I've been testing Rovo in my Premium organisation lately. I've noticed that queries about Confluence Databases and Whiteboards do not get reliable answers. Are these not supported by Rovo as a knowledge source?
Hi @Israel Padilla
before going further, could you share the exact use case you’re trying to solve? For example:
A quick summary to ground the follow-up: Rovo can ingest Confluence content when you give an agent a space or page as a knowledge source—so if your database lives on a page or in a space, that content can be included.
Whiteboards aren’t currently a generic “selectable” knowledge source for arbitrary agents in the same way, but there are related capabilities: you can summarize a whiteboard using Atlassian Intelligence, and the Brainstorm Facilitator agent works directly with whiteboards for ideation.
If you need a regular agent to “see” whiteboard content, a common workaround is to copy or export key parts of the whiteboard into a standard Confluence page and then add that page as the agent’s custom knowledge.
With your specific goal, I can tailor the next step more precisely.
— Mia Tamm from Simpleasyty
Thanks for your reply, @Mia Tamm
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hi @Israel Padilla ,
Thanks so much for the follow-up — that context really helps.
For the Confluence database part: you're right — Rovo currently struggles to understand structured content like databases or tables as queryable entities. Even if the database lives on a Confluence page, agents can’t yet interpret the schema or perform operations like counting rows or filtering entries (e.g., “how many lessons?”). A possible workaround is to summarize the content in plain text below or above the database block — some teams create a daily or weekly summary in list form to help the agent pick up the key data.
For the whiteboards: since your use case is exploratory (e.g., product ideation), the Brainstorm Facilitator agent might help more than regular agents. However, it's not yet able to interact with specific branches or sticky notes unless the content is copied to a regular Confluence page. I’d suggest exporting a summary of the “opportunity solution tree” and linking it to a Rovo agent trained with that context.
Let me know if you’d like a sample structure to help prep content for Rovo.
— Mia Tamm from Simpleasyty
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.