DiagramPreview
AI betaAI generatedPreviewDownload

Grafana Dashboard Generator

Generate Grafana dashboard JSON for Prometheus, Loki, Node Exporter, API, and Kubernetes monitoring.

サンプル
Preview
The preview will appear here.

Continue with a related tool

Move this output into a nearby workflow for preview, conversion, or debugging.

デザイン仕様書ジェネレーター

Markdown 仕様書を作成し、要件の不足を確認してフロー図を出力できます。

Open
AI 図表ジェネレーター

自然言語プロンプトから Mermaid 図表を生成します。

Open
Mermaid AI 修正

壊れた Mermaid 構文を AI で修正し、読みやすくします。

Open

AI Review beta

Need deeper AI diagram review?

Free tools stay open. We are validating syntax checks, readability scoring, fix suggestions, and publish-ready checklists.

使い方

  1. 1Describe the service, datasource style, and metrics you want to monitor.
  2. 2Generate the dashboard JSON and inspect panel titles, queries, units, and time ranges.
  3. 3Copy or download the JSON, then import it into a Grafana test folder before using it in production.
  4. 4Adjust datasource UIDs and PromQL labels to match your real environment.

主な用途

Grafana dashboard bootstrapPrometheus panel designKubernetes monitoringAPI latency dashboardsObservability documentation

FAQ

Can I import the generated JSON directly into Grafana?

Use it as a draft. Import into a test folder first, then adjust datasource UIDs, PromQL labels, units, thresholds, and folder permissions.

Which metrics should an API dashboard include?

A useful first dashboard usually includes request rate, error rate, latency percentiles, saturation, dependency failures, and recent log or trace links.

Why generate a dashboard instead of building one manually?

Generation gives you a quick baseline for repeatable service dashboards. Human review is still needed to make the panels match real SLOs and labels.

Grafana Dashboard Generator creates editable dashboard JSON for common Prometheus, Loki, Node Exporter, API service, Kubernetes, and database monitoring scenarios.

The generated dashboard should be treated as a starting point: review panel titles, PromQL expressions, thresholds, units, and datasource names before importing it into Grafana.

It pairs well with Prometheus alert rules because the same service signals should appear both as visual panels and actionable alerts.

Demo: baseline API service dashboard panels

For a service dashboard, start with the signals that answer whether the service is receiving traffic, failing, slowing down, or running out of capacity.

  • Traffic: rate(http_requests_total[5m]) grouped by route or status.
  • Errors: 5xx rate and error ratio, not only raw error count.
  • Latency: p50, p95, and p99 from histogram buckets when available.
  • Saturation: CPU, memory, queue depth, or connection pool usage.
histogram_quantile(0.95, sum by (le, route) (rate(http_request_duration_seconds_bucket[5m])))
sum(rate(http_requests_total{status=~"5.."}[5m])) / sum(rate(http_requests_total[5m]))

These expressions are examples; label names need to match your instrumentation.

Dashboard review: avoid panels that look useful but do not answer an incident question

A good generated dashboard should map to real operational questions: is the service alive, is traffic normal, are users seeing errors, and where is the bottleneck. Pretty panels are not enough if they do not guide response.

  • Pair request rate with error rate so traffic drops are not mistaken for success.
  • Use latency percentiles instead of averages for user-facing APIs.
  • Add variables only when they reduce duplication, not when they hide the primary service view.
histogram_quantile(0.95, sum(rate(http_request_duration_seconds_bucket{job="$service"}[5m])) by (le))

Grafana Dashboard Generator のレビューチェックリスト

ドキュメント、PR、障害メモ、引き継ぎの前にソースを視覚的に確認する場合は Grafana Dashboard Generator を使えます。Generate Grafana dashboard JSON for Prometheus, Loki, Node Exporter, API, and Kubernetes monitoring.

エクスポート前に、ラベルの読みやすさ、関係の正しさ、機密情報の有無、編集後のプレビュー整合性を確認してください。

制限とトラブルシュート

プレビューに失敗したら、入力を最小の完全な例に縮小し、構文を確認してから少しずつ戻してください。

プレビューはレビュー面であり、唯一の正解ではありません。重要な出力は人が確認してください。

試せるサンプル入力

  • API service: Generate a Grafana dashboard JSON for an API service with request rate, error rate, latency p95, saturation, and instance health panels.
  • Node Exporter: Generate a Grafana dashboard JSON for Node Exporter CPU, memory, disk, network, and filesystem saturation.
  • Loki logs: Generate a Grafana dashboard JSON for Loki logs with error logs, warning logs, top services, and log volume panels.

ツール成熟度

AI beta

AI 支援 beta

このツールは AI で編集可能なソースを生成します。ローカルでプレビューし、調整してから公開してください。

ラベルは、安定エクスポート、深いデバッグ、素早い解析、AI 支援生成のどれに向くかを示します。