The actual wall everyone hits in production is domain-level state and credentials. When an agent holds valid API tokens and hallucinates a destructive action inside an authorized session, OS-level sandboxing won't save you.
For folks actually running agents in prod: are you gating every state-mutating API call with human approvals, or spinning up dynamic shadow environments to execute and verify side effects on mirrored state before committing?
AppD offers some SIEM. Splunk does much more than SIEM. Splunk Observability Cloud has nothing to do with Splunk Enterprise, it's a fully fledged AppD competitor.
They have some many parts of spinning up/change the infrastructure (Terraform), connect the services (Consul), run the apps (Nomad) but not their own way to tell you how well they do. Also monitoring is quite sticky and high margin. I think it makes sense but have no special insight.
It will be interesting to watch what new products the launch.
I guess a monitoring solution would be next. Netdata could be a good fit with agent-first approach.