openworker/deploy/reference/dashboards/dashboard.json

{
  "uid": "openworker-harness",
  "title": "OpenWorker · Harness operations",
  "schemaVersion": 39,
  "version": 1,
  "editable": false,
  "tags": [
    "openworker"
  ],
  "timezone": "browser",
  "refresh": "10s",
  "time": {
    "from": "now-30m",
    "to": "now"
  },
  "panels": [
    {
      "id": 1,
      "title": "Operations / second",
      "description": "Actual observed operations, not unique user turns.",
      "type": "timeseries",
      "gridPos": {
        "x": 0,
        "y": 0,
        "w": 24,
        "h": 8
      },
      "datasource": {
        "type": "prometheus",
        "uid": "prometheus"
      },
      "targets": [
        {
          "refId": "A",
          "expr": "sum(rate(openworker_operation_count_total[$__rate_interval])) by (openworker_operation)",
          "legendFormat": "{{openworker_operation}} {{openworker_outcome}} {{direction}}"
        }
      ],
      "fieldConfig": {
        "defaults": {
          "unit": "ops",
          "color": {
            "mode": "palette-classic"
          },
          "custom": {
            "lineWidth": 2,
            "fillOpacity": 8,
            "showPoints": "never",
            "drawStyle": "line",
            "spanNulls": false
          }
        },
        "overrides": []
      },
      "options": {
        "legend": {
          "displayMode": "list",
          "placement": "bottom"
        },
        "tooltip": {
          "mode": "multi"
        }
      }
    },
    {
      "id": 2,
      "title": "Operation latency · p95",
      "description": "Measured active operations; not wall-clock approval waiting.",
      "type": "timeseries",
      "gridPos": {
        "x": 0,
        "y": 8,
        "w": 12,
        "h": 8
      },
      "datasource": {
        "type": "prometheus",
        "uid": "prometheus"
      },
      "targets": [
        {
          "refId": "A",
          "expr": "histogram_quantile(0.95, sum(rate(openworker_operation_duration_seconds_bucket[$__rate_interval])) by (le, openworker_operation))",
          "legendFormat": "{{openworker_operation}} {{openworker_outcome}} {{direction}}"
        }
      ],
      "fieldConfig": {
        "defaults": {
          "unit": "s",
          "color": {
            "mode": "palette-classic"
          },
          "custom": {
            "lineWidth": 2,
            "fillOpacity": 8,
            "showPoints": "never",
            "drawStyle": "line",
            "spanNulls": false
          }
        },
        "overrides": []
      },
      "options": {
        "legend": {
          "displayMode": "list",
          "placement": "bottom"
        },
        "tooltip": {
          "mode": "multi"
        }
      }
    },
    {
      "id": 3,
      "title": "Errors / second",
      "description": "Error outcomes; cancellations and unknown effects are separate.",
      "type": "timeseries",
      "gridPos": {
        "x": 12,
        "y": 8,
        "w": 12,
        "h": 8
      },
      "datasource": {
        "type": "prometheus",
        "uid": "prometheus"
      },
      "targets": [
        {
          "refId": "A",
          "expr": "sum(rate(openworker_operation_count_total{openworker_outcome=\"error\"}[$__rate_interval])) by (openworker_operation)",
          "legendFormat": "{{openworker_operation}} {{openworker_outcome}} {{direction}}"
        }
      ],
      "fieldConfig": {
        "defaults": {
          "unit": "ops",
          "color": {
            "mode": "palette-classic"
          },
          "custom": {
            "lineWidth": 2,
            "fillOpacity": 8,
            "showPoints": "never",
            "drawStyle": "line",
            "spanNulls": false
          }
        },
        "overrides": []
      },
      "options": {
        "legend": {
          "displayMode": "list",
          "placement": "bottom"
        },
        "tooltip": {
          "mode": "multi"
        }
      }
    },
    {
      "id": 4,
      "title": "Main-model reported tokens",
      "description": "Absent usage stays absent. Excludes reviewer/compaction accounting; not a billing ledger.",
      "type": "timeseries",
