Dev.to
7/3/2026

The same root cause keeps coming back because nobody tracks it. I built a zero-dep CLI that does.
Short summary
rootecho detects when incident root causes repeat and shows whether previous action items were completed. It compares local JSON postmortems using tag-weighted Jaccard similarity scoring, stores history in .rootecho/history.jsonl for Git integration, and supports both Node.js and Python with byte-identical output. Zero dependencies, MIT licensed, designed for teams managing incidents without paid platforms.
- •Flags when new incident root causes echo past ones and reveals which previous action items were never completed
- •Supports Node.js and Python with dual-language byte-identical output; stores history locally in Git-tracked JSONL
- •Zero dependencies, runs in milliseconds, no ML or network calls—fills the gap for teams using spreadsheets instead of platforms like Rootly
Generated with AI, which can make mistakes.
Is this a good recommendation for you?



