Developers Forum

Hello,I noticed that there is not a place to register in advance for sessions for the pathfinder summit. Does registration open soon? Will that information be communicated soon?Thank you

Curious if anyone has experimented with AARI forms of variable size. To be specific, I’m considering having an AARI form generated partway through an automation that houses a table of information, but the table size (num rows) will be determined by the automation at runtime. It seems like as I create the AARI form, I have to create a pre-determined size of the table field.Curious if anyone has experimented with tables in forms whose size is generated dynamically. Or if not an AARI table, some other sort of data table of variable size that acts as a human in the loop element that an end user can interact with mid automation. Perhaps a CSV would work that a user could move into a “validated” folder, but it would be nice to have some control over the values that they can input into cells.

I am facing an issue with Automation Anywhere Document Automation where a document repeatedly goes back to the Validation stage.The document is processed and sent for validation, but when I open it in the Validation Station, it shows “0 Fields to Validate.” Even though there are no fields requiring validation, the document keeps returning to the validation queue instead of completing the process.Current behavior: Document is processed by Document Automation. It is sent to the Validation stage. Validation shows “0 Fields to Validate.” After completing/submitting the validation, the same document again appears for validation. This cycle keeps repeating. Expected behavior:If there are 0 fields requiring validation, the document should complete the validation stage and proceed to the next stage instead of returning to validation.Could you please advise what could cause this behavior and how we can prevent the document from repeatedly entering validation?

Hi everyone,We have bots in production that have run for several months without any particular deployment problem On this week-end for half a day we had a systematic deployment error “An unexpected problem occurred” on both production VMs that we can’t explain. There is no information about updates or maintenance on AWS side. Moreover on a third VM not linked with that project, bots ran correctly.The error solved itself alone and at the same time bots started running correctly on both VMs.It was during the week-end and to my knowledge nobody did connection, admin action or maintenance on the VMsDo you have an idea on what could be the cause of this problem? 

Cannot add a dictionary (any) to a list variable (any). Getting an exception that says “Data type of to-be-added item is different from the data type of List” Does anyone know of a workaround? Thank you

We have two unattended Bot Runner accounts running on the same Windows Server 2022 RDP multi-user machine, with separate Windows sessions and RDS CALs. Automation works normally under Bot Runner 1 but Recorder/DOMX browser interactions consistently fail under Bot Runner 2Environment (identical for both accounts): OS: Windows Server 2022 Datacenter Azure Edition Chrome: 155.0.8059.40 (64-bit) Automation 360 Chrome extension: 4.5.0.0 Extension ID: bboccnojhhoegfffpcllbeigipedjedf Both accounts use separate RDP sessions on the same machine. Symptoms: Bots launch successfully and Chrome opens under both accounts. On Bot Runner 2, Recorder/DOMX consistently errors, saying it cannot find various elements but the error is often received after the element is successfully located and filled/clicked, but subsequently reports that the element cannot be found. Identical automation works normally under Bot Runner 1 on the same multi-user RDP server. Elements appear to flicker or be repe

Hello,  I am building an automation that processes different transfers in a Workday report.  The report lives in Workday and if a record in this report needs to be skipped I want it to go to the next record.  I currently have it set up for a Loop based on the total records in the report.  Has anyone build an automation that would skip various records if needed.  Thanks! 

Hi ​@Aaron Gleason I’ve purchased the Summit ticket, what next.. When will I get the certification vochure ? Receive an email that “Supply Drop” Agentic Arcade is open.   But It has nothing.  Pls guide, Is there any special hands-on exercises we can do before the actual summit. 

Hi guys,Im tying to run a parent bot that calls several child bot in a VM. but when the parent bot get half way it just stop. Im getting There is no active user session available for user for the bot to run error in the control room. If im running the child bot individually it run without this issue. Any idea what is causing this and how to fix this

Using OCR Capture (area) to assure a page has updated properly. I am getting an error: This may be due to the following reason: Failed to do OCR - 'Exception occurred while OCR Engine initialization with message Unspecified error '.To continue, edit the bot and fix the error. Then, try again. If you continue to see this message, please contact your system administrator.Code: ba.bot.execution.errorI can copy the OCR line out of my bot and use it in a test bot with just that line and the OCR runs without issue. I can also create a new test bot and do an OCR capture on this page/area with no errors. What would cause this to fail this way, but run all by itself?  

I am trying to find the string “tax invoice” on documents, but I need to make the regex case insensitive. Google keeps telling me to use (?i)tax invoice but the TEST REGEX PATTERN does not seem to accept this syntax.  I am hoping that someone can provide some guidance on the support regex syntax library. Thanks in advance, Cameron 

Hi,We are receiving this error in one of our instances. The same bot works just fine in other VMs. This seems to be the Database Package executable. Any insight on the root cause and the fix is appreciated.

