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

typesafe-jev-bridge

Zero-dependency Node bridge that exposes Jev through an OpenAI-compatible API, plus a CLI, code-audit tool and auto-fix loop.

CLIbeta
View on GitHub ↗

Overview

typesafe-jev-bridge makes Jev usable from any tool that speaks the OpenAI API, such as 9Router, Claude Code, Cursor, Cline or an OpenAI SDK. It is a small zero-dependency Node.js bridge, plus a CLI for asking noul, choice and score questions, a code-audit tool and an auto-fix loop.

The README shows real output: a billing ticket routed with confidence 1, and an ambiguous message where Jev split between billing and technical with confidence 0.29, which the CLI flags as low confidence (below 0.55) and exits with code 3 so scripts can react. It is an unofficial community project and requires your own TypeSafe key. Score questions over user-defined rubrics are also supported, and text can be passed from a file to avoid shell quoting problems with characters such as dollar signs.

What the decision model decides

TypeQuestionAnswers
ChoiceWhich team should handle this?billing · technical · sales
ScoreScore over a user-defined rubric—
Yes / noYes/no likelihood questionyes · no

How it works

  1. Run the bridge or CLI with a TypeSafe key.
  2. Send state and typed questions via CLI flags or the OpenAI-compatible endpoint.
  3. Read answers with confidence.
  4. Low confidence exits with code 3 for scripts to handle.

Strengths

  • Plugs Jev into OpenAI-compatible tools.
  • Exit code signals low confidence.
  • No dependencies.

Caveats

  • Unofficial.
  • Mapping a decision model onto a chat API is a workaround and may surprise some tools.

Getting started

node ask-jev.cjs --text '...' --q "Which team should handle this?" --type choice --criteria "billing=...,technical=..."

At a glance

Type
CLI
Status
beta
Language
JavaScript
Runtime
Node CLI and HTTP bridge
Has code
yes

Use cases and integrations

How curated lists describe it

  • Use the TypeSafe Jev decision model (System One) anywhere: zero-dependency OpenAI-compatible bridge for 9Router, Claude Code, Cursor, Cline & any OpenAI SDK. Typed yes/no, choice & score judgments via CLI or HTTP.

    hellogumbo/awesome-jