web and Mobile applications.

You can now leverage deep client, network, and server execution breakdowns alongside APDEX scoring across your entire application portfolio. Paired with Code Quality (formerly AI Mentor Studio), live performance data connects directly with codebase insights—allowing you to trace a degraded screen in LifeTime Analytics straight to the underlying code findings for rapid troubleshooting within O11. 

This capability is currently available in Beta and requires manual activation.

Agentic Loan Applications is now available, expanding the OutSystems Agentic Industry Solutions portfolio for banking. It brings customer-facing applications, governed agents, and deterministic workflows together into a single pre-built system for consumer lending that banks adapt to their own products, policies, and existing banking systems.

The solution layers over your existing banking systems and grounds its agents in your operational context - your policies, systems, and dependencies - so agents operate more efficiently and accurately while every lending decision stays under the bank's control.

What you get:

• A guided, agent-assisted application journey. A conversational agent walks applicants through a personal unsecured loan application: it captures intent, collects and validates documents, runs identity and sanctions screening, and compiles a complete application that passes to your existing loan origination system.
• Agents evaluated before they reach production. Every banking agent is tested against dozens of agent evaluations covering criteria such as relevance, accuracy, and PII protection, and re-tested after any model, prompt, or tool change so agents stay within guardrails.
• Model selection tested for cost and accuracy. OutSystems evaluated multiple models through Amazon Bedrock and pre-selected models that perform optimally at up to 82% lower cost. Banks configure the system around their own products, policies, and existing banking systems while retaining control over their data.

Inside the solution:

• Apps: Mobile Banking, the customer channel for starting and tracking loan applications, and Lending Console, the back-office workspace where lending staff review, intervene in, and progress applications with AI assistance.
• Agents: Lending Conversation Agent guides applicants from intent to submission; Lending Policy Agent answers product, fee, rate, and eligibility questions strictly from approved documentation; Banking Document Agent validates loan evidence and flags only concrete, traceable discrepancies; Lending Fraud Agent surfaces evidence-anchored fraud, KYC/AML, and credit-risk indicators to inform human decisions.
• Workflow and services: Lending Application Workflow orchestrates the full lifecycle - submission, regulatory screening, risk and fraud assessment, credit checks, loan-officer validation, final offer, and either account creation or an adverse-action notice - with durable loan state, status history, and audit centralized in the lending services layer.
• Integration gateways: Stable contracts isolate credit bureau, regulatory and compliance (OFAC, FinCEN/BSA, SSA eCBSV, DMDC MLA/SCRA, and fair-lending checks), and core-banking integrations from the origination flow.

The Banking Agents Kit is also available on the OutSystems Forge, with building blocks for creating banking apps and agents on the OutSystems platform.

ODC Self-Hosted now supports the complete interoperability suite—enabling organizations running Self-Managed OutSystems 11 deployments to adopt ODC Self-Hosted while retaining full control within their own customer-managed infrastructure. Connect seamlessly with your existing O11 environments with zero migration required.

What’s Included:

• Data Interoperability: ODC Self-Hosted includes the Data Fabric Connector for O11, allowing developers to read and write (create, read, update, delete) O11 data directly through a secure connection. Ground your new ODC AI agents and apps in live O11 core data without building custom pipelines or copying databases.
• Logic Interoperability: ODC Self-Hosted apps can invoke O11 REST APIs through a dedicated private tunnel, eliminating public internet exposure. Reuse existing O11 business logic in your ODC applications without rebuilding from scratch.
• User Interoperability: Enables End-User Single Sign-On (SSO) for seamless navigation between O11 and ODC Self-Hosted apps. Users authenticate once for a frictionless experience across both estates.

Why it matters: This update enables any O11 customer to deploy ODC Self-Hosted on their own infrastructure and leverage the full power of O11/ODC interoperability—keeping O11 as their operational foundation while unlocking AI-driven, cloud-native innovation.

