Hi @leoncalverley,
I will definitely be sharing the thoughtful feedback on the agent with our engineering team!
Our very wonderful and amazing @bobthebear has addressed in great depth a similar issue in another thread on our forums which I will link right here.
The gist being, there is some room for the "prompt engineering" to drill down and enforce a tool to always be used.
Making sure that the prompt isn't too big is a key factor in reducing the occurrence of any surprises for the final agent action. As @bobthebear points out delegating tasks out to different agents can help to greatly reduce this.