Expose BearClaw runtime health on infrastructure dashboards

config/dashboards/infrastructure/partials/home_sections.yaml

@@ -13,6 +13,7 @@
 # Notes: Power & Solar shows realtime flow; accumulated history remains in Energy.
 # Notes: Scheduled Proxmox maintenance and sustained WAN quality/reset warnings render amber; critical conditions remain red.
 # Notes: Active Issue rows enumerate faults, scheduled maintenance, and reset urgency so drill-down is optional for initial triage.
+# Notes: BearClaw scheduled-job warnings/errors roll up under Joanna and appear as an exceptions-only Active Issue row.
 # Notes: Home Water and Systems surface normalized Rheem faults; telemetry-only loss renders amber.
 # Notes: Back to Overview is the first control in the landing-view navigation strip.
 ######################################################################
@@ -186,12 +187,15 @@
 
     - type: custom:button-card
       template: bearstone_infra_status_kpi
-      entity: sensor.bearclaw_status_telemetry
+      entity: sensor.bearclaw_scheduled_job_health
       name: Joanna
       icon: mdi:robot-outline
       variables:
         ok_state: ok
+        warning_state: warning
         healthy_text: HEALTHY
+        warning_text: JOB WARNING
+        critical_text: JOB ERROR
       tap_action:
         action: navigate
         navigation_path: /dashboard-infrastructure/joanna
@@ -339,6 +343,24 @@
           action: navigate
           navigation_path: /dashboard-infrastructure/kiosks
 
+      - type: custom:button-card
+        template: bearstone_infra_alert_row
+        entity: sensor.bearclaw_scheduled_job_health
+        name: BearClaw scheduled job needs attention
+        icon: mdi:calendar-alert
+        variables:
+          alert_kind: state_not
+          ok_state: ok
+          alert_severity: >
+            [[[ return entity?.state === 'warning' ? 'warning' : 'critical'; ]]]
+          exception_detail: >
+            [[[ return String(entity?.attributes?.summary ?? 'SCHEDULED JOB STATUS REQUIRES REVIEW').toUpperCase(); ]]]
+        state_display: >
+          [[[ return entity?.state === 'warning' ? 'WARNING' : 'ERROR'; ]]]
+        tap_action:
+          action: navigate
+          navigation_path: /dashboard-infrastructure/joanna
+
       - type: custom:button-card
         template: bearstone_infra_alert_row
         entity: binary_sensor.proxmox1_runtime_healthy

config/dashboards/infrastructure/partials/joanna_sections.yaml

@@ -4,11 +4,12 @@
 # Original Repo : https://github.com/CCOSTAN/Home-AssistantConfig
 # -------------------------------------------------------------------
 # Infrastructure Partial - Joanna
-#  Focused dispatch health, latest activity, rolling volume, and diagnostics.
+#  Focused dispatch health, latest activity, runtime issues, and memory-system freshness.
 # -------------------------------------------------------------------
 # Notes: BearClaw Admin provides the detailed runtime drill-down while this page stays summary-focused.
 # Notes: This view changes HA presentation only; BearClaw runtime behavior is untouched.
-# Notes: Reset expiry is amber within 7 days and red within 3 days; the four reset slots remain visible.
+# Notes: Compact scheduled-job health provides an HA-visible failure summary with Admin as the detailed source of truth.
+# Notes: BearClaw working memory and the federated OneNote KB are separate indexes and render separately.
 ######################################################################
 
 # -------------------------------------------------------------------
@@ -63,7 +64,7 @@
         raw_suffix: ERRORS
 
 # -------------------------------------------------------------------
-# Trend, latest state, and reset summary
+# Trend, latest activity, and runtime health
 # -------------------------------------------------------------------
 - type: grid
   column_span: 4
@@ -109,16 +110,6 @@
       name: HA Dispatches 30D
       icon: mdi:calendar-month
 