AI Mentor Studio in O11 is now called Code Quality, matching ODC's naming for the same underlying engine. The categories, findings, and status model you already know haven't changed, only the name.

O11 now supports Azure SQL Managed Instance as a platform database. Azure SQL MI is a fully managed cloud database service running on the latest stable SQL Server engine, with close to 100% SQL Server feature compatibility - including capabilities not available in Azure SQL Database. Teams can configure their O11 environment against an Azure SQL MI instance and gain an Azure native infrastructure. 

To activate, simply run the Configuration Tool and select Azure SQL Managed Instance as your database option.

Gain full visibility into your deployment history. You can now programmatically retrieve and query publish comments directly via O11 system entities to automate your compliance, tracking and auditing workflows. 

Tracking the what, when and why of application changes is critical for enterprise governance, but keeping a clean history hasn't always been easy. Previously, 1-Click Publish comments were locked within the platform database, making them inaccessible to external tools. This forced teams to rely on manual tracking workarounds or direct database queries just to maintain a basic audit trail which slowed down delivery and broke end-to-end traceability. 

To eliminate this friction, O11 now exposes published comments through public, read-only system entities. This will allow you to bridge the gap between O11 and the rest of your enterprise ecosystem.

Agent Monitoring is now available in Beta. Agent Monitoring gives developers and AI engineers full visibility into how their agents behave across every stage, from development to production. Plus, it's configurable per agent, giving you total control to choose which ones you monitor.

Once enabled for an agent, you’ll be able to see:

• Input and output tokens for every agent call, with historical usage trends over time
• Two new quality signals, relevance and safety, scored automatically on your agent traffic alongside your usage data
• The full input and output of every LLM call, plus which tools the agent used to get there

Together, this means you can inspect exactly what happened on any past execution, and track whether your agents are getting slower, more costly, or less accurate before your users notice.

This feature is fully integrated into the Analytics console, where you can now view detailed agent information.

AI Guardrails protect your agents using a set of predefined categories, but every organization has content, compliance, and data-protection rules that go beyond the defaults. To support your unique business needs, we're evolving Guardrails so you can set custom rules across your topics, terms, and data formats.

Custom guardrail policies

System Administrators can now define three kinds of custom policies that apply as an organizational baseline across every agent:

• Denied topics described in plain language. Detection evaluates meaning and intent, not just keywords, so content is blocked even when it's paraphrased or implied.
• Word and phrase filters for exact terms such as competitor names, internal codewords, or sensitive phrases.
• Custom PII patterns are defined with regular expressions, so sensitive data formats unique to your organization, like employee IDs, account numbers, or booking codes, are detected and either blocked or masked.

Developers can extend this organizational baseline with rules specific to their own agent, adding stricter controls where a use case demands it, but they can never weaken the baseline administrator set. The result is seamless, organization-wide protection with flexibility for targeted, agent-level hardening. Every intervention is logged and traced to the exact rule that triggered it, so it remains easy to audit and troubleshoot guardrail behavior.

In regulated environments, trust is never assumed. For governed industries and B2B ecosystems, every incoming connection must verify its identity, not just be encrypted. O11 Cloud now supports mTLS passthrough for REST API endpoints, giving your applications the ability to authenticate clients by certificate, not just credentials.

Built for compliance driven environments, partner integrations, and zero trust architectures. Available on request.

System Administrators can now configure an AWS Bedrock AI Model connection using IAM role-based authentication. IAM Role is the recommended default for new connections. When selected, OutSystems generates a unique External ID for the tenant, which the administrator uses to configure a trust relationship in their AWS account with the help of ready-made setup templates provided directly in the portal. OutSystems issues short-lived credentials at runtime without storing any AWS access keys.

Before going live, administrators can test the connection directly in the portal to confirm everything is set up correctly, with clear guidance if something needs fixing. The External ID can be rotated at any time for added security. Agents using Bedrock models over IAM role-based authentication run completely on their own at runtime, requiring zero developer action. Access Keys authentication remains available for existing connections and is not affected.