{"wiki":{"id":340,"slug":"app-watchdog-01","entity_type":"droid","entity_id":"app-watchdog-01","title":"App Watchdog","prose_md":"## About\n\nApp Watchdog is a scheduled droid owned by [[agt_029]] that keeps Carol's shared apps alive. Every five minutes it probes the critical services the ecosystem depends on — such as the initiatives tracker [[initiatives]], the gateway [[gateway]], and access management [[access-management]] — and, if any have stopped responding, relaunches them automatically. It exists so a crashed app never sits unnoticed waiting for someone to look.\n\nIts runs are triggered by the scheduler operated by [[agt_016]], which means every execution is recorded and audited: the watchdog's supervisor is itself supervised. After each probe it restarts whatever is down and emits a run-audit through its wrapper, giving the stewardship team a clean trail of what happened and when.\n\nThe name \"App Watchdog\" is unique to this droid. It is technically part of the (none) droid family, a set of 222 similar droids run by many Carol agents, but no other droid shares its exact name. It is pure software: its source makes no calls to Claude, so no model is involved in its work.","namesake_json":"{\"engine\": \"software\", \"model\": null, \"claude_purpose\": null, \"family\": \"(none)\", \"family_size\": 222, \"is_unique\": true}","profile_pic_path":"avatars/droid/app-watchdog-01.png","source_hash":"7de5c7695f05f5e01ec7ac6d605e07064088d6862a675686901ff238d146987f","status":"being_built","last_generated_at":"2026-08-05 02:37:49","created_at":"2026-07-24 03:30:16","updated_at":"2026-08-05 02:37:49"},"facts":{"id":"app-watchdog-01","name":"App Watchdog","machine_name":"app_watchdog","owner":"agt_029","function":"Restarts critical shared apps if down (initiatives, planner, gateway, auth)","process_type":"scheduled","schedule":"Every 5 min","process_name":"app-watchdog","avatar_color":"#94a3b8","created_for":"CAROL-INI-0991","purpose":"Shared system service (steward Radagast): health-checks the critical shared apps and relaunches any that are down. Now triggered by the Hermione Scheduler so its runs are audited and the supervisor is itself supervised.","duties":"Probe each critical app; relaunch if down; emit run-audit via the wrapper.","constraints":"","status":"running","gender":"","archetype":"","building_block":"support","service_override":null,"enabled":1,"task_key":"system-services.app_liveness_watch","model_free":0},"page":{"type":"droid","page_class":"main","class_label":"Main page","kind_label":"Droid","kind_gloss":"","listed":true}}