GONZO
Gonzo is a powerful, real-time log analysis terminal UI inspired by k9s and Dstl8. Analyze log streams with beautiful charts, AI-powered insights, and advanced filtering – all from your terminal.
Star us to stay updated!
Open Source · MIT License · Terminal Native · OTLP Ready · Built for AI Coders
Log Analysis Tool for Live Investigation
Gonzo is a log analysis tool for engineers who need answers while systems are running. Instead of sending you into a slow dashboard workflow, it brings real-time log analysis into the terminal. Stream logs from Vercel, Supabase, Railway, Render, Kubernetes, OTLP, CloudWatch, Loki, or local files, then quickly move from raw output to usable signal with top words, severity counts, heat maps, and regex filtering.
It’s built for engineers who need answers while systems are running and prefer terminal-native workflows over dashboard-heavy tools.
Learn more about our integrations…
Integrations: Gonzo Works With The Tools You Already Have
Gonzo fits naturally beside the CLI tools you already use. Pipe kubectl logs, work from K9s or Stern, receive structured logs directly over OTLP, tail AWS CloudWatch, follow Grafana Loki in real time, or stream vercel logs -j into Gonzo and keep the metadata intact.
Stream Render service logs through a jq normalizer and inspect deployments, cron jobs, and web services in real time.
Zero-config JSONL pipe – railway logs –json | gonzo is all it takes to get structured log analysis.
Poll and normalize logs across all nine Supabase services – Auth, Realtime, Storage, Edge Functions, PostgREST, and more.
Pipe Vercel logs -j into Gonzo and inspect deployment logs with readable structure and filtering.
Stream Netlify serverless function and deploy logs into Gonzo for structured analysis of your JAMstack builds.
Unwrap Fly.io’s double-encoded JSON log format and tail multi-region machine logs with full structure preserved.
Flatten wrangler tail output from Workers and Pages Functions into structured JSONL for real-time edge log analysis.
Bring cloud, serverless, and application logs into the same terminal workflow.
Use Gonzo with kubectl, K9s, Stern, or cluster-native workflows to understand live pod and namespace activity faster.
Pipe docker logs or Docker Compose service output directly into Gonzo for container-level log visibility.
Receive OpenTelemetry logs directly and preserve attributes for structured investigation.
Add Gonzo as a K9s plugin and launch terminal log analysis on any pod with a single keyboard shortcut.
Use Claude as Gonzo’s built-in AI analysis provider for root cause suggestions and pattern detection on live logs.
Connect GPT-5 or any OpenAI-compatible model to power Gonzo’s AI log analysis and anomaly detection.
Use OpenAI models powering Codex as Gonzo’s AI backend for log pattern analysis and debugging suggestions.
Use Gemini as Gonzo’s AI analysis backend via any OpenAI-compatible endpoint for pattern detection and root cause suggestions.
Launch Gonzo from Cursor’s integrated terminal and pair AI-powered log analysis with your editor-native debugging workflow.
GCP Cloud Logs
Stream Google Cloud Logging entries through gcloud logging tail and analyze them with Gonzo’s structured dashboard.
See Gonzo in Action
Gonzo’s Main Dashboards
Stats!
Heatmaps!
Why Engineers Love Gonzo
Key Features
Real-Time Analysis
Process logs from multiple sources with OTLP and Victoria Logs native support, automatic format detection, and color-coded severity tracking.
Interactive Dashboard
k9s-inspired 2×2 grid interface with real-time charts, stay in-flow with vim-style navigation, and expandable modal views for deep log analysis.
Advanced Filtering
Drill down faster, surface what matters with regex support, attribute search, severity filtering, and interactive keyboard/mouse navigation.
AI-Powered Insights
Get intelligent pattern detection, anomaly analysis, and root cause suggestions. Works with OpenAI, Anthropic Claude, Ollama, LM Studio, and more.
Multiple Log Sources
Stream from Victoria Logs, receive via OTLP (OpenTelemetry), tail log files, or pipe from kubectl and other tools.
Customizable Themes
11+ beautiful themes available including Dracula, Nord, and GitHub Light. Create custom skins with YAML or use ControlTheory’s original themes.
Quick Installation
Using Go
Install directly with Go (requires Go 1.21+)
go install github.com/control-theory/gonzo/cmd/gonzo @latest
Using Homebrew
Available on macOS and Linux
brew install gonzo
Download Binary
Get the latest release for your
platform
Build from Source
Clone and build the project yourself
git clone https://github.com/control-theory/gonzo.git
cd gonzo
make build
Quick Start
Start analyzing logs in seconds:
# Analyze logs from a file
$ cat application.log | gonzo
# Stream Kubernetes logs
$ kubectl logs -f deployment/my-app | gonzo
# Stream from Victoria Logs
$ gonzo –vmlogs-url=”https://localhost:9428″ –vmlogs-query=”*”
# Receive OTLP logs
$ gonzo –otlp-enabled
# With AI analysis
$ export OPENAI_API_KEY=sk-your-key-here
$ cat logs.json | gonzo
Trusted by Engineers
Gonzo is built for engineers debugging real runtime failures, not polishing dashboards after the fact. It is already used by teams working across AI-generated code, Vercel deployments, OpenTelemetry pipelines, and modern terminal-first workflows where fast log analysis matters.
