Skip to content

Jev directory

Tocsin

A Rust CLI and service for log triage at ingest: it collapses a stream into Drain templates, asks Jev a policy-driven set of typed questions once per new pattern, and routes every line to page, ticket or log.
Stars
4
Language
Rust
License
MIT
Reading time
2 min
Published by
Community
Author
tpateeq
Last verified
Oct 2026

tpateeq/tocsin

User repository on GitHub

View on GitHub

Log triage at ingest. Group logs into patterns, judge each pattern once, page only on what matters.

Forks
1
Last pushed
Updated Sep 2026

Repo stats from the GitHub API, cached Oct 2026.

Highlights

  • The author's run collapsed 22,833,750 log lines into 11,812 templates and 13,080 Jev calls for $0.64.
  • Each template gets four typed questions: pageable and detail as Noul, severity as a 0–3 Score, area as Choice; attention is computed in code.
  • Numbers, IPs, UUIDs, hex values and emails are masked first; HTTP status classes are preserved, so 5xx never merges with 200s.
  • Routes follow a paging policy in plain English: at attention ≥0.75 page, ≥0.5 ticket, otherwise log — pages at most once per pattern per cooldown.
  • Serve mode accepts OTLP/HTTP JSON logs or plain NDJSON lines and emits Slack-compatible webhook payloads with template, count, severity and area.

Quickstart

bash
cargo install --git https://github.com/TPAteeq/tocsin
export TYPESAFE_API_KEY=...
kubectl logs deploy/api --since=1h | tocsin triage

Watch out

Copy reviewed against the repo README on 2026-10-15; the capture ends at the webhook payload table. The cost comparison figures are the author's own single run, and the README links a full write-up at tpateeq.github.io/tocsin that includes a run where the approach lost.

Questions about this entry

Short answers drawn from the linked project or page.

From the Jev guides

Original write-ups on what this entry is for.

From the Jev AI model community

Posts from builders shipping with Jev right now.

Follow @typesafeai