Why does the code live in jev-tools instead of this repo?

This repo is only the wrapper so the action can sit on the Marketplace. The implementation lives in jev-tools under packages/triage.

What happens when Jev is not confident about a label?

Jev returns a label with a confidence score. When confidence is low the issue is flagged for a human, or for an LLM if you supplied a key, instead of guessing.

What trigger and permissions does the workflow need?

Trigger on issues of type opened with issues:write permission, then call CMaintz/jev-triage@v2 with jev-api-key from secrets.

Can I stay on the older standalone behavior?

Yes. Version one still points at the older standalone version and keeps working while v2 uses the jev-tools code.

How do I add it to a workflow?

Use the action on issues opened with issues:write permission: a job on ubuntu-latest that checks out the repo and calls CMaintz/jev-triage@v1 with jev-api-key from a secret. Config, inputs and the escalation setup are documented in the jev-tools triage README.

How should I pin the action version?

Pin @v1 to follow updates, or @v1.0.1 for an exact release. @v1.0.0 is the last standalone version from before the move to jev-tools, and the README says v1 still points at the older standalone code and keeps working.

What key does the action need, and what resolves a flagged issue?

It takes a jev-api-key input, typically mapped to a repository secret, for the Jev call. When confidence is low the issue is flagged for a human, or for an LLM if you supplied an LLM key, rather than being guessed at.

CommunityJev Tools & CLIs

Jev Triage — CMaintz

GitHub Action that labels newly opened issues with Jev, returning a label plus confidence and flagging low-confidence cases for a human or an LLM instead of guessing.

See highlights

Open github.com

Stars: 1
License: MIT
Published by: Community
Author: CMaintz
Last verified: Oct 2026

CMaintz/jev-triage

User repository on GitHub

View on GitHub

Near-free GitHub issue triage powered by TypeSafe AI's Jev - typed, confidence-gated labels that escalate only the uncertain cases.

Forks: 1
Last pushed: Updated Oct 2026

Repo stats from the GitHub API, cached Oct 2026.

This is one of several Jev listings for the same project. The canonical entry is Jev Triage — boldbug1.

About this project

Label GitHub issues for basically nothing, using TypeSafe AI's Jev. Jev gives a label plus a confidence score; if it's not sure, the issue gets flagged for a human (or an LLM, if you give it a key) instead of guessing. The code lives in jev-tools now. This repo is just the wrapper so the action can sit on the Marketplace. Config, inputs and the escalation setup are documented in the jev-tools triage README.

Quoted from the project's README. Read the full README.

Tagscommunity github

Works forJev for Classification

Used inJev for Software & DevTools

Highlights

Runs on issues of type opened and requests issues:write permission.
Implementation lives in jev-tools packages/triage; this repo is the Marketplace wrapper.
Configures via jev-api-key input; setup details are in the jev-tools triage README.
Version one still points at the older standalone version and keeps working.

Watch out

Needs a Jev API key; low-confidence issues need a human or your own LLM key to resolve. Copy reviewed against the repository README on 2026-10-04, not from running it.

Questions about this entry

Short answers drawn from the linked project or page.

More Jev resources like this

SwiftJev

Swift provider adapter that sends typed noul, choice, and score prompts to the hosted System One API through SwiftDecision DecisionBackend, returning validated probability distributions.