I am building an AI-based resume screening/quotation analysis solution using UiPath Cloud Studio, Storage Buckets, GenAI Activities, and Content Generation.
My use case is as follows:
Users upload a Job Description PDF and multiple Vendor/Resume PDFs into Storage Buckets through a UiPath App.
In the Cloud Studio workflow, I need to retrieve all the files dynamically from the Storage Bucket, download them, extract their text content, and then use Context Grounding + Content Generation to generate an analysis/comparison report.
The challenge I am facing is that the uploaded files have dynamic names, so I cannot hardcode the file path in the Download Storage File activity.
Could you please guide me on the recommended approach in Cloud Studio for:
Listing all files available in a Storage Bucket at runtime.
Iterating through those files dynamically.
Downloading each file without specifying a fixed path.
Passing the extracted content to GenAI activities for analysis.
Is there a standard pattern or best practice for handling this scenario in UiPath Cloud Studio workflows?
Use List Storage Files activity to get all the files into a collection of files.
Use For Each activity to loop through files collection retrieved in step 1
Inside for each to process the file data (like reading an Excel or PDF file), place a Download Storage File Activity inside the loop to save the file locally first, using currentItem.Filepath as the source path.
Use Read PDF Text activity to extract downloaded pdf text and then use Content Generation GenAI activity for the analysis and further steps.
Use List Storage Files to fetch all files from the bucket at runtime
Loop through them using a For Each activity
Inside the loop, use Download Storage File and pass item.FullPath as the File Key (this handles dynamic names automatically)
Extract the text from each file (Read PDF / Document Understanding) and store it in a collection like a Dictionary
Optionally, classify files (JD vs resumes) either via naming convention or a small GenAI classification step
Finally, pass the JD + all resume contents into Context Grounding + Content Generation to generate your comparison output
If you’re dealing with many resumes, you might also want to summarize each one first before sending it to GenAI ,it helps with token limits and improves performance.
I have used List Storage Files to fetch all files from the bucket at runtime—
but I’m not able to get the pdfs present in the bucket .
I’m not getting list of pdf after using message box to
I have used List Storage Files to fetch all files from the bucket at runtime—
but I’m not able to get the pdfs present in the bucket .
I’m not getting list of pdf after using message box to