Community Central

Hi Community!After an exciting first quarter, April is packed with major model upgrades, smarter search and a brand new developer homebase. We’re kicking things off with GPT-5.2 going GA, rolling out more powerful multilingual capabilities, and making it easier than ever to build in Agent Studio. Plus, don’t miss Knowledge26 in May and our lineup of April office hours to stay ahead of what’s next.📩 Save the Date: Knowledge May 5–7, 2026 | Las VegasDon't miss your chance to be part of Knowledge, where you and AI get to work.  Learn how Moveworks from ServiceNow is powering the AI front door. Learn more and sign up. 📹 April Webinar and Office HoursApril 2 — Developer Office Hours: 

We’ve refreshed the Program Owner Certification to better support you beyond launch and throughout your entire Moveworks journey.The updated course is designed for:• New program owners or champions ramping up• Teams transitioning ownership post-implementation• Anyone looking to get more value from their Moveworks programThe learning path now focuses on three key areas:• Foundations of Program Ownership• Program Execution• Optimization & Value RealizationYou’ll find guidance on driving adoption, expanding use cases, measuring impact, and more — all tailored to where you are today.👉 Access the certification here: Program Owner Certification

The thing I keep seeing at events like this: developers assume connecting a new tool to an AI assistant is a multi-sprint project. It's not anymore.I'll be running a hands-on mini lab at CreatorCon where you'll go from zero to a deployed, production-ready tool in ServiceNow EmployeeWorks in minutes. Connectors configured, permissions set, tested through natural language.First come, first seated. If you're curious how fast this can actually move, come find out.   

Most AI stops at the edge of one system. We're going to break that habit at CreatorCon this year.In our hands-on lab, you'll build tools that search and act across external business applications through the AI Assistant in EmployeeWorks. Real connectors, real permission enforcement, natural language testing. Not a demo. You'll actually build it and leave with a blueprint you can repeat.If you want to see what cross-system AI tooling looks like when it's done right, this is the lab.Learn more here.  

Whether you just shipped something small or something wild, we want to hear about it. Use this format to keep it quick and useful for everyone: Here's what I built: (1–2 sentences) What problem did you solve or idea did you bring to life? Here's what I learned: (1–2 sentences) What surprised you or stretched your skills? Here's my #1 tip: (1–2 sentences) What advice would you give to the next builder?Here’s an example of a similar share out for your reference! 

I’m excited to announce that I’ll be co-hosting a session alongside my colleague ​@Ajay Merchia at ServiceNow Knowledge ‘26. You don't want to miss out - sign-up today!We’ll explore how ServiceNow EmployeeWorks, powered by the Moveworks agentic platform, serves as the AI-powered front door for work, enabling employees to search and take action across HR, IT, finance, and other business applications through conversation. We’ll discuss the role developers play in building and extending tools and enabling AI agents to act across applications. Bring your questions on tool development and designing AI-driven workflows that operate across systems.Learn more about our session and register here. 

It's almost always the output mapper.When you return structured JSON, the reasoning engine parses it — extracts the schema, filters fields, routes to code execution if needed. When you return a flat string via RENDER, it can't do any of that. The whole blob gets dumped into context as-is.The fix: use $MAP() for lists and direct field mapping for single objects. Save RENDER for cases where you genuinely need static text nothing will process downstream.Kevin digs into exactly how the engine handles both paths — including what triggers the ~7K token structured data analysis mode — in this week's newsletter. subscribe here.!-->

First, you can subscribe to email notifications for product updates. Head to the product updates section and hit subscribe so changes land in your inbox instead of getting buried in the feed. (Pssst this is a feature available for all forums in the community) Second, there are front-end filters so you can narrow updates down to the specific products you're building with. No more scrolling past things that don't apply to you. Small setup, big payoff if you're trying to stay current without living in the community 24/7.!--> 

