Enforce the allowed versions of code that can perform an action

[thesocialdev]

We need to improve the security posture of this process beyond allowing selected clients to guarantee information integrity.

The ideas is that we can have guardrails to avoid an actor to modify the code and process the AI task with bad intent.