Skip to content
System One

BeLazy167/typesafe-mod

Claude Code mod that routes decisions to Jev: it ranks installed skills against each prompt and answers the agent's own this-or-that questions when confidence is high enough, otherwise falling back to asking the user.

BeLazy167/typesafe-mod on GitHub

Stars
3
Forks
0
Language
TypeScript
License
MIT
Last push
17 Sep 2026

Read from GitHub on . Counts change daily.

How it uses Jev

/** Key under which the session's roster is cached in `$.store`. */
export const ROSTER_KEY = 'skill-roster.v1';

export const DEFAULTS = {
  endpoint: 'https://api.typesafe.ai/v1/systemone',
  model: 'jev-latest',
  /** Below this Choice confidence, suggest nothing. Tune on your own turns. */
  minConfidence: 0.6,
  /** Below this Noul, the turn does not want a procedure at all. */
  minNeedsSkill: 0.5,
  // A cap that is too low drops real skills without saying so. At 120 this
  // list lost `resolving-merge-conflicts`, and the router then said nothing
  // about merge conflicts. This sits well above the 194 skills seen here.
  // Raise it if a scan reports hitting it.
  maxSkills: 500,

View in hooks/suggest.ts

Open on GitHub

More like this

Use cases this is tagged with