AI and Agentic Automation

Hi, overview : 1. I am sending a email to process model and store the data's in pv 2. With the use of (Create Eml File from OOTB Poller) smart service generating a document in EML format 3. In classify email AI - passing the generated document as input. I am getting an error saying " The user does not have permission to view document " in classify email All objects(AI,process model,document folder) have Admin access Also the generated doc from step 2 is showing "not in the application" but it is in document folder within the application Pla suggest a solution to resolve this. Thanks in Advance

Hello Everyone ,  so i was having query regarding Doc Extraction skill's , anyone is having idea that how we can achieve Checkbox field extraction without having Reconcile DOC task ? because of this issue i have to look for AWS services for the same , it would be so useful if checkbox and table field we can extract automatic from Appian itself without having reconcile task! 

I have this array {100, 50, 20,50,10} and a variable x = 600, I need help to create a rule that creates an array with the results that must be calculated in this way: The first value of the array must be subtracted with the variable x, the result must be subtracted with the second value of the array, the previous result must be subtracted with the third position and so on. The results must be stored in an array. At the moment I have this code but it does not work correctly a!localVariables( local!regularizaciones: {100, 50, 20,50,10}, local!initialValue: 600, local!resultArray: {local!initialValue}, a!forEach( items: a!forEach( items: local!array, expression: fv!item ), expression: a!localVariables( local!currentValue: fv!item, local!lastResult: index(local!resultArray, length(local!resultArray)), local!newResult: local!lastResult - local!currentValue, local!resultArray: append(local!resultArray, local!newResult) ) ), local!resultArray ) I appreciate any help

Hi everyone,I need to control the font size in an email whose body is generated by an expression rule that generates the corresponding html. The problem is that the font size that I specify in the html does not correspond to what the email finally has. For exmaple, this HTML <p style='font-size: 14px;'> Hola, la solicitud se ha completado y puede generarse el acuerdo en Vistex (adjunto fichero de carga)<p> should be 14 px. However in the emali the font-size is 10.5px.

Hello  I uploaded 10 pdfs, did mark the labeling then submitted for training how long does it take before i can publish, is it 30mins ?  Is there some way to calculate this time 10 docs, 50 fields each page one page doc, structured pdf , ex:insurance Acord form

{ 'a': true, 'b': true, 'c': false, 'd': false } The code above should dynamically be converted to  { { 'a': true }, { 'b': true }, { 'c': false }, { 'd': false } } is that possible? Thanks in advance

Hi All, I am trying to fetch data from Oracle DB using a query rule with batch size as -1, but the system is throwing the following error : Expression evaluation error at function a!queryEntity [line 2]: An error occurred while retrieving the data. Details: Unexpected error executing query (type: [DataTypeName], query: [queryentity expression], order by: [[Sort[primaryKey asc]]], filters:[((Ts= TypedValue[it=9,v=3000-01-01 00:00:00.0]) AND (Valid= TypedValue[it=3,v=Y]))]) It had data of 147 rows and if I use batch size of 100 with start index 1 it's sometimes working but when I use startindex as 101 it's always failing with the above error.I am facing the same issue with some more tables from the same Oracle schema.Here I am attachinga!queryEntity( entity: cons!DSE_TABLE_NAME, query: a!query( pagingInfo: a!pagingInfo(startIndex: 1, batchSize: -1, ) ), fetchTotalCount: true ) the sample Code I have used.

Which types of objects can be called using the "rule domain" prefix? A.   Connected systems, records, expression rules, and interfaces B.  Decisions, expression rules, interfaces, and integrations, C.   Process models, integrations, expression rules, and constants

Hi all, I've been attempting to create a simple AI skill for document extraction on 23.4. I created a new AI skill with basic security, hit "Create First Model" and started going through the steps. First, I uploaded 25 sample documents that fit the Appian requirements (must be .pdf, must be smaller than 7mb individually, must be unique). Once the upload was complete, I proceeded to the next step "Add Fields to Extract." From there on out I will randomly get an Error 403 - Access Denied. It can happen while editing a new field name, sometimes it happens once I start labeling the documents (if I even get that far). I tried scanning the tomcat log file for info but couldn't find anything related to AI Skills in there. Anyone know what's going on? As far as I know there shouldn't be a special privilege group for editing AI skills or I should have seen an Appian error before that, right?

I have a datetime field in DB. When user selects datetime from interface, i want to make sure that date doesn't already exist in DB. Basically, i am trying to fetch duplicate requests for a day.  I cannot use "=" operator since field is of datetime type not date and if i use other operators such as startswith, in, includes , it gives me error i.e. Cannot apply operator [INCLUDES] to field [dateTime] when comparing to value [TypedValue[it=9,v=2024-07-05 05:00:00.0]].

Hi experts, We have been able to establish connectivity to a HANA database successfully. The HANA stored procedure is expecting the values to be passed, enclosed in single quotes - 'Value1, Value2'. A query like below works perfectly fine. SELECT * from DB_SCHEMA.VIEW_NAME('placeholder' = ('$$list_name$$','Value1,Value2')) However when I try to pass this list as a parameter, I am getting an error. SELECT * from DB_SCHEMA.VIEW_NAME('placeholder' = ('$$list_name$$',{parameter})) Has anybody integrated with HANA DB before and has a reference to how we can achieve this by passing dynamic values via parameter with the main criteria being include parameter values in single quotes?