-    - type: custom:button-card
-      template: bearstone_infra_list_row
-      entity: sensor.bearclaw_status_telemetry
-      name: BearClaw Admin
-      icon: mdi:shield-crown-outline
-      state_display: OPEN ADMIN
-      tap_action:
-        action: url
-        url_path: http://192.168.10.13:8124/admin
-
   - type: custom:vertical-stack-in-card
     card_mod:
       style: !include /config/dashboards/infrastructure/card_mod/infra_panel.yaml
@@ -165,6 +156,89 @@
       name: All Dispatches 30D
       icon: mdi:calendar-month
 
+  - type: custom:vertical-stack-in-card
+    card_mod:
+      style: !include /config/dashboards/infrastructure/card_mod/infra_panel.yaml
+    cards:
+    - type: custom:button-card
+      template: bearstone_infra_panel_header
+      name: Runtime & Memory
+
+    - type: custom:button-card
+      template: bearstone_infra_status_kpi
+      entity: sensor.bearclaw_scheduled_job_health
+      name: Scheduled Jobs
+      icon: mdi:calendar-alert
+      variables:
+        ok_state: ok
+        warning_state: warning
+        healthy_text: HEALTHY
+        warning_text: JOB WARNING
+        critical_text: JOB ERROR
+      state_display: >
+        [[[
+          const count = Number.parseInt(entity?.attributes?.issue_count ?? '0', 10);
+          if (entity?.state === 'ok') return 'HEALTHY';
+          return `${Number.isFinite(count) ? count : '?'} ${String(entity?.state ?? 'unknown').toUpperCase()}`;
+        ]]]
+
+    - type: custom:button-card
+      template: bearstone_infra_alert_row
+      entity: sensor.bearclaw_scheduled_job_health
+      name: >
+        [[[
+          const issue = Array.isArray(entity?.attributes?.issues) ? entity.attributes.issues[0] : null;
+          return issue?.label || 'Scheduled job needs attention';
+        ]]]
+      icon: mdi:alert-circle-outline
+      variables:
+        alert_kind: state_not
+        ok_state: ok
+        alert_severity: >
+          [[[ return entity?.state === 'warning' ? 'warning' : 'critical'; ]]]
+        exception_detail: >
+          [[[
+            const issue = Array.isArray(entity?.attributes?.issues) ? entity.attributes.issues[0] : null;
+            return String(issue?.detail || entity?.attributes?.summary || 'SCHEDULED JOB STATUS REQUIRES REVIEW').toUpperCase();
+          ]]]
+      state_display: >
+        [[[ return entity?.state === 'warning' ? 'WARNING' : 'ERROR'; ]]]
+
+    - type: custom:button-card
+      template: bearstone_infra_status_kpi
+      entity: sensor.joanna_memory_docs
+      name: BearClaw Working Memory
+      icon: mdi:database-search
+      variables:
+        status_severity: >
+          [[[ return states['binary_sensor.joanna_memory_index_stale']?.state === 'on' ? 'critical' : 'healthy'; ]]]
+      state_display: >
+        [[[
+          const docs = entity?.state ?? '0';
+          const chunks = states['sensor.joanna_memory_chunks']?.state ?? '0';
+          const status = states['binary_sensor.joanna_memory_index_stale']?.state === 'on' ? 'STALE' : 'CURRENT';
+          return `${status} · ${docs} DOCS · ${chunks} CHUNKS`;
+        ]]]
+
+    - type: custom:button-card
+      template: bearstone_infra_status_kpi
+      entity: sensor.joanna_onenote_kb_health
+      name: Federated OneNote KB
+      icon: mdi:notebook-multiple
+      variables:
+        ok_state: ok
+        warning_state: warning
+        healthy_text: CURRENT
+        warning_text: SYNC STALE
+        critical_text: INDEX ERROR
+      state_display: >
+        [[[
+          const docs = Number(entity?.attributes?.docs ?? 0).toLocaleString();
+          const chunks = Number(entity?.attributes?.chunks ?? 0).toLocaleString();
+          const status = entity?.state === 'ok' ? 'CURRENT' : String(entity?.state ?? 'unknown').toUpperCase();
+          return `${status} · ${docs} DOCS · ${chunks} CHUNKS`;
+        ]]]
+
     - type: custom:button-card
       template: bearstone_infra_list_row
       entity: sensor.openai_response
@@ -177,13 +251,10 @@
 
