What is the nikojim/jev app?

A Python web application that triages customer support tickets with Jev and compares its output with other models through a single OpenRouter API key.

Which Jev model does it call?

The README names typesafe/jev-1.13 as the Jev model. Every ticket is sent to that model and to each comparison model configured in COMPARE_MODELS.

What does it decide for each ticket?

It picks a team, urgency with a reason, severity score and level, category, sentiment, a summary, a draft reply, next steps, escalation and whether assignment can be automatic.

How does the side-by-side comparison work?

Results from Jev and the comparison models appear together with response time, time to first token, token counts and estimated cost, all through one OpenRouter key.

What are Agent Skills in this app?

They are SKILL.md files with team- or situation-specific instructions that are added to the prompt when a ticket matches, letting triage behaviour be steered without code changes.

How do I install and configure it?

Use Python 3.13+ and uv: run uv sync, copy .env.example to .env, then edit the API key plus the JEV_MODEL and COMPARE_MODELS values, and run uv run main.

What is missing from the captured README?

The setup block is captured only through uv run main, so the interface details, data storage, testing approach and licence are not visible; metadata shows 0 stars and no licence.

Jev directory

CommunityJev Demos & Experiments

jev — Support Ticket Triage & LLM Comparison

A Python web app that sends every support ticket to Jev (typesafe/jev-1.13) and to configured comparison models through OpenRouter, showing team routing, urgency, severity, sentiment and draft replies side by side with latency and cost.

See highlights Open github.com

Stars: 0
Language: Python
Reading time: 2 min
Published by: Community
Author: nikojim
Last verified: Oct 2026

nikojim/jev

User repository on GitHub

View on GitHub

Forks: 0
Last pushed: Updated Oct 2026

Repo stats from the GitHub API, cached Oct 2026.

Tagscommunity github python openrouter triage

Works forJev for Classification Jev for Routing Jev for Scoring

Used inJev for Customer Support

Languages: Python
Platforms: OpenRouter Browser
Capabilities: Open sourceRequires API key

Highlights

Every ticket goes to Jev (typesafe/jev-1.13) and to each comparison model configured through one OpenRouter key.
Decides team, urgency, severity score and level, category, sentiment, summary and whether to auto-assign.
Drafts a reply in the ticket's language, suggests next steps and flags escalation or human intervention.
Shows response time, time to first token, token counts and estimated cost side by side.
Agent Skills, SKILL.md files with team or situation instructions, are added to the prompt when a ticket matches.

Quickstart

bash

uv sync
cp .env.example .env
uv run main

Watch out

README reviewed 2026-10-17; the captured text cuts off inside the setup block after uv run main, so the exact JEV_MODEL and COMPARE_MODELS settings, UI behaviour and any test results are not visible. Metadata shows 0 stars and no licence.

Questions about this entry

Short answers drawn from the linked project or page.

Reactions & coverage

Posts, threads, and videos about this entry from around the web.

Instagram: A claim triage tool that routes suspicious claims and never denies one.

Instagram

X: Intent-based search in Gmail

X: 400 companies matched to one candidate

X: Watching 25 customer groups on WhatsApp

From the Jev guides

Original write-ups on what this entry is for.

Per-Tick Decisions in Games

Sep 28, 2026

A game tick is a state plus a legal action list. How Clash Royale, Minecraft, and the earlier Mario, Pokémon, StarCraft, and drone loops ask Jev one step at a time.

Jev Demos & Experiments#gaming#agents#control

Read article

From the Jev AI model community

Posts from builders shipping with Jev right now.

Follow @typesafeai