Skip to content
Metatext Decide — typed decisions for your agents, with a probability on every answerGet early access
Metatext
Sign Up

Jev AI Tools

Educational site with six bounded recipes and a custom builder for AI SDK Choice, Score and Boolean evaluations, exportable as TypeScript.

Service
Open source ↗Live demo ↗

This summary is based on how curated lists describe the project; its README was not available to review.

Overview

Jev AI Tools at jevai.tools is a developer-education site listed in yibie/awesome-jev. It hosts six bounded recipes plus a custom builder for Choice, Score and Boolean evaluations written against the AI SDK. Each recipe displays the answer probabilities separately from the provider's confidence value, and uses deterministic local thresholds to pause uncertain routes for human review.

Every configuration can be exported as TypeScript, so a recipe built on the site can be dropped into an application. It is useful for developers who want to learn how to set thresholds and review paths around typed decisions. No README or source was available, so the recipe contents and hosting details are not documented here.

What the decision model decides

TypeQuestionAnswers
ChoiceRecipe-defined Choice over bounded options(inferred)—
ScoreRecipe-defined Score on an ordered rubric(inferred)—
Yes / noRecipe-defined Boolean check(inferred)true · false

How it works

  1. Pick a recipe or open the custom builder
  2. Run an evaluation through the AI SDK
  3. Compare answer probabilities with provider confidence
  4. Apply local thresholds; pause uncertain routes for review
  5. Export the configuration as TypeScript

Strengths

  • Teaches thresholding and review paths explicitly
  • Exportable TypeScript configurations

Caveats

  • No README or source available
  • Third-party educational site

At a glance

Type
Service
Language
TypeScript
Runtime
web app (AI SDK)

Use cases and integrations

How curated lists describe it

  • Developer education: hosts six bounded recipes plus a custom builder for AI SDK `Choice`, `Score`, and `Boolean` evaluations; recipes display answer probabilities separately from provider confidence and use deterministic local thresholds to pause uncertain routes for review, while every configuration exports as TypeScript.

    yibie/awesome-jev

Frequently asked questions

What is on the site?

Six recipes and a builder for AI SDK Choice, Score and Boolean evaluations.

Why separate probability from confidence?

So users see both the answer distribution and the provider's confidence value before setting thresholds.

Can I reuse a recipe?

Yes, each configuration exports as TypeScript.

Related projects

  • jev-orchestrator

    Python library to batch, gate, trace and calibrate thousands of typed Jev decisions, maximising questions per call.

    ProjectprototypePython
  • jev-pipeline-skill

    Agent skill and CLI that chains Jev question design: find decision calls, check fit, draft, lint, calibrate thresholds and plan a cascade.

    Agent skillprototypePython
  • discern

    Effect library turning Jev Choice, Noul and Score answers into typed branches with a compiler-enforced Uncertain case.

    LibrarybetaTypeScript
  • typesafe-ai-rails

    Community Rails integration for System One with configuration, persisted usage and cost telemetry, and fail-closed confidence policies.

    LibrarybetaRuby
  • jev-playground

    Runnable Jev cookbooks, a plain-English guide and sixteen interactive simulations of how a typed decision model changes architecture.

    AppbetaPython
  • typesafe-ai-jev-example

    Chinese-language tutorial repo with six runnable Python demos and four notes on Jev primitives, with offline mock mode and live samples.

    ServiceprototypePython