     - type: custom:button-card
       template: bearstone_infra_list_row
-      entity: sensor.joanna_memory_docs
-      name: Memory Index
-      icon: mdi:database-search
-      state_display: >
-        [[[
-          const docs = entity?.state ?? '0';
-          const chunks = states['sensor.joanna_memory_chunks']?.state ?? '0';
-          const stale = states['binary_sensor.joanna_memory_index_stale']?.state === 'on';
-          return `${docs} docs | ${chunks} chunks${stale ? ' | STALE' : ''}`;
-        ]]]
+      entity: sensor.bearclaw_status_telemetry
+      name: BearClaw Admin
+      icon: mdi:shield-crown-outline
+      state_display: OPEN ADMIN
+      tap_action:
+        action: url
+        url_path: http://192.168.10.13:8124/admin

config/packages/README.md

@@ -65,7 +65,7 @@ Kitchen Show note: `kiosk_tablet.yaml` uses Kiosk Satellite's native ESPHome ent
 | [![YAML source: processmonitor](https://img.shields.io/static/v1?label=YAML&message=processmonitor&color=lightgrey&logo=github&logoColor=181717)](processmonitor.yaml) | Root filesystem disk-pressure monitoring with immediate digest/logbook notes at 80%, Joanna review after 10 minutes above 80%, and delayed phone alerts only if the issue stays unresolved after dispatch. | `sensor.disk_use_percent`, `repairs.create`, `script.joanna_dispatch`, `tts.clear_cache` |
 | [![YAML source: tugtainer_updates](https://img.shields.io/static/v1?label=YAML&message=tugtainer_updates&color=lightgrey&logo=github&logoColor=181717)](tugtainer_updates.yaml) | Tugtainer container update notifications via webhook + persistent alerts, with immediate Joanna recovery dispatch for `### Failed:` / `### Rolled-back:` reports and every `### Available:` report. Home Assistant core participates in the 07:00 automatic rollout after a mandatory Tugtainer pre-update config check, with HTTP health and image rollback safeguards. Protected Tugtainer control-plane containers are updated one host at a time through their owning Compose stacks; duplicate Portainer image-update entities are auto-disabled because Tugtainer and Compose own rollout decisions. [![Watch on YouTube](https://img.shields.io/badge/Watch-YouTube-FF0000?logo=youtube&logoColor=white)](https://youtu.be/4NNOkXzUyYw) [![vCloudInfo Blog Post](https://img.shields.io/static/v1?label=vCloudInfo&message=Blog%20Post&color=21759B&logo=wordpress&logoColor=white)](https://www.vcloudinfo.com/2026/07/home-assistant-tugtainer-docker-updates.html) | `automation.tugtainer_disable_duplicate_portainer_image_updates`, `persistent_notification.create`, `event: tugtainer_available_detected`, `script.joanna_dispatch`, `input_datetime.tugtainer_last_update` |
 | [![YAML source: printer](https://img.shields.io/static/v1?label=YAML&message=printer&color=lightgrey&logo=github&logoColor=181717)](printer.yaml) | Epson ink watchdog with one-day and one-week mobile snooze actions for low-ink reminders. | `input_datetime.printer_ink_snooze_until`, `sensor.epson_*`, mobile app action events |
-| [![YAML source: bearclaw](https://img.shields.io/static/v1?label=YAML&message=bearclaw&color=lightgrey&logo=github&logoColor=181717)](bearclaw.yaml) | Joanna/BearClaw bridge automations forward Telegram commands to codex_appliance, include LLM-first routing context for freeform text, relay replies, deliver compact adaptive TeslaMate road-trip milestone, destination-arrival, and final summaries to Carlo and Stacey, emit an active-trip vacation signal, ingest `/api/bearclaw/status` telemetry, and expose dispatch, BearClaw memory-index, user-keyed 7/28/90 step trends, and personal records for Overview Health. Full Joanna reports add directional route comparisons, charging strategy, trip-aware morning context, and a private final route map. Retired