Product Forums

I’m facing an issue with the search functionality in the Moveworks setup when trying to check ingested files.Currently, when I type a file name in the search field, the cursor automatically jumps out of the search box. Because of this, I have to click back into the search box repeatedly to continue typing. This makes it very difficult to type the full file name, and the experience becomes quite frustrating.At the moment, the only workaround I’ve found is to copy and paste the file name directly into the search box to avoid the cursor jumping issue. Can you please help look into this or suggest if there’s a known fix.

Hey builders!We wrapped 2025’s Office Hours series yesterday with a cozy, high-signal session – I was out sick (stay safe from the flu!), so Steve Rankich stepped in as host with Ryan Geroulo riding shotgun. Even with holidays in full swing, the questions were gold – all about making your agents smarter, more reliable, and less noisy.Here’s the recap of what we unpacked:Available Data Bank Pop-Up – Your New DSL Best Friend  → No more hunting through actions to remember output keys! The new “Available Data” modal shows exactly what you can reference: action outputs, slots, and the all the user fields (time zone, employee ID, etc.). → Drag-and-drop or copy-paste – instant data.user.time_zone wins. Huge quality-of-life upgrade for chaining actions cleanly. GPT-5.2 Now Live in LLM Actions → Fresh model support dropped: GPT-5.2 (pinned & unpinned), plus 5-Mini/Nano/04 flavors.

Hi, I was looking into the Agent Marketplace use case for Workday – Notify on Time Off Request Approval. I got stuck at a point where we need to fetch event steps using the URL/ccx/api/businessProcess/v1/{tenant}/eventStepsThis endpoint requires parameters such as businessProcess and stepType, which are mentioned as IDs in the documentation. Could you please clarify from where we can fetch these IDs for our tenant/environment?

Hey Team,We want to know if there’s a way to disable the ability for users to block messages that come from the Employee Communications module? Those are critical messages and while we can understand people may want to block the general ticketing messages but we need to make sure those are sent to people.Anyone else experience this, or find a way to get through, or should we raise this as a feature request?

Hello Moveworks Community,We’re excited to share an update on the timeline for the new improvements to our SIEM logs pipeline, designed to be more streamlined, secure, and structured.The new log pipeline is available now for self-service. Please refer to this help documentation to learn more about the enhancements, the new log structure, and any updates required for custom workflows that extract data from SFTP to your data store. Please start migrating your workflows to follow the new log structure.The current SIEM log export will remain available un

Our current manual process is as follows: Step 1: create user accountStep 2: license the email account with a G1 licenseStep 3: convert the mailbox to a shared mailbox Hoping somebody may have done this or used a combination of moveworks and powershell scripting.  

Hey everyone,We're thrilled to announce a significant enhancement to how we track feedback telemetry in your analytics dashboard.We heard from many of you that the feedback counts in your dashboards were low. You told us this made it challenging to get a full picture of employee sentiment, potentially causing you to miss key opportunities for improvement. What's Changing?To give you a truer and more immediate signal, we are upgrading our feedback tracking. Previously: We counted feedback after a user clicked thumbs up/down and submitted the follow-up form. Now: We are capturing every single thumbs-up and thumbs-down click as a feedback event. This change ensures that even the quickest, most immediate reactions from your employees are counted, giving you a much richer dataset to work with. The Impact

Hi, I am encountering API Not found Error while accessing DATA API to extract Interaction Data from Moveworks.  I have passed the Token after authenticating as mentioned in the help doc. Please guide me

Hey everyone, I have a question for the group that I’m hoping someone may have experience with. We’re planning a governance overhaul this coming year, which will likely include implementing an identity governance (IG) system such as Okta OIG, Lumos, or Zluri IG. Has anyone successfully done this or currently have a setup where Moveworks integrates with an IG platform? Our current plan is to keep Moveworks as the entry point for all software access requests, but then hand off the backend processing and approval workflow to OIG (or whichever IG tool we adopt). Okta and similar platforms offer more advanced capabilities—like time-based access and improved audit controls—which we need for compliance, upcoming security certifications, and more robust user access reporting. If anyone has insights, lessons learned, or is open to sharing how they approached this, we’d love to connect!

