JevMade Sign in
← Back to experiments

Agent tooling

jeff

A Go CLI that asks Jev a question from the shell and turns thresholds into exit codes your scripts and CI jobs can act on.

Bookmark: jeff Keep this in your collection.
Leave a noteWhat would you try with this? : jeff

Only you can see your notes.

Source screenshot of jeff
SOURCE SCREENSHOTFull screenshot ↗

What it does

Rank mode scores every item on every weighted dimension of a YAML spec in one request and sums the weights; the name is a 22 Jump Street joke.

How you can use it

Start by listing project ideas or work items you want to compare. Write down several questions with fixed answer levels to evaluate them, such as expected revenue or required effort. Assign each question an importance value, keeping the calculations in your own plan rather than asking the system to do math.

Your developer can run the tool to score each item across your questions in one step. To connect, you will need an access key from TypeSafe. Keep in mind that the underlying model cannot count or compare dates reliably, and misleading text in your input can skew its answers.

Keep this for later

Sign in to bookmark experiments, guides and videos, and keep notes only you can see.

Continue to sign in

We’ll bring you back to this listing.