Forums

Articles
Create
cancel
Showing results for 
Search instead for 
Did you mean: 

Anyone Else’s Automation Break Since the New Rovo Model?

Patricia Modispacher _K15t_
Community Champion
April 7, 2026

 

I’m running into a strange issue with a Confluence automation that uses a Rovo Agent, and I’m curious whether anyone else has seen the same behavior.

I built an automation rule that works like this:

  • a page gets a label (seo-check)
  • the rule triggers
  • a Rovo Agent is used to audit the page
  • the result should be posted as a page comment
  • the label is then removed

This setup used to work for me.

Since switching the agent to the new model, the automation no longer completes properly. In the audit log, the runs stay stuck on WAITING under In Progress and never finish.

What I already checked:

  • the rule structure itself looks fine
  • {{agentResponse}} is correctly used in the comment step
  • I tested with my original prompt
  • I also tested with a much shorter and simpler prompt
  • same result: still stuck on WAITING

What’s even stranger is that switching back to the old model did not seem to fix it either.

So now I’m wondering:

  • has anyone else seen Confluence automations with Use agent getting stuck on WAITING?
  • did this start for anyone else after the new model rollout?
  • is this a known issue with Rovo Agents in automation?

Would really appreciate any experiences or ideas. 

2 answers

1 accepted

1 vote
Answer accepted
Arkadiusz Wroblewski
Rising Star
Rising Star
Rising Stars are recognized for providing high-quality answers to other users. Rising Stars receive a certificate of achievement and are on the path to becoming Community Champions.
April 9, 2026

@Patricia Modispacher _K15t_  @Tomislav Tobijas 

I still believe it will take quite some time before agents behave in a truly predictable and reliable way. That is probably my biggest concern at the moment. The core issue is not only capability, but consistency: the results are still not stable enough from one case to another, which makes it difficult to fully trust them in more critical or repeatable Scenarios.

Patricia Modispacher _K15t_
Community Champion
April 9, 2026

Yes, you're right, @Arkadiusz Wroblewski ! Feel that as well ...

2 votes
Tomislav Tobijas
Community Champion
April 8, 2026

Hi @Patricia Modispacher _K15t_ ,

I've actually experienced the opposite right now.

2026-04-08_09-11-01.png

We have an agent within automation that posts weekly highlights in the Teams channel and up until today, everything worked perfectly (on old model). I've re-run the old agent twice today, and it said it didn't find any highlights, but when I switched the model (to the new one), it published everything successfully. 👀

Patricia Modispacher _K15t_
Community Champion
April 8, 2026

haha :D That's interesting! Thanks for that info, @Tomislav Tobijas 

Like # people like this

Suggest an answer

Log in or Sign up to answer
TAGS
AUG Leaders

Atlassian Community Events