I have been getting error since this morning while testing a compound action: "AttributeError('link_context_infos')"This was working perfectly fine until Thursday night before I logged off for the week; nothing changed since then but running into this weird error. I tried to build from scratch but still same issue.  any insights are greatly appreciated. 

I’m the Sr. Technical Content Writer on my team and was hoping to use Knowledge Studio to help identify knowledge gaps so that I could proactively update and create new material. However, I’m not finding the suggestions helpful at all. When I open the suggestions, I make sure to open every associated ticket. If present, I also open the referenced existing documentation. I then recreate the prompts in the bot and every single time I’ve received a perfect answer. Knowledge Studio doesn’t seem to be finding any real gaps. Additionally, the text it suggests is often completely unusable. Most of the time, the suggested content says to refer to existing documentation. 😐 EXI has been proven more useful, specifically in APPAS AND SERVICES DETAIL  > Generate Insights. I wish the two tools could actually be combined. 

Referencing the March 2025 Product Release Notes and implementing the following compound action:steps:  - action:      action_name: mw.get_user_by_email      output_key: output_user_to_notify      input_args:        user_email: data.email      delay_config:        seconds: data.delay_in_seconds  - notify:      output_key: output_test      recipient_id: data.output_user_to_notify.user.id      message: data.message We cannot suppress the message after the reminder notification is sent.  The message states: “you’ll receive: <message content>” after the no

🎉 Triage 2.0 is now General Available (GA), available in Moveworks Setup What is Triage?For issues that cannot be resolved instantly, Moveworks Triage can categorize and route these tickets to ensure that they reach the service desk team best equipped to resolve the issue. Triage intercepts newly created tickets that passes a filter, configured by the customer. Then, Triage will attempt to classify the configured ticket fields, e.g. such as assignment group, category, subcategory, cmdb_ci, etc by retrieving the top 10 most similar tickets and select the most common field value (assignment group = WW Manufacturing). If the prediction confidence level passes the threshold, Triage will automatically update the ticket - all within 1 minute of the ticket being created. What changed from Triage 1.0?Triage 2.0 is a dramatic improvement from its predecessor (Triage 1.0) as its architecture is

Hey Agent builders!Session #3 dropped today and we were cooking – we dove into fresh releases, live demos, and those "aha!" unblocks that make building feel like magic.Quick hits from the hour: List Slots – No more prompt gymnastics → Tick the "Data type is expected to be a list" checkbox and voila: collect arrays of strings, ints, booleans, or custom types without hand-holding the assistant. → We did a live demo showcasing the power of this new feature! Dynamic Resolver Context Passing – Databank to Resolver, Seamlessly → Feed data bank outputs (e.g., activity results from your data bank) directly into resolvers without brittle input args. → We also did few live demos of this new feature! Taming Giant Payloads: Data Mappers + DSL Filter Magic → API spewing massive JSON? Slice it down with output mappers and DSL functions like $FILTER – trim n

The first red box is a script activity that will return a list of systems the user has a locked account in.The second red box I need to run against that list, up to 4 times. So, if a user has a locked account in 2 different systems, I need to run the unlock action for each one. Does anyone have any ideas how I might achieve this?

We are using OAuth 2.0 with Authorisation Grant in our Moveworks plugin integration. On first authorization, we successfully receive both an access token and a refresh token. However, the access token expires after 24 hours, and when users query the plugin after that, the plugin directly fails instead of prompting the user to re-authorize.Expected behaviour :  When the access token expires, the plugin should use the refresh token to fetch a new access token silently. If that also fails (e.g., refresh token expired), it should prompt the user to re-authorize.Current behaviour :  The plugin fails immediately on expired access token without asking users for re-authorization, resulting in poor user experience.Request:  Could you please advise the best practice or configuration in Moveworks to handle access token expiration and ensure the plugin prompts users for rea

