I hit a an automation skill-routing issue. An existing scheduled run mentioned “automation,” which activated the creation-oriented skill. The agent then tried to inspect the scheduler instead of executing the actual task. And then everything went sideways. I fixed the prompt and that fixed it. It might make sense adding an run-guard to the existing automation skill to prevent this when creating / modifying automations.

Do you have any ideas on the best place to address this @modern-fireman-99755?

I think the best possible fix would be to update the Agent Profile to select only a specific set of skills for the automation. That way, you can remove this skill from the profile and the automation will run normally. Otherwise, short of changing the prompt, there isn't much we can do to prevent LLMs from getting confused. But I would push to implement the above proposed feat/fix, as could improve performance for the automation.