Author test

jev-skill-router

Test whether Jev routing helps Claude Code choose a skill

shimo4228 · Sources checked 2026-10-09

Added to this collection 2026-10-09

How this project uses Jev

Input: Each prompt and the installed skills’ names and descriptions; the narrow stage also sends the full SKILL.md bodies for the top three
Jev decides: Choice ranks candidates and Noul thresholds gate the request, then Jev selects from a shortlist; the plugin can log in shadow mode or append a suggestion in inject mode
Code executes: Includes scripts that join decision logs to transcripts to check whether Claude Code later called the suggested skill; the author logged 1,242 decisions and 539 suggestions over a week, followed by 28 calls

Evidence and limitations

The MIT license, routing and logging code, and the author’s one-week measurement were reviewed without installing the plugin or reproducing results. About 5% of suggestions in that sample were followed by a call, and the author removed the router; this single setup and short experiment do not prove other environments behave the same. Prompts and skill descriptions are sent to TypeSafe, and the code does not redact secrets pasted into a prompt.

This project has not been run independently here. Author-reported results are not independently verified results.

Original sources

Explore related use cases

Browse this category →