Hi ​@Ajay Merchia  Are the other two content type activities disabled?Knowledge and Form Referencing the 

Hello Community!I wanted to share a useful architecture for managing Agent Studio Notifications that allows end-users to opt-in or opt-out of specific proactive messages.The Solution: A Moveworks Notifications TableTo support custom user notifications, this strategy involves provisioning a Moveworks Notifications Table (e.g., in ServiceNow or any iPaas tool) dedicated to housing user notification settings. This table tracks user preferences for specific proactive use case notifications (e.g., approval reminders, training due reminders, direct report birthday reminders). Any proactive message sent—whether triggered by a Moveworks Scheduled Job, a ServiceNow Flow Designer Flow, an Azure Function, or any other system—has the ability to cross-check this Moveworks Notifications Table. The message is only sent if the user has actively

Does Movework allows to create and access a datastore, this data store can be - SQL, NOSQL or VECTOR STORE. This would allow users to store and access data from persistent store which is indexed.Use case: If we refresh CMDB and pick CI directly from MW Platform and doesnt have to make any API call to ServiceNow or other ITSM platform to fetch CI Data.

We are excited to share an upcoming improvement in search quality: Moveworks AI Assistant will now be able to better answer questions from long KB articles with the help of an improved chunking algorithm that handles long articles better. This change will be rolled out gradually to all customers and expected to available for all customers in all regions by Thursday, December 18. What problem are we addressing?We are improving search performance on documents that contain very long chunks, which are Moveworks-extracted sub-parts of the article.“Chunking” is an important part of processing KB articles or web format documents (it is also sometimes referred to as “snippetization” in Moveworks terminology), which breaks down every article into smaller parts when it is ingested.Every chunk is converted into an embedding, which is a semantic representation of the text that captu

We’re using Live Agent Message Brokering and I’m trying to understand how often users are using the translation feature.I know users can set their preferred language before starting a live agent chat, and I also know I can filter to see how many users are starting live chats overall. What I’m looking for is something more direct and easy to consume:A way to quickly see how many users are using Live Agent Message Brokering with a language set to something other than English, without having to do a lot of manual filtering, coding, or custom calculations.Ideally, I’d like a dashboard or out‑of‑the‑box report where I can see something like:“10 users had their language set to Italian when using Live Agent Message Brokering” “8 users had their language set to German when using Live Agent Message Brokering” “5 users had their language set to Spanish when using Live Agent Message Brokering”In other w

Hello, We are facing an issue with the MW generic approval template(mw.create_generic_approval_request). We are able to build the approval request successfully, but when we trigger it to the user, the notification contains the header as approval_details. Whatever we pass as “approval_details” appears as the header. This is fine when we add one or two lines of approval details, but when we add a list of information for approval, it looks bad for the end user and it’s duplication valuesIs there any way to eliminate or truncate these details in the header?  

Sharing an important update for everyone using the Snowflake Dynamic Query / Snowflake Cortex Analyst plugin.Snowflake has announced that they are deprecating single-factor password sign-ins, including for service accounts. Once this enforcement begins, password-only authentication will stop working, and Snowflake will require MFA or non-password mechanisms for programmatic access.Reference:🔗 https://docs.snowflake.com/en/user-guide/security-mfa-rollout  Our Snowflake connectors today are set up using username + password. If these connectors remain unchanged, they may start failing with authentication errors when Snowflake completes this rollout. Snowflake recommends migrating to one of the supported authentication methods for automated integrations:Key-pair authentication (public/private key)

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

I’m working on a use case to ingest a Google Drive that requires profile boosting. I see in this documentation clear guidelines for how to label and thereby boost geographic specific information based on user profile location.  Does this hold true for both MIME types of content, articles and files? Context for what an article vs file is taken from this documentation:Knowledge Base Articles (KBA): HTML-content such as .aspx pages & .md/.mdx (markdown) Files: Such as .pdf, .ppt, .doc, & .txt