Thanks for the contribution, @ygd58! This fix has been superseded by a larger refactor already on main.

This is an automated hermes-sweeper review.

• The PR patches a systemctl --user is-active block in hermes_cli/status.py at ~line 295, but that code no longer exists on main. The entire gateway status section was refactored to delegate to get_gateway_runtime_snapshot() in hermes_cli/gateway.py.
• get_gateway_runtime_snapshot() already handles Docker/Kubernetes environments: it calls is_container() (hermes_constants.py:196) which checks /.dockerenv, /run/.containerenv, and /proc/1/cgroup. When inside a container, it returns manager="docker (foreground)" and derives liveness from find_gateway_pids() — no systemctl call is made.
• GatewayRuntimeSnapshot.running (hermes_cli/gateway.py:55) is True whenever gateway_pids is non-empty, which correctly reflects a running PID-1 gateway process without any pgrep fallback needed.

The original issue #4776 is addressed by the current architecture. No further changes to status.py are needed for this case.