QMD telemetry is not imported. Road-trip companion: [![Watch on YouTube](https://img.shields.io/badge/Watch-YouTube-FF0000?logo=youtube&logoColor=white)](https://youtu.be/9-9T6v17NEw) [![vCloudInfo Blog Post](https://img.shields.io/static/v1?label=vCloudInfo&message=Blog%20Post&color=21759B&logo=wordpress&logoColor=white)](https://www.vcloudinfo.com/2026/07/teslamate-road-trip-reports-codex.html). | `rest_command.bearclaw_*`, `event: tesla_road_trip_active`, `sensor.bearclaw_status_telemetry`, `sensor.joanna_*`, `binary_sensor.joanna_*`, `automation.bearclaw_*`, `script.notify_engine`, `script.send_to_logbook` |
+| [![YAML source: bearclaw](https://img.shields.io/static/v1?label=YAML&message=bearclaw&color=lightgrey&logo=github&logoColor=181717)](bearclaw.yaml) | Joanna/BearClaw bridge automations forward Telegram commands to codex_appliance, include LLM-first routing context for freeform text, relay replies, deliver compact adaptive TeslaMate road-trip milestone, destination-arrival, and final summaries to Carlo and Stacey, emit an active-trip vacation signal, ingest `/api/bearclaw/status` telemetry, and expose dispatch, distinct BearClaw working-memory and federated OneNote KB health, compact scheduled-job warning/error health for Systems, user-keyed 7/28/90 step trends, and personal records for Overview Health. Full Joanna reports add directional route comparisons, charging strategy, trip-aware morning context, and a private final route map. Retired QMD telemetry is not imported. Road-trip companion: [![Watch on YouTube](https://img.shields.io/badge/Watch-YouTube-FF0000?logo=youtube&logoColor=white)](https://youtu.be/9-9T6v17NEw) [![vCloudInfo Blog Post](https://img.shields.io/static/v1?label=vCloudInfo&message=Blog%20Post&color=21759B&logo=wordpress&logoColor=white)](https://www.vcloudinfo.com/2026/07/teslamate-road-trip-reports-codex.html). | `rest_command.bearclaw_*`, `event: tesla_road_trip_active`, `sensor.bearclaw_status_telemetry`, `sensor.bearclaw_scheduled_job_health`, `sensor.joanna_onenote_kb_health`, `sensor.joanna_*`, `binary_sensor.joanna_*`, `automation.bearclaw_*`, `script.notify_engine`, `script.send_to_logbook` |
 | [![YAML source: telegram_bot](https://img.shields.io/static/v1?label=YAML&message=telegram_bot&color=lightgrey&logo=github&logoColor=181717)](telegram_bot.yaml) | Legacy Telegram transport marker for BearClaw; the shared `joanna_send_telegram` helper now forwards through the codex_appliance direct Telegram API. | `rest_command.bearclaw_telegram_send`, `script.joanna_send_telegram` |
 | [![YAML source: rheem_econet](https://img.shields.io/static/v1?label=YAML&message=rheem_econet&color=lightgrey&logo=github&logoColor=181717)](rheem_econet.yaml) | Rheem EcoNet water-heater monitoring with normalized alert and telemetry-stale states, event snapshots, persistent Repairs, parent notifications, Joanna diagnostic dispatch, and safe electric-mode enforcement during the current warranty repair. | `binary_sensor.rheem_wh_active_alert`, `binary_sensor.rheem_wh_telemetry_stale`, `binary_sensor.rheem_wh_problem`, `sensor.rheem_wh_diagnostic_status`, `sensor.rheem_wh_last_alert_snapshot`, `repairs.create`, `script.joanna_dispatch` |
 | [![YAML source: phynplus](https://img.shields.io/static/v1?label=YAML&message=phynplus&color=lightgrey&logo=github&logoColor=181717)](phynplus.yaml) | Phyn shutoff automations with leak-test guard, Activity Feed context, Repairs tracking, and critical push recovery when the valve closes. [![Watch on YouTube](https://img.shields.io/badge/Watch-YouTube-FF0000?logo=youtube&logoColor=white)](https://youtu.be/xbhgWnomFYI) | `valve.phyn_shutoff_valve`, `binary_sensor.phyn_leak_test_running`, `script.phyn_send_actionable_leak_notification`, `repairs.create` |

