feat(sidechats): headless gateway new channel auto-provisioning and dedicated heartbeat channel

jobs/box-deep-health.json

@@ -2,7 +2,6 @@
   "agent": "646",
   "chain_next": null,
   "description": "Box deep health \u2014 daily comprehensive check",
-  "dm_target": "646 tasks",
   "followup": {
     "escalate": "opm",
     "expect_reply": true,
@@ -15,7 +14,9 @@
   "prompt_template": "Box deep health check (daily).\nJob ID: {job_id}\nTime: {datetime}\n\nRun the FULL Box health suite on the VM:\n  ssh dev-operator-646@34.139.37.135 \"/srv/box/bin/box-health-check.sh all\"\n\nThen verify these endpoints and services:\n1. https://box.muse-dev.online/ loads (Fleet Console UI, assets box.js and box.css 200 OK)\n2. https://box.muse-dev.online/api/box/fleet returns live fleet status\n3. https://box.muse-dev.online/api/box/dm/log returns live DM log entries\n4. https://box.muse-dev.online/followups loads (operator view, follow-up dashboard)\n5. https://box.muse-dev.online/api/box/followups/summary returns valid JSON\n6. Check /srv/box/dm-log.jsonl tail for errors in the last 24h\n7. Confirm the request-sweeper processed follow-ups in the last hour\n   (grep dm_followup /srv/box/box_requests.jsonl | tail -5)\n\nReport any degradation, even if checks pass (slow responses, log warnings).\n\nReply with OK or FAIL <summary>, plus any observations worth tracking.",
   "schedule": "0 9 * * *",
   "sidechat": {
-    "create": false
+    "create": true,
+    "name_template": "box-deep-health",
+    "reuse_key": "box-deep-health"
   },
   "timeout": 1800
 }