Gonzo: An Open-Source Terminal UI That's Changing How I Analyze Logs

Why Another Log Tool? The Terminal Gap ## Hands-On Impressions from the KubeCon Demo ## Developer-Friendly Features That Stand Out ## Why This Open-Source Tool Feels Different ## Final Take: Worth Adding to Your Toolkit ## kubernetes #devops #observability #opensource #golang #logging #tui #kubecon I am still catching up on the amount of content I captured from KubeCon + CloudNativeCon North America 2025 in Atlanta (always packed with great cloud-native discoveries). One open-source tool that really impressed me at the ControlTheory booth was Gonzo which is a Go-based TUI (terminal UI) for real-time log analysis. I caught the live demo, fired it up myself, and came away thinking: this is exactly the kind of practical, developer-friendly tool the community needs more of. No heavy web app and dashboards. Just a fast, interactive terminal experience that feels right at home in my workflow. Here's why I'm excited about it from a pure dev/ops perspective. We all spend tons of time in the terminal debugging Kubernetes pods, services, or local apps. Tools like tail -f, kubectl logs, or even k9s are staples, but when logs get noisy (mixed severities, patterns buried in volume), it turns into manual grep hell. Gonzo bridges that gap with a beautiful, interactive dashboard – all without leaving your terminal. It's inspired by k9s (same navigation feel), but laser-focused on logs: real-time charts, pattern detection, filtering, and even optional AI insights. The best part? It's fully open source (MIT license) on GitHub: https://github.com/control-theory/gonzo (already over 2k stars and growing fast). I piped some sample logs into it during the demo, and the UX hooked me immediately. Here's the main dashboard in action: Another cool view is the heatmap for severity over time: And detailed stats on selected entries: Installation is super simple: Or via Homebrew: brew install gonzo Then just kubectl logs -f my-pod | gonzo and you're in. Gonzo nails the developer experience: lightweight, no runtime deps beyond Go, runs anywhere you have a terminal. The community aspect shines too – active Slack, clear contributing guide, and modular architecture that invites PRs. At KubeCon, chatting with other devs, the consensus was: we need more tools like this that empower individual engineers without forcing enterprise pipelines. If you live in the terminal and deal with logs daily (who doesn't?), give Gonzo a spin on a side project or next debug session. It's refreshed how I think about quick log triage. Star it on GitHub if it helps – open source wins when we support solid projects like this. Have you tried Gonzo yet, or built something similar? What's your go-to terminal log setup? Drop thoughts in the comments! Follow for more open-source finds, Kubernetes tips, and DevOps takes. Check out the DiscoPosse Podcast for deeper chats. Templates let you quickly answer FAQs or store snippets for re-use. Are you sure you want to hide this comment? It will become hidden in your post, but will still be visible via the comment's permalink. Hide child comments as well For further actions, you may consider blocking this person and/or reporting abuse CODE_BLOCK: go install github.com/control-theory/gonzo/cmd/gonzo@latest Enter fullscreen mode Exit fullscreen mode CODE_BLOCK: go install github.com/control-theory/gonzo/cmd/gonzo@latest CODE_BLOCK: go install github.com/control-theory/gonzo/cmd/gonzo@latest - Dashboard layout: A clean 2x2 grid with live log stream, severity pie chart, word frequency, and timeline views. - Navigation: Vim-style keys, mouse support, global pause (Space), fullscreen modes – super responsive. - Filtering: Quick modals for severity, regex, or Kubernetes namespace/pod selection. - Themes: Tons of built-in skins (Dracula, Nord, etc.) – makes it fun to match your terminal setup. - Kubernetes integration: Native flags for namespace/pod selection, or plugin for k9s (Ctrl-L to launch directly on selected pod logs). - Input flexibility: Stdin pipe, files, tail -f mode, or even built-in OTLP receiver for OpenTelemetry logs. - AI option: Hook in OpenAI-compatible models (including local via Ollama) for pattern summaries – works offline if you want. - Extensible: Clean Go codebase with Bubble Tea for the TUI – easy to hack on or contribute.