Nintex Agent Designer

Introducing the Sentiment and Severity Agentflow Template When support requests start piling up, one of the biggest challenges is ensuring they're routed to the right team as quickly as possible. Not every request requires the same level of attention, but identifying which ones to prioritize often relies on someone reviewing and assessing each submission.The Sentiment and Severity Agentflow Template helps automate that decision-making process. Using AI, the agent analyzes the content of a request and evaluates factors such as sentiment, business impact, and urgency signals. Based on its analysis, it assigns a severity rating of Critical, High, Medium, or Low (P1 being critical and P4 low).On its own, this provides a consistent way to assess incoming requests. When combined with Nintex Workflow, the severity rating can automatically triage requests to the appropriate support tier or team. Information collected through forms, service requests, or other business processes can be passed to

AI agents are changing how organizations automate work. Unlike traditional automation, agents can interpret context, generate responses, use tools, and make decisions dynamically. That flexibility is what makes agents so powerful, but it also raises an important question: How do you know your agent is performing as intended? A successful demo is a great start, but real-world environments are far more dynamic. Agents can support a wide range of user requests, business data, tools, and workflows, making it important to evaluate performance across representative scenarios. That's where agent evaluations (now generally available in Nintex Workflow), come in. What are agent evaluations? Agent evaluations provide a structured way to assess and measure agent performance. Rather than relying on ad hoc testing or subjective opinions, teams can define what success looks like, create representative test cases, run evaluations, and review results against consistent criteria. With an evaluation fra

How do multiple AI agents work together to drive business processes? In our latest Nintex Agents video, we take a high-level look at how a multi-agent Agentflow can be designed, incorporated into a workflow, and ultimately used within an orchestration, helping you understand where these capabilities fit within a broader automation solution. Using a loan pre-assessment example, the video demonstrates a supervisor-worker agent pattern in which specialist agents collaborate to evaluate information, make decisions, and route work based on the outcomes. It also shows how an Agentflow can be called from a workflow and how those results can be used to drive progression within an orchestration. Watch the Video      Whether you're exploring agentic automation for the first time or looking for ideas on how agents can enhance existing solutions, this video provides a practical introduction to how multi-agent Agentflows, workflows, and orchestrations work together to create intelligent automation

AI agents are great at getting work done. But once they're in production, a different question emerges: How do you know they're doing the right work, and doing it reliably? Think about a typical scenario: You've built a support ticket triage agentflow. A support request is submitted, an agent summarises, categorizes them, assigns priority, and another routes it to the right team. Everything works great during testing, so you publish it and make it available to your customers. A week later, the real questions start:  How often is the agentflow being used?  Is it completing successfully?  Are failures occurring?  Why did a specific ticket get routed incorrectly?  How many AI tokens is the agentflow consuming?  Answering these questions requires two capabilities: monitoring and observability.  Monitoring: Understanding agentflow activity and health Monitoring gives you a high-level view of how your Agentflow is performing over time. For a support ticket triage agentflow, you might wan

Bring AI-Powered Intelligence to Accounts Payable Finance teams spend countless hours reviewing invoices, reconciling purchase orders, and investigating exceptions before payments can be approved. While many organizations have automated parts of the process, invoice matching remains one of the most time-consuming and exception-heavy tasks within Accounts Payable.Today, we're excited to introduce the Invoice & Purchase Order Matching Agent, a new Nintex Agentflows gallery template that uses AI to streamline Invoice-to-PO matching and help finance teams reduce manual effort, lower processing costs, and accelerate payment workflows. Modernizing a Core Accounts Payable ProcessThe agent is designed to automate one of the most fundamental finance operations: comparing incoming invoices against purchase orders and detecting discrepancies for overcharges.  Using AI, the agent can analyze invoices and purchase orders, compare the information, and determine whether discrepancies require furt

Faster Triage, Smarter Escalation, Better Customer Outcomes Support and Service desk teams are under constant pressure to respond quickly, prioritize accurately, and identify high-risk customer issues before they escalate. But as ticket volumes grow, manually reviewing every case for urgency, emotional intensity, and business impact becomes increasingly difficult.That is why we are excited to introduce the Support Ticket Sentiment and Severity Agent, a new agent template designed to help teams analyze incoming support cases and tickets faster, more consistently, and with greater intelligence. The template uses Nintex Agentflows to analyze sentiment in support-reported cases and tickets, helping support teams triage requests more accurately and quickly.Helping support teams identify what needs attention firstThe mission of the Support Ticket Sentiment and Severity Agent is simple: help support teams identify the most urgent, high-risk, or emotionally distressed customers before a situat