config/packages/bearclaw.yaml

@@ -17,6 +17,9 @@
 # Notes: Progress and destination-arrival callbacks emit tesla_road_trip_active so Vacation Mode can use the signal when the whole family is away.
 # Notes: Status telemetry polling expects !secret bearclaw_status_url (token header stays !secret bearclaw_token).
 # Notes: Status telemetry includes the latest 12 dispatches in the rolling 24-hour window for the Joanna Activity drill-down.
+# Notes: Status telemetry exposes compact scheduled-job warning/error health for the Systems dashboard.
+# Notes: Status telemetry distinguishes the compact BearClaw working-memory index from the federated OneNote KB.
+# Notes: sensor.joanna_onenote_kb_health marks the federated KB warning after 90 minutes without a successful sync.
 # Notes: Status telemetry also exposes compact, user-keyed step trends and personal records for the Health dashboard.
 # Notes: Retired QMD health telemetry is no longer imported or materialized in Home Assistant.
 # Notes: TeslaMate road-trip walkthrough: https://youtu.be/9-9T6v17NEw
@@ -146,10 +149,87 @@ sensor:
       - transports
       - followups
       - memoryIndex
+      - oneNoteKb
       - healthInsights
+      - scheduledJobsHealth
 
 template:
   - sensor:
+      - name: BearClaw Scheduled Job Health
+        unique_id: bearclaw_scheduled_job_health
+        icon: mdi:calendar-alert
+        availability: >-
+          {% set health = state_attr('sensor.bearclaw_status_telemetry', 'scheduledJobsHealth') %}
+          {{ health is mapping and health.get('status') in ['ok', 'warning', 'error'] }}
+        state: >-
+          {% set health = state_attr('sensor.bearclaw_status_telemetry', 'scheduledJobsHealth') | default({}, true) %}
+          {{ health.get('status', 'unknown') }}
+        attributes:
+          issue_count: >-
+            {% set health = state_attr('sensor.bearclaw_status_telemetry', 'scheduledJobsHealth') | default({}, true) %}
+            {{ health.get('issueCount', 0) | int(0) }}
+          error_count: >-
+            {% set health = state_attr('sensor.bearclaw_status_telemetry', 'scheduledJobsHealth') | default({}, true) %}
+            {{ health.get('errorCount', 0) | int(0) }}
+          warning_count: >-
+            {% set health = state_attr('sensor.bearclaw_status_telemetry', 'scheduledJobsHealth') | default({}, true) %}
+            {{ health.get('warningCount', 0) | int(0) }}
+          summary: >-
+            {% set health = state_attr('sensor.bearclaw_status_telemetry', 'scheduledJobsHealth') | default({}, true) %}
+            {{ health.get('summary', 'Scheduled-job health unavailable') }}
+          issues: >-
+            {% set health = state_attr('sensor.bearclaw_status_telemetry', 'scheduledJobsHealth') | default({}, true) %}
+            {{ health.get('issues', []) }}
+          generated_at: >-
+            {% set health = state_attr('sensor.bearclaw_status_telemetry', 'scheduledJobsHealth') | default({}, true) %}
+            {{ health.get('generatedAt') }}
+
+      - name: Joanna OneNote KB Health
+        unique_id: joanna_onenote_kb_health
+        icon: mdi:notebook-check-outline
+        availability: >-
+          {% set kb = state_attr('sensor.bearclaw_status_telemetry', 'oneNoteKb') %}
+          {{ kb is mapping }}
+        state: >-
+          {% set kb = state_attr('sensor.bearclaw_status_telemetry', 'oneNoteKb') | default({}, true) %}
+          {% set last_sync_ts = as_timestamp(kb.get('lastSyncAt'), 0) %}
+          {% set age_minutes = ((as_timestamp(now()) - last_sync_ts) / 60) if last_sync_ts > 0 else 999999 %}
+          {% if not kb.get('ok', false) or kb.get('lastSyncError') not in [none, ''] or kb.get('lastSchedulerError') not in [none, ''] %}
+            error
+          {% elif age_minutes > 90 %}
+            warning
+          {% else %}
+            ok
+          {% endif %}
+        attributes:
+          docs: >-
+            {% set kb = state_attr('sensor.bearclaw_status_telemetry', 'oneNoteKb') | default({}, true) %}
+            {{ kb.get('docs', 0) | int(0) }}
+          chunks: >-
+            {% set kb = state_attr('sensor.bearclaw_status_telemetry', 'oneNoteKb') | default({}, true) %}
+            {{ kb.get('chunks', 0) | int(0) }}
+          last_sync_at: >-
+            {% set kb = state_attr('sensor.bearclaw_status_telemetry', 'oneNoteKb') | default({}, true) %}
+            {{ kb.get('lastSyncAt') }}
+          last_modified_at: >-
+            {% set kb = state_attr('sensor.bearclaw_status_telemetry', 'oneNoteKb') | default({}, true) %}
+            {{ kb.get('lastModifiedAt') }}
+          source_accounts: >-
+            {% set kb = state_attr('sensor.bearclaw_status_telemetry', 'oneNoteKb') | default({}, true) %}
+            {{ kb.get('sourceAccounts', []) }}
+          pending_digest: >-
+            {% set kb = state_attr('sensor.bearclaw_status_telemetry', 'oneNoteKb') | default({}, true) %}
+            {{ kb.get('pendingDigest', 0) | int(0) }}
+          pending_synthesis: >-
+            {% set kb = state_attr('sensor.bearclaw_status_telemetry', 'oneNoteKb') | default({}, true) %}
+            {{ kb.get('pendingSynthesisAll', 0) | int(0) }}
+          carlo_docs: >-
+            {% set kb = state_attr('sensor.bearclaw_status_telemetry', 'oneNoteKb') | default({}, true) %}
+            {{ kb.get('sources', {}).get('carlo', {}).get('docs', 0) | int(0) }}
+          stacey_docs: >-
+            {% set kb = state_attr('sensor.bearclaw_status_telemetry', 'oneNoteKb') | default({}, true) %}
+            {{ kb.get('sources', {}).get('stacey', {}).get('docs', 0) | int(0) }}
+
       - name: Joanna HA Dispatches 24h
         unique_id: joanna_ha_dispatches_24h
         icon: mdi:calendar-clock