Most device health issues don't start as tickets, they start as slow laptops and frustrated employees who can't figure out why their machine is struggling.One customer connected their device monitoring platform to their AI assistant to flip that script. Instead of waiting for a ticket, the AI proactively reaches out when a device is running low on disk space or needs a restart — identifies large unused apps, surfaces recommendations in chat, and lets employees take action with a single click. No IT involvement required."Employees completing device health actions conversationally — no ticket, no IT call, no back-and-forth."These agents are used 1,500+ times per month. The win here isn't just efficiency — it's removing friction before employees even realize there's a problem.We'd love to hear from you: What device management tasks would you most want to handle proactively rather than reactively? Drop your r

Hi everyone! I’m new here to sharing my introduction. Here’s a bit more about me:What I’m working on: I’m leading efforts to scale AI-driven support across multiple business units, focusing on smarter ticket routing, better self-service adoption, and faster resolution times with Moveworks.What I want to learn or discuss: Excited to hear how others are driving AI adoption in enterprise teams, especially non-technical ones, and measuring real business impact. Tips for balancing automation with a human touch are always welcome.Fun fact: When I’m not at work, I’m rock climbing, or training my dog to sit through virtual meetings, he’s now the best (not so) silent observer on the team.

Hey,Why would a plugin that returns the right data produce a wrong answer? I keep seeing the same thing: builders format their response into a clean paragraph using an LLM action, return it as text, and the assistant loses the ability to work with it.It comes down to what your output mapper returns.How the Engine Processes Your Tool ResponseWhen a plugin runs and returns data, that response passes through a processing step before the reasoning engine sees it. What the engine can do with your data depends entirely on whether it's structured JSON or flat text.Two-path comparison: structured JSON gets parsed, filtered, and routed while plaintext gets forwarded as-isStructured JSON gets parsed. The engine extracts the schema, trims irrelevant fields, and routes piec

260408-016.prod3.us -  This is my ID I can log into Okta but cannot see the manage apps within moveworks. This sends me to help and documentation page. I have already completed a few labs 0-2 and was working perfectly okay today. 

Hello,I am new to the community and getting my feet wet by starting withMoveworks Agent Studio Developer: Level 1 Certificationfor which I will need access to Agent Studio as per the instructions. Could you please guide me with the next steps to get access.Much appreciated. Thanks,Rekha

An incorrect offboarding can lock out an active contractor for hours or even days, disrupting live projects and generating urgent escalations across IT, HR, and management. It's one of those processes that seems simple until it isn't.One customer tackled this head-on by automating contractor expiry and offboarding through their AI assistant. They added real-time compliance alerts to flag potential issues before they snowballed into disruptions.The outcome: contractor lockouts essentially eliminated, and compliance risk reduced without layering on more manual oversight."Contractor lockouts essentially eliminated. Compliance risk reduced. No added manual overhead."The win here wasn't just efficiency. It was proactive risk reduction before issues ever surfaced.We'd love to hear from you: What offboarding processes would you most want to automate? And where does manual handling create

It's been here all along: a dedicated space within our community to share opportunities, find talent, and support each other's growth. If you haven't checked it out yet, now's a great time.For customers: Post consulting projects, contract roles, and full-time job openings you want to put in front of a trusted professional network. For partners & professionals: Browse opportunities for yourself or your organization — whether you're actively looking or simply keeping an eye open. For everyone: Share posted opportunities with your broader network. Help great talent and organizations find each other.It's a low-pressure, professional space, perfect for those open to new possibilities without formally "job hunting." Think of it as a warm network, not a job board.A few simple guidelines:Share relevant, professional opportunities only — consulting work, proje

The AI Assistant you already know is coming to a new home – on the web.So much work happens in the browser, and now your Assistant will be there too. We're making it effortless for employees to get the help they need, when they need it, and wherever they need it. By bringing the Assistant to the web, Moveworks is stepping right into the center of work. We're meeting employees directly in their flow, helping them stay focused, move faster and unlock even more impact from every interaction.From chat companion to the center of workToday, our AI Assistant helps employees inside tools like Teams & Slack, service portals, and intranets. But employees spend far more of their day in the browser than they do in chat. And as we evolve

