- ./check stress spins a fresh cube-server+cubec built from this tree on a
throwaway socket/store and drives ~150s of real prog/run traffic while
sampling the per-command latency + per-C telemetry.
- tools/stress.sh: never touches the production daemon; STRESS_SECONDS override.
- Fixed latent control-flow bug: the old '|| exit 0' guards made 'bench' and
'mount' subcommands unreachable after the gate (they exited early). Now uses
explicit run_* flags and only exits at the true end.
User clarified: the earlier prune was a ONE-TIME cleanup of an over-backfill
bug, NOT a recurring retention cap. Saved history must persist. Removed
cmd_prune / --prune-older-than-hours entirely; the script now only captures
forward (default) or seeds a bounded window once. No code path deletes cube
entries. Verified: help has no prune arg, forward/seed dry-runs change no
state, live timer runs forward-only.
Captures user/assistant messages from /root/.hermes/state.db into the
'hermes' cube namespace as TYPE: HIST-QNA entries. Forward-only by default
(no backfill); --seed-hours N for bounded one-time population; idempotent;
--prune-older-than-hours to undo over-backfill. systemd timer runs every
5 min. Tested: 24h seed = 1171 msgs; prune cleared a 3164-entry backfill
to 44, then re-seeded correctly.
hermes