config/packages/infrastructure.yaml

@@ -28,6 +28,7 @@
 # Notes: Garage compute health uses the same 145 F threshold as the existing temperature notification.
 # Notes: Immediate website-down states create Repairs; the installed UptimeRobot integration exposes current state, not trend rollups.
 # Notes: Dashboard aggregate issue count is live-only; its underlying degraded sensors retain actionable history.
+# Notes: Dashboard aggregate issue count includes compact BearClaw scheduled-job warning/error health.
 # Notes: Home Water rollup includes Rheem faults/telemetry loss plus active Phyn alerts and valve closure.
 ######################################################################
 
@@ -241,7 +242,9 @@ template:
             'binary_sensor.home_water_problem',
             'binary_sensor.kiosk_devices_need_attention'
           ] %}
-          {{ expand(categories) | selectattr('state', 'eq', 'on') | list | count }}
+          {% set category_count = expand(categories) | selectattr('state', 'eq', 'on') | list | count %}
+          {% set bearclaw_issue = states('sensor.bearclaw_scheduled_job_health') in ['warning', 'error'] %}
+          {{ category_count + (1 if bearclaw_issue else 0) }}
         attributes:
           active_entities: >-
             {% set categories = [
@@ -253,7 +256,11 @@ template:
               'binary_sensor.home_water_problem',
               'binary_sensor.kiosk_devices_need_attention'
             ] %}
-            {{ expand(categories) | selectattr('state', 'eq', 'on') | map(attribute='entity_id') | list }}
+            {% set active = expand(categories) | selectattr('state', 'eq', 'on') | map(attribute='entity_id') | list %}
+            {% if states('sensor.bearclaw_scheduled_job_health') in ['warning', 'error'] %}
+              {% set active = active + ['sensor.bearclaw_scheduled_job_health'] %}
+            {% endif %}
+            {{ active }}
 
   - binary_sensor:
       - name: "Home Water Problem"