Hello,I am Ganesh, and I am new to Moveworks. I have been in ServiceNow ecosystem for more that 13 years. Currently working one of elite partners of ServiceNow. Looking forward to learn, adapt and utilize the knowledge from this community.Thanks,Ganesh

Want to get smarter about managing analytics access in Moveworks? Our latest Academy course walks you through everything you need to know about data governance, roles, and Custom Views.You’ll learn how to: Structure analytics resources safely Apply the principle of least privilege Manage permissions with confidence Share insights across teams without exposing sensitive data Perfect for program owners, admins, and analysts, this course combines practical examples and best practices to help you make informed decisions about analytics access. Click here to take the course.

If your plugin starts repeating questions, forgetting what the user said, or hallucinating field values after a few turns, the config probably isn't the problem. The context window is.Every plugin has a context budget: the total information the AI can hold in working memory during a conversation. Slot descriptions, action outputs, system instructions, conversation history. They're all competing for the same finite space. And here's the thing: before your user says a single word, your plugin has already burned thousands of tokens just loading up.Every turn after that is a fight for what's left.This is why plugins that work fine in testing fall apart in real conversations. Short demos never push the budget. Real users do.The good news: most context waste comes from just two places. Overstuffed slot descriptions and action outputs returning way more data than the AI actually needs. Clean those up and your plugin can hold a 20-turn conversation without losing

Hey,Your plugin didn't break in production. It was always going to degrade after a few turns. You just never had a conversation long enough to notice.By turn three or four, the reasoner — the agentic reasoning engine that orchestrates your plugin's conversation — starts repeating questions, hallucinating field values, forgetting what the user said two messages ago. The config didn't change. The context window filled up. Before your user said a single word, your plugin had already consumed thousands of tokens in slot descriptions, system prompts, action schemas, and resolver metadata. Every turn after that is a fight for what's left.We covered the architecture fix for chained actions in Issu

Most teams launch their AI Assistant, see decent early numbers, then plateau. Sound familiar?We analyzed data across thousands of deployments and found a clear pattern: many organizations achieve good adoption but stall before reaching great adoption, because they treat it as a change management exercise rather than an organizational transformation.The top performers don't just launch and move on. Here's what they do differently.The two metrics that matterMAU% — breadth of adoption (average customer: 22%, top performers: 51%) WAU/MAU — depth of engagement, i.e. how often monthly users return weekly (average: 35%, top performers: 42%)Where you sit on these two dimensions tells you exactly what to focus on next.The three-stage playbookStage 1 — Build Your User Base If you're below 40% MAU, this is your only job. Awa

​Objective:  The objective of this design is to enable the consumption of Moveworks Large Language Model (LLM) capabilities from enterprise platforms such as ServiceNow, rather than restricting LLM usage solely to a conversational chatbot interface. This document uses ServiceNow as a reference integration platform to illustrate the approach.  Business Use Case Overview The proposed integration enables multiple high‑value business use cases designed to improve operational efficiency, service quality, and knowledge reuse across IT and business teams.The solution supports, but is not limited to, the following cases like: 1. Incident Summarization a

If your Agent Studio plugin used to respond in 2 seconds and now takes 8, you might not have a bug, you might have an architecture problem.When you stack multiple actions back-to-back in a conversation process, the AI has to stop and process everything after each one. More actions, more data, more thinking, and your user is stuck watching a spinner.The fix: either add a natural user interaction between actions (like collecting a required field), or bundle your chained actions into a single compound action that runs behind the scenes.One reasoner turn. Cleaner context. Faster plugin.📬Want more tips like this? Subscribe to the Agent Studio Developer Newsletter in Community or via email.

We recently hosted a session on how to move from AI ambition to real enterprise automation — and what it actually takes to deploy AI agents that work across the business.Here are the key takeaways 👇 🔎 The Core ChallengeEnterprise work rarely lives in one system.Even simple tasks often span:• Multiple applications• Several approvals• Manual handoffs• Status chasing and follow-upsThe result? Operational friction, slower execution, and teams stuck in low-value work. 🤖 What Makes AI Agents Actually