# Jev Triage

> Sort a batch of similar items into labelled buckets and flag only the ones worth a second look. Use when you are about to make the same judgement call dozens of times — review findings to confirm, failing tests to classify as flaky or real, issues to route, diff hunks to rate for risk — and especially when the alternative is spawning a subagent per item. Returns a label plus a calibrated confidence for each item; anything under the confidence floor is handed back for you to decide properly. Do NOT use for a single judgement, or when each item needs its own investigation rather than a classification.

- Skill: `wanlanglin/jev-triage` (Agent Skill)
- Install (CLI): `npx skillmds@latest add wanlanglin/jev-triage`
- Raw SKILL.md: https://api.skillmd.com/api/skills/wanlanglin/jev-triage/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- License: MIT
- Author: WanLanglin (https://skillmd.com/u/wanlanglin)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/wanlanglin/jev-triage

---


# jev-triage

The value here is not the labels — you can label things yourself. The value is
that the obvious cases cost nothing, so you only spend real reasoning where it
is needed. **Read the confidence caveat below before you trust the split.**

## Use it

```bash
python3 "$CLAUDE_SKILL_DIR/../../scripts/triage.py" \
  --question "Is this review finding a real defect?" \
  --label real="a genuine bug that would cause wrong behaviour or a crash" \
  --label style="naming, formatting, or preference only" \
  --label wrong="the finding misreads the code and is not a defect" \
  --escalate-below 0.9 \
  --items findings.txt
```

Multi-line items (stack traces, diff hunks, findings with code):

```bash
python3 .../triage.py --sep $'\n---\n' --items hunks.txt \
  --question "How risky is applying this hunk automatically?" \
  --label safe="isolated, easy to revert, well covered by tests" \
  --label review="touches shared state or public API" \
  --label danger="security, auth, migrations, or data loss"
```

Add shared background once with `--context` — it is billed once for the whole
batch, not per item, so be generous with it.

## The part that matters

Output splits into decided buckets and an **escalate** list. The escalate list
is the whole point: those are the items where a confident-looking answer would
have been wrong. Send exactly those to a subagent or handle them yourself, and
accept the rest.

`--escalate-below` defaults to **`0.90`**, and that number is measured, not
chosen for feel. On a 500-decision labelled task, `choice` confidence was close
to uninformative between 0.4 and 0.9 — accuracy wandered between 26% and 50%
across five bins — and only separated above 0.9, where it reached 80.9%
(ECE 0.226).

| escalate below | automated | accuracy of that automated set |
|---:|---:|---:|
| 0.7 | 77% | **65%** — one in three wrong |
| 0.9 | 41% | **81%** |

So: **do not lower this without measuring on your own data.** `choice`
confidence is meaningfully weaker than the `noul` probability `jev-shortlist`
uses. Even at 0.9 you are accepting roughly one error in five, which is fine for
triage and not fine for anything irreversible. Full curve and caveats:
`references/calibration.md`.

## Cost

Up to 256 items per call, content billed once. A 256-item batch runs well under
a cent and under a second. Compare that with one subagent per item.

## Setup

```bash
export TYPESAFE_API_KEY=...   # from typesafe.ai
```