      "gridPos": {
        "x": 0,
        "y": 16,
        "w": 8,
        "h": 8
      },
      "datasource": {
        "type": "prometheus",
        "uid": "prometheus"
      },
      "targets": [
        {
          "refId": "A",
          "expr": "sum(increase(openworker_model_tokens_total[$__range])) by (direction)",
          "legendFormat": "{{openworker_operation}} {{openworker_outcome}} {{direction}}"
        }
      ],
      "fieldConfig": {
        "defaults": {
          "unit": "short",
          "color": {
            "mode": "palette-classic"
          },
          "custom": {
            "lineWidth": 2,
            "fillOpacity": 8,
            "showPoints": "never",
            "drawStyle": "line",
            "spanNulls": false
          }
        },
        "overrides": []
      },
      "options": {
        "legend": {
          "displayMode": "list",
          "placement": "bottom"
        },
        "tooltip": {
          "mode": "multi"
        }
      }
    },
    {
      "id": 5,
      "title": "Main-model retries",
      "description": "Attempts after the first, not automatically retried tools.",
      "type": "timeseries",
      "gridPos": {
        "x": 8,
        "y": 16,
        "w": 8,
        "h": 8
      },
      "datasource": {
        "type": "prometheus",
        "uid": "prometheus"
      },
      "targets": [
        {
          "refId": "A",
          "expr": "sum(increase(openworker_model_retry_count_total[$__range]))",
          "legendFormat": "{{openworker_operation}} {{openworker_outcome}} {{direction}}"
        }
      ],
      "fieldConfig": {
        "defaults": {
          "unit": "short",
          "color": {
            "mode": "palette-classic"
          },
          "custom": {
            "lineWidth": 2,
            "fillOpacity": 8,
            "showPoints": "never",
            "drawStyle": "line",
            "spanNulls": false
          }
        },
        "overrides": []
      },
      "options": {
        "legend": {
          "displayMode": "list",
          "placement": "bottom"
        },
        "tooltip": {
          "mode": "multi"
        }
      }
    },
    {
      "id": 6,
      "title": "First visible output · p95",
      "description": "No data is expected from a purely scripted model.",
      "type": "timeseries",
      "gridPos": {
        "x": 16,
        "y": 16,
        "w": 8,
        "h": 8
      },
      "datasource": {
        "type": "prometheus",
        "uid": "prometheus"
      },
      "targets": [
        {
          "refId": "A",
          "expr": "histogram_quantile(0.95,sum(rate(openworker_model_first_output_duration_seconds_bucket[$__rate_interval])) by (le))",
          "legendFormat": "{{openworker_operation}} {{openworker_outcome}} {{direction}}"
        }
      ],
      "fieldConfig": {
        "defaults": {
          "unit": "s",
          "color": {
            "mode": "palette-classic"
          },
          "custom": {
            "lineWidth": 2,
            "fillOpacity": 8,
            "showPoints": "never",
            "drawStyle": "line",
            "spanNulls": false
          }
        },
        "overrides": []
      },
      "options": {
        "legend": {
          "displayMode": "list",
          "placement": "bottom"
        },
        "tooltip": {
          "mode": "multi"
        }
      }
    },
    {
      "id": 7,
      "title": "Policy & human decisions",
      "description": "Committed transitions and policy reviews, not current pending count. Use Console for durable pending decisions.",
      "type": "timeseries",
      "gridPos": {
        "x": 0,
        "y": 24,
        "w": 12,
        "h": 8
      },
      "datasource": {
        "type": "prometheus",
        "uid": "prometheus"
      },
      "targets": [
        {
          "refId": "A",
          "expr": "sum(increase(openworker_operation_count_total{openworker_operation=~\"approval.*|question.*|policy.review\"}[$__range])) by (openworker_operation,openworker_outcome)",
          "legendFormat": "{{openworker_operation}} {{openworker_outcome}} {{direction}}"
        }
      ],
      "fieldConfig": {
        "defaults": {