Hi everyone - I’m building an automation to read values in a column using the terminal emulator. The screen I’m reading is 131 characters across. I’ve figured out how to increase the visible width of the terminal emulator using some of the terminal emulator settings, but any attempt to use GetText with the entire screen or any character above column 80 results in an error. I’ve seen various fixed suggested elsewhere about turning on extended properties, but that results in my terminal not connecting at all. Has anyone else run into this situation? TIA

I'm trying to use the Microsoft 365 Excel - Connect action with a Control Room managed OAuth 2.0 connection in Automation Anywhere Cloud.EnvironmentOAuth Authentication Mode: Control Room managed Grant Type: Authorization Code Flow Shared token successfully generated Delegated Microsoft Graph permissions granted and admin consented: User.Read Files.ReadWrite.All Sites.ReadWrite.All Mail.ReadWrite Mail.Send offline_access openid profile What worksOAuth connection tests successfully in Control Room. Microsoft 365 Outlook Connect works successfully using the same connection and account. The service account can access Outlook, SharePoint and OneDrive directly through Microsoft 365 in a browser on virtual machineWhat doesn't workThe following actions fail immediately on the Connect action:Microsoft 365 Excel: Connect Microsoft 365 OneDrive: Connect Microsoft 365 Calendar: ConnectError:We are unable to authorize the application with provided credentials. Please check your credentials

Hi Everyone, I am unable to create any custom agent from ai.automationanywherre.com. It is showing below error. 

I’m trying to connect the bots to AAI Enterprise Knowledge using Enterprise Knowledge package but getting the below error messagenote : it is on premise and I’m sure about API Key and secret Key     

I could not able to run the bot. please help me. The bot agent is connected as it is howing the green tick in the below, only the issue is whenever i am trying to run the bot it is showing ‘user not logged into the machine’. Please help asap. I need to test the bot.   

Hello , We have a bot that logs into a web site and clicks various objects  (calendar picker, values from drop down lists, etc). Every so often one or more of the object capture actions fails and i have to recapture them especially after a google chrome update.  I was wondering if maybe i’m just not selecting the right attributes. I recall there was a document that listed the best attributes to use and to de-select some of the ones the recorder automatically selects for me. For example, I leave the Dom Path selected but unselect the Path attribute.Can anyone help me with this. I tried the AA document link but i keep getting an error stating it is unavailable or has moved.

Hallo,is the Outlook application presence necessary on BOT machine when using e-mail:connect Email server or EWS action?greetings MT

Recently, our Automation Anywhere Recorder Capture actions started failing right after a browser update. At first glance, this looked like an Automation Anywhere platform issue, especially because the functional team was not facing any problems with the application. Everything worked fine for them, but automation was failing.This naturally leads people to think something is wrong with AA. But in reality, it wasn’t an Automation Anywhere issue at all. 🌐 What Went WrongAfter the browser auto update to a newer version, the web application we were automating simply did not work as expected. Elements behaved differently, certain interactions failed, and the overall UI didn’t respond the way it normally does.Because Recorder Capture relies on the web application’s behavior, it also started failing. Not due to an AA issue, but because the application itself wasn’t compatible with the updated browser. 🔍 How We Identified the Root CauseWe compared browser versions across machines: Machines wi

Hi everyone,I am currently using Automation Anywhere Community Edition (Free version) and I am trying to learn how to send an email from an Outlook/Microsoft 365 email address using Automation Anywhere.I would appreciate a step-by-step explanation of the recommended way to configure this in the Community Edition, because I am not completely sure whether I should use the Microsoft 365 Outlook package or the Email package.1. Microsoft 365 Outlook packageI found the native Microsoft 365 Outlook package, which seems to be the recommended option for Microsoft 365.I understand that I first need to use the Connect action.However, when configuring it, I see several options that I am not completely sure how to configure: Azure Cloud: Azure Global Azure US GCC High OAuth2 Authentication mode: Authorization Code with PKCE Client Credentials Control Room managed Client ID Tenant ID Redirect URI Username I understand that these settings are related to Microsoft Entr

I’m planning to integrate an Ollama LLM with Automation Anywhere using the REST Web Service action to automatically analyze bot failure logs.The goal is to group and categorize errors such as: Credential / access issues Application not responding Selector or object change Data format mismatch Environment or profile corruption Network / timeout errors Excel/Office activation popups Unexpected UI popups Once the logs are classified, we can automate the production support workflow, including: Auto‑creating PRD support tickets only when required Routing tickets to the correct support queue based on error category Providing recommended fixes directly inside the ticket Reducing manual triage time for support teams I’m curious if anyone here has used AI models (Ollama, OpenAI, Claude, etc.) with Automation Anywhere for similar production support automation or log intelligence.Would love to hear your use cases, lessons learned, and model choices.Tagging for visibility:  @De

We’ve had a few bots running for several months now that use many Recorder: Capture actions to extract info from and input info into certain websites. For months, they’ve been running fine daily or multiple times per day without issue. As of Saturday 9/26, they've been failing.The cause of the failure is that Recorder: Capture steps are failing. They are failing inconsistently, and I can’t identify any pattern as to why this is occurring. If I rerun the same bot multiple times, it might fail at a different Capture step each time, but the point of failure is always a Capture step.  If the bot fails during its run, I’ll get a “unable to find CLIENT. Search criteria did not match” error. Sometimes, the bot won’t stop at the error - it will just fail to get a property that it was supposed to capture and cause issues downstream (e.g. a variable it was supposed to capture will be empty, so it will think it failed to input a particular value).  This happens whether the bot is being run attend

Working in a setup where the bots are running in unattended mode on runners. The bots are either scheduled specific time of the day, if they need input, they run based on new emails. The business user does not access Control room in any way, they either raise tickets or send emails. Want to know how FORMS work in prd? Let's say if a process needs a form input in an intermediate step, how will the process be implemented in setup like mine? Will those processes have to be run locally? Can we use webhooks which can take approval over teams, but then will it keep the runner engaged? 

I am facing issue in the production bots relating to recorder package after the chrome updates. It is failing abruptly although there is no change on the web application. Is anybody facing the same issue ? What measures have you taken to ensure the bot run.