Hey Guys, We are trying to use recordsChatField component in the interface but it is giving the below error. Any help will be appreciated.

Hi All, I have recently utilized the AI skill where I have used it as Document Extraction. Once the Extraction is done I am able to publish the AI skill and utilize it in my Process Model. I am also able to extract the data. I now used the Reconciling SMart service and able to Map certain fields which are not picked by the AI skill. Now How do i train my Model to pick those fields going forward for the Next set of Documents.? Basically How do I retrain my Model. Thanks  Harsha [mention:a11f77a729fb4db2af76b09948583328:e9ed411860ed4f2ba0265705b8793d05] [mention:31db00b785704865b14e5d9b9bef3863:e9ed411860ed4f2ba0265705b8793d05]  Really appreciate your help in this.

I would need to run two wherecontains within this dictionary array, for example to get a country with a POPULATION greater than 1,000,000 and an AREA greater than 150. how can I proceed with two wherecontains? thanks

Hello all, I have a question that I think it should be easy but I cannot figure it out, and also I cannot see any response in this forum. If I have a a!forEach within another a!forEach, is there a way of refering to the items of the outer loop, within the first loop? Example to understand with non real data:  I have two lists: local!text: {"1111 - GREAT", "1000 - BAD", "5600 - MEDIUM"} local!words: {"GREAT", "MEDIUM"} as you can see, "text" is a list of text, and each string contains a code a "-" and a word. I want to create a nested loop that returns a list with the codes only when the words contained in the list "words" are in the list a.In this particular example: {"GREAT", "MEDIUM"} because "GREAT" is included in "1000 - GREAT" and "MEDIUM" is included in "5600 - MEDIUM". I woud like to iterate though every item of list &quo

Hello All, Is there a way to pass system message to Appian Records chat to set the tone to the interactions? To be more specific, I want to display a value as chat response that is dependent on a column of the record type instead of displaying the exact value from the column. I am able to achieve it using prompts, but I cannot expect users to write this prompt every time they interact with the bot. 

I have a field in my form named as description. The description contains value of paragraphs spanning over a couple of pages. While training the model, I selected the data to be extracted from multiple pages by selecting the continue drawing option, but when the model was put into practice, it fetched the data from only one page. Even in the Reconciliation Task, I cannot select data from more than one page. Is there any workaround available for this issue?

Hello Everyone, I am using text() or any  function in type!{urn:com:appian:types}Test'(). But it seems not working. Can anyone suggest how I can do this? Thanks

Hi Team,  I have a used case, where an employee goes to a specific option on a website and clicks on the upload button, as we know once the upload is clicked file explorer pops up, and then he chooses a document > and clicks on submit. I am trying to automate this so that robot can pick documents one by one and submits them. I know clicking buttons is easy with interactive elements of the browser but once the Upload button is clicked the file explorer is opened, and I am not finding any way that robot can interact with File Explorer and selects that document from some location. I looked at the documentation of File system action https://docs.appian.com/suite/help/20.3/rpa-8.0/modules/file-management-module.html and I cannot find anything relevant so please help. Best, Khaiser

II want the result of this expression rule to be in the below format or the highlighted format in the editor. currently I am passing name and comment from one CDT ,now along with this 2 fields  I need to pass the colleagues array with add array with in the colleague array. I have created other new  CDT for colleague array with id , typeid and addcolleague array in it ,how to call it in the expression rule to get the result in the below format . Please adviseHow can I get this result in the below format ,can someone please adviseExpected RESULTa!toJson ({ "name": "string","Comment": "string",Colleagues:{ {"TypeId": 1,addColleague: {"email": "string","displayName": "String","eFlag": false,"dFlag": false,"ProviderID": 2,"idFromProvider": "string"   }}}})

Hi Team,               In my application i am using one expression rule that fetches the data from database view,and that database view fetch the data from 4 tables and each table have different number of records(646183,2185098 ,159583,166000) , however when i go to database and open the view it's working fine but when comes to expression rule it's not working  Note: Previously we didn't face any issue as we didn't  had much records  and it was working well for load testing we inserting lakh's of records now it's not working. Could you please suggest me a solution. Thanks & Regards, Prasad.

We had a decimal value in text format in local!value variable as below But converting it to decimal is rounding off the value as below But I want the exact value in decimal data type as we need to send this value to an external system that doesnot accept text format. Is ther any way that I can sent exact value in decimal format without rounding it.

Hi Experts, I need your help! I trained an AI skill to extract data from PDFs and the model is trained with 91%recall. In the process model I am getting an error: This operation is not supported in Extract from Document node. Need your help in this regard. Unable to identify the cause of the error.

If it's CDT rule input, we can give "{}" simply to provide test values. For recordType! rule input, how to provide value?

So I was practicing new skill of document Extraction , but I am stuck at one place while I am trying to run a script task for code below  so I am targeting value to isValidate variable as my all values is true then too isValidate Variable is showing false and I checked my XOR decision too but I guess something is wrong with Script Task can anyone pls suggest me best Practice For this [emoticon:c4563cd7d5574777a71c318021cbbcc8] and I am using Cdt