          "unit": "short",
          "color": {
            "mode": "palette-classic"
          },
          "custom": {
            "lineWidth": 2,
            "fillOpacity": 8,
            "showPoints": "never",
            "drawStyle": "line",
            "spanNulls": false
          }
        },
        "overrides": []
      },
      "options": {
        "legend": {
          "displayMode": "list",
          "placement": "bottom"
        },
        "tooltip": {
          "mode": "multi"
        }
      }
    },
    {
      "id": 8,
      "title": "Recovery & uncertain effects",
      "description": "Observed recovery transitions, not scans. Unknown effects must not be replayed blindly.",
      "type": "timeseries",
      "gridPos": {
        "x": 12,
        "y": 24,
        "w": 12,
        "h": 8
      },
      "datasource": {
        "type": "prometheus",
        "uid": "prometheus"
      },
      "targets": [
        {
          "refId": "A",
          "expr": "sum(increase(openworker_operation_count_total{openworker_operation=~\"session.recovery|tool.unknown_outcome|turn.cancellation\"}[$__range])) by (openworker_operation)",
          "legendFormat": "{{openworker_operation}} {{openworker_outcome}} {{direction}}"
        }
      ],
      "fieldConfig": {
        "defaults": {
          "unit": "short",
          "color": {
            "mode": "palette-classic"
          },
          "custom": {
            "lineWidth": 2,
            "fillOpacity": 8,
            "showPoints": "never",
            "drawStyle": "line",
            "spanNulls": false
          }
        },
        "overrides": []
      },
      "options": {
        "legend": {
          "displayMode": "list",
          "placement": "bottom"
        },
        "tooltip": {
          "mode": "multi"
        }
      }
    },
    {
      "id": 9,
      "title": "Telemetry dropped",
      "description": "Cumulative local drops per process. During collector outage inspect API /readyz as well.",
      "type": "timeseries",
      "gridPos": {
        "x": 0,
        "y": 32,
        "w": 12,
        "h": 8
      },
      "datasource": {
        "type": "prometheus",
        "uid": "prometheus"
      },
      "targets": [
        {
          "refId": "A",
          "expr": "max(openworker_telemetry_dropped) by (instance)",
          "legendFormat": "{{openworker_operation}} {{openworker_outcome}} {{direction}}"
        }
      ],
      "fieldConfig": {
        "defaults": {
          "unit": "short",
          "color": {
            "mode": "palette-classic"
          },
          "custom": {
            "lineWidth": 2,
            "fillOpacity": 8,
            "showPoints": "never",
            "drawStyle": "line",
            "spanNulls": false
          }
        },
        "overrides": []
      },
      "options": {
        "legend": {
          "displayMode": "list",
          "placement": "bottom"
        },
        "tooltip": {
          "mode": "multi"
        }
      }
    },
    {
      "id": 10,
      "title": "Telemetry export failures",
      "description": "Failed batches, not a precise count of failed HTTP requests.",
      "type": "timeseries",
      "gridPos": {
        "x": 12,
        "y": 32,
        "w": 12,
        "h": 8
      },
      "datasource": {
        "type": "prometheus",
        "uid": "prometheus"
      },
      "targets": [
        {
          "refId": "A",
          "expr": "max(openworker_telemetry_export_failures) by (instance)",
          "legendFormat": "{{openworker_operation}} {{openworker_outcome}} {{direction}}"
        }
      ],
      "fieldConfig": {
        "defaults": {
          "unit": "short",
          "color": {
            "mode": "palette-classic"
          },
          "custom": {
            "lineWidth": 2,
            "fillOpacity": 8,
            "showPoints": "never",
            "drawStyle": "line",
            "spanNulls": false
          }
        },
        "overrides": []
      },
      "options": {
        "legend": {
          "displayMode": "list",
          "placement": "bottom"
        },
        "tooltip": {
          "mode": "multi"
        }
      }
    }
  ]
}