Small decisions. Interesting possibilities.Submit contentSubmit

How it uses Jev

A user task like "which of these 187 files handle auth" becomes a fixed candidate set in the state and one typed question per candidate, sent in parallel to Jev. The labels, yes/no answers or scores come back with their confidence, and only the shortlist enters Claude's context, so the expensive model reads four paths instead of 187 files.

What Jev decides

File shortlist for one task (illustrative)Example answers · not a recorded Jev response · Source ↗
Question 1 · f042 auth
YOUR APP
INSTRUCTION

Is this file part of the authentication flow the task asks about?

STATE

The state carries one candidate file, its path and a slice of its content, and the task being answered. The documented benchmark sends 187 Hono files for one discovery task; three of the question set are shown here.

JEV · NOUL
YesNo
Question 2 · f043 auth
YOUR APP
INSTRUCTION

Is this file part of the authentication flow the task asks about?

STATE

The state carries one candidate file, its path and a slice of its content, and the task being answered. The documented benchmark sends 187 Hono files for one discovery task; three of the question set are shown here.

JEV · NOUL
YesNo
Question 3 · f044 auth
YOUR APP
INSTRUCTION

Is this file part of the authentication flow the task asks about?

STATE

The state carries one candidate file, its path and a slice of its content, and the task being answered. The documented benchmark sends 187 Hono files for one discovery task; three of the question set are shown here.

JEV · NOUL
YesNo

App workflow

  1. Recognise the shape

    Claude uses the skill when a task means reading a lot to decide a little, such as finding files, triaging logs or routing tickets.

  2. Build the candidate set

    The skill collects the pile to judge, from a directory listing, a log file or a ticket batch, and keeps each item's identity in the state.

  3. Ask in parallel

    One typed question per candidate goes to Jev with the task text, and the answers return with their confidence.

  4. Return a shortlist

    Only the items that pass enter Claude's context, with their confidence, so the model spends tokens on the answer rather than the pile.

Why it is interesting

It turns context waste into a bounded decision problem with a measured trade: the published benchmark cuts Claude's tokens by a median 82% and matches its accuracy on eight of twelve real tasks, while two log-triage cases show the accuracy actually falling when the cheap shortlist misses.