Jev in production › Message and ticket triage

jev-inbox-triage

Jev decision model sorts a store's incoming emails into routing categories inside an email inbox pipeline. 19/20 emails routed correctly (author).

Open on GitHub ↗

19/20measured, as published by the source
Use
Message and ticket triage
Industry
E-commerce and retail
Form
Open-source tool
Stage
In production
Listed
2026-10-06
Found via
discord
Repository
pentasir/jev-inbox-triage
Stars
0
Forks
1
Last push
2026-10-11
Language
HTML
License
MIT
jev-inbox-triage screenshot
docs/workflow-map.png in the pentasir/jev-inbox-triage README, MIT; shown from GitHub.

The README opens with

An experiment in giving a small online store's inbox a quiet clerk: a decision model reads every customer email first and decides which ones a person needs to see now, which can wait, which a template could answer, and which are only sales pitches.

Result ------ Emails sent to the right lane 19 / 20 Urgent or risky emails sent to automation 0 Median time to judge an email 338 ms Input per email ~1,000 tokens (8 questions, one call) Cost at 600 emails a month ~2.5 US cents

Badge

For the project's own README, linking back here:

Listed in Jev in production

Also used for message and ticket triage