Open Source
Free tools we build in the open
Everything we learn about real production infrastructure goes into open, free, properly-licensed tools. Use them, fork them, and send a pull request — contributions are very welcome.
AI-powered, read-only DevOps troubleshooting — one engine, three interfaces.
A read-only troubleshooting engine for logs, YAML, Terraform, Kubernetes, OpenStack and 20+ technologies. The same analysis engine powers a Typer CLI, a Python SDK, and a FastAPI REST API — offline-first and deterministic, with optional LLM enrichment.
- ▹CLI + Python SDK + REST API on one shared engine
- ▹80+ error signatures across 19 technologies
- ▹Plugin architecture — add a technology without touching the core
- ▹Offline-first; optional Anthropic / OpenAI / Gemini / Ollama
The open knowledge base of real DevOps & cloud infrastructure errors.
Every error is one structured Markdown document — exact message, root causes, diagnostic commands, fixes, prevention and references. Searchable from the terminal with the bundled errlib CLI, validated in CI, and architected to scale to thousands of errors.
- ▹Structured docs: cause → diagnostics → fix → prevention
- ▹Covers Kubernetes, Docker, Terraform, OpenStack, Linux & more
- ▹Fast local search with the errlib CLI
- ▹CI-validated; designed to grow to 10,000+ errors
Why we open-source it
The best DevOps tooling is transparent — you can read exactly what it does before you run it near production. Open code builds trust, invites real-world fixes, and keeps us honest. Both projects are permissively licensed (MIT and CC BY 4.0), so you can use them at work without a procurement conversation.
How to contribute
- • Star a repo so more engineers discover it.
- • Open an issue with an error you hit or a tool you wish existed.
- • Add an error doc or a plugin — both repos have contribution guides and CI checks.
- • Each PR is reviewed and credited.
Prefer the hosted experience? Try the free config validators, the AI Incident Assistant, or the troubleshooting guides — all powered by the same thinking behind the open-source tools.