Hi!
I work in a product team where we use JIRA to document our software requirements. Due to regulatory standards, our requirements must follow a specific format in both language and structure.
Since our team consists of five people, each with their own writing style, I developed a small agent that takes a JIRA link as input, refines the content to match our standards, and if the user approves, automatically updates the JIRA work item with the improved version.
The agent performs well when refining text, tables, anchor links, and overall structure. However, I’m running into issues with media elements (e.g., images etc.).
Problem
Media content is not preserved between the input and output. The agent seems unable to retain or correctly reposition media within the refined requirement. This breaks the flow and completeness of the documentation. The attached screenshot below shows how the image presented in Rovo after the refinement. The same image is then posted to the updated JIRA work item.
Question
Is this a known limitation of Rovo? Or is there a specific way to instruct the agent to handle media correctly, perhaps a formatting convention or metadata tag I’m missing?
Any guidance or suggestions would be greatly appreciated!
