jevgrep
grep by meaning: pipe in any text, ask a yes/no question in plain English, get only the matching lines. Works behind tail -f, about $0.004 per 1,000 lines, powered by TypeSafe's Jev.
CLI that filters logs and text by meaning using plain-English yes/no questions and Jev probabilities.
How it uses Jev
- Filters by meaning, not keywords. On the benchmark log, it kept the 503 access-log line, the Traceback and the OOMKilled event, none of which contain "error". It dropped error rate=0.00 , 0 failed payments and error handler registered . - Fast enough for a pipe. A match shows up about 0.7 s after the line is written, so it works behind tail -f , kubectl logs -f or journalctl -f . - Cheap. About $0.004 per 1,000 li…
Getting started
jevgrep --json --stats --model jev-1.13 "Mentions a security problem" auth.log \ jq -r 'select(.probability 0.8) "\(.line no)\t\(.text)"'
- Language
- Python 100%
- License
- MIT
- Latest release
- jevgrep 0.1.0
- Last activity
- Sep 2026
- Added
- 2026-09-22
- Stars at snapshot
- 1
- Status
- Community listing
Source: community catalog. Details and metrics reflect the published community snapshot and are not a performance endorsement. Documentation details extracted from GitHub · the README · the project website.