Hello Nintex Community! 👋We're excited to announce that Agent Designer and Orchestration entered General Availability (GA) on July 13, 2026, as part of Nintex CE and Workflow. Together, these capabilities make it significantly easier to manage complexity, adapt to change, and scale automation.  🚀 What these unlock for you:Break down complex, monolithic workflows into coordinated, manageable phases with Orchestration Introduce AI agents that can execute tasks with real-time decisioning, with or without human input Reduce bottlenecks and maintain context across the full process lifecycle Gain a more flexible, scalable low-code foundation to evolve workflows and accelerate time to value  📍 Stay connectedThe Nintex Agent Designer and Nintex Orchestration forum sections remain your primary destination for all announcements, best practices, learning resources, and community support. Whether you're looking for the latest information or have questions, be sure to subscribe to receive notifi

As you start building your Nintex AI agents, the first step is knowing where to configure and manage them. All agent setup happens in Nintex Agent Designer, a dedicated experience where you define how your agent thinks, behaves, and interacts with tools and data.  For more details and instructions use any of the following resources:Nintex HelpConfigure an Agent blog  (by Andy Hayes)

Configuring an agent is one of the foundational steps in building effective Agentflows. In this short how-to video, you'll learn how to configure a single Agent Action and understand the key components that influence how an agent behaves, accesses information, and returns results.  Whether you're building your first AI-powered workflow or refining an existing Agentflow, this overview will help you understand the core configuration options available. What You'll LearnThis video provides a high-level walkthrough of:Adding an Agent Action to an Agentflow. Understanding the four key configuration areas: Model Tools Knowledge Output Selecting and configuring an AI model and connection. Adjusting model settings such as: Reasoning effort Temperature Defining the agent's: Role and rules Work request Adding and using tools to enable agent actions. Connecting agents to knowledge sources and document stores. Understanding shared context and external vector embeddings. Configuring user mes

A question that keeps coming up: when do I use an agent versus a workflow?It's exactly where most teams get tripped up when adopting AI. Here's a clean mental model and some practical guidance you can actually use.Summary: Use agents to interpret and choose. Use workflows to execute. Use both together for scale and safety. The Core Distinction  Agent Workflow Role Interprets, reasons, decides Executes, routes, enforces Best for Ambiguity, judgment, adaptation Precision, repeatability, governance Output A choice or recommendation A deterministic action  When to Use an AgentUse agent tools when human-like judgment, adaptation, decisioning or choosing can be a solution without needing a human in the loop and/or less effort to build.Signals:The path forward isn't known in advance Inputs are unstructured (emails, documents, messages) Decisions depend on nuance or context Multiple valid next steps exist The

You are invited to join the weekly Agent and Orchestration Office Hours session focused on learning how to design, build, and operationalize AI agents.When? Weekly - Tuesdays 3:00 PM CT (Americas) / Wednesdays 8:00 AM AEDT (APAC)Where? Teams call: Join here!Each session will walk through a specific topic with a short demo and practical guidance you can apply immediately, such as:Build AI agents and understand when to use agents vs. simple actions Embed agents into workflows and orchestrations Design multi-agent solutions and master agents Connect agents to knowledge sources, tools, and external systems Implement guardrails, testing strategies, and human-in-the-loop patterns Deploy agents safely and monitor performance in productionThe office hours are designed to be interactive. Bring your questions, ideas, or scenarios you’d like help with.

AI is on everyone’s radar; the question is knowing where to start? The good news, you don’t need a massive overhaul to begin benefiting from AI. In fact, most companies find success by identifying just one clear use case and running a small instance or project. This introductory blog post – the first in our series on building AI agents for business applications – lays out practical best practices for getting started. We’ll explore why it’s smart to start with small, simple projects (think five steps or fewer), how to focus on straightforward, deterministic tasks for quick wins, and which high-impact, low-complexity use cases (like document processing, translation, or sentiment analysis) are perfect for early AI successes. Finally, we’ll discuss the importance of beginning at the department level and making AI a natural part of your design process. By focusing on clear, actionable steps and avoiding unnecessary complexity, you can dip your toes into AI confidently and set the stage for

Runtime AI brings AI-powered capabilities directly into the execution of a Nintex Workflow or App. Unlike design-time AI, which assists with building workflows, Runtime AI enhances how your live workflows operate by analyzing content, generating summaries, detecting entities, and making light, context-aware decisions as the workflow runs. Runtime AI Actions are lightweight, single-step automations that use large language models (LLMs) such as GPT and services like Azure Document Intelligence to perform specific content generation or analysis tasks. They are easy to configure and fit naturally into workflow execution, allowing you to add intelligence without complexity. All actions are delivered through the Runtime AI Platform Service and are accessible via API for integration across Nintex products. In this release, there are 11 out-of-the-box actions available in Nintex Workflow. VideosAnalyze sentiment Detect language Extract information from document or image Extract text from docum