dockit vs kibana
Kibana is the standard UI for dashboards and observability, but for daily query development, it's overkill. You have to run an extra service, wait 30 seconds for it to boot, and let it eat 500MB of RAM just to reach the Dev Tools tab.
DocKit is a native desktop app built for Elasticsearch queries. It starts in under two seconds, works offline, and saves your queries as local files so you can actually version control them.
Head-to-head comparison
| DocKit | Kibana | |
|---|---|---|
| Type | Native desktop app | Web app (requires server) |
| Startup time | < 2 s | 10–30 s |
| RAM usage | ~150 MB | 500 MB+ |
| Requires Kibana server | ❌ | ✅ |
| Offline mode | ✅ | ❌ |
| Query editor | Monaco + JSON5 + autocomplete | Basic Dev Tools console |
| AI query assistant | ✅ (OpenAI, DeepSeek) | ❌ |
| Query persistence | ✅ Local files (Git-friendly) | ✅ Saved queries (server-side) |
| Index & cluster management | ✅ | ✅ |
| Visualizations / dashboards | ❌ | ✅ |
| APM / observability | ❌ | ✅ |
| Alerting | ❌ | ✅ |
| DynamoDB / OpenSearch | ✅ | ❌ |
| Open source | ✅ Apache 2.0 | Mixed (Basic free, paid tiers) |
| ES version support | 1.x – 9.x | Version-locked to stack |
When dockit is the better choice
You just need the query editor
Kibana Dev Tools is fine, but you don't need a whole server cluster running just to write a GET /my-index/_search query. DocKit gives you a better editor. It uses Monaco (the VS Code engine) with JSON5 support, provides field autocomplete from your live mapping, and handles auto-formatting and curl exports without any server overhead.
You work across clusters frequently
DocKit stores connection profiles so you can switch between dev, staging, and prod instantly. Kibana is usually pinned to the cluster where it was deployed. Switching environments in DocKit is just a click in the sidebar.
You want ai-assisted query generation
You can describe what you need in plain English and DocKit uses your actual index mapping as context to generate the DSL. Kibana doesn't have an equivalent for this. It works with OpenAI or DeepSeek using your own API key, so your data stays on your machine.
You need offline capability
DocKit works without an internet connection once it's installed. Your queries, history, and credentials stay local. Kibana requires a network path to both its own server and the Elasticsearch cluster, which is a pain when you're on a VPN or working locally.
Your team uses dynamodb or opensearch too
DocKit supports Elasticsearch, OpenSearch, and DynamoDB in one app. If you're touching more than one type of database, DocKit can replace two or three other tools.
When kibana is the better choice
Keep Kibana if you need:
- Dashboards and visualizations: Kibana Lens and Canvas are the gold standard for building charts for stakeholders.
- APM and tracing: Integration with Elastic APM, distributed tracing, and service maps are native to Kibana.
- Alerting: Rule based alerting with connectors for PagerDuty or Slack is built in.
- Machine learning: Elastic ML anomaly detection lives in Kibana.
- Shared browser access: If non-developers need a UI, the web interface is the way to go.
Most engineers I know use both. DocKit handles query development and index management, while Kibana handles the dashboards and observability.
The query editor difference
Kibana Dev Tools uses a CodeMirror console. It works, but it's got some annoying limits. No JSON5 means you have to be strict about double quotes and you can't add inline comments. Field autocomplete is often slow or partial, and there's no AI help. Plus, your saved queries are stuck on the server.
DocKit's editor is built on Monaco. It supports JSON5, so you can use comments and trailing commas. Autocomplete is drawn directly from your cluster mapping when you connect. It also includes DSL syntax validation, auto-formatting, and a one-click copy as curl. Since it supports multiple tabs, you can run independent queries in parallel without losing your spot. Your full history is also saved locally so you can search and replay anything you've run.
Setup: connecting dockit to your elasticsearch cluster
- Download DocKit for macOS, Windows, or Linux.
- Open DocKit, go to New Connection, and select Elasticsearch.
- Enter your host, port, and credentials.
- Click Connect. Your indices will show up in the sidebar.
- Open Dev Tools and start querying.
If you're using clusters with TLS or client certificates, check the connection guide.
Elasticsearch version support
DocKit uses the standard REST API and works with Elasticsearch 1.x through 9.x. It supports both Apache 2.0 and Elastic License distributions. Kibana, on the other hand, is version locked to its specific stack release.
Frequently asked questions
Is DocKit a full Kibana replacement? No — and it doesn't try to be. For query development and index management it covers the same ground (and adds AI + offline). For dashboards, APM, and alerting, Kibana has no equivalent in DocKit. Most teams keep both.
Does DocKit support Kibana's saved queries format? No. DocKit stores queries as local files (JSON). You can copy DSL from Kibana Dev Tools into DocKit and vice versa — the query language is identical.
Can I use DocKit with Elastic Cloud? Yes. Enter your Elastic Cloud endpoint and API key in the connection dialog. Works with all Elastic Cloud tiers.
Does it work with self-hosted Elasticsearch behind a VPN or bastion? Yes, as long as DocKit can reach the Elasticsearch REST port (default 9200). For SSH tunneling, set up the tunnel externally and point DocKit at localhost.
→ DocKit full feature overview · Elasticsearch GUI client page · ES GUI deep-dive