We have multiple SF to SF integrations and, as we know, the secure agent must be defined in Step 6 for these integrations. (This too could be improved, but that's not what my idea is about. That's what this one is about though: https://community.informatica.com/ideas/1021). The problem we have is that we use Sandbox and Production IC orgs and migrate tasks back and forth. When we migrate a SF to SF task across orgs the "first" secure agent is assigned to these tasks. This can cause problems as today's "first' secure agent is not necessarily the same secure agent that will be "first" in the future.
We would like to be able to define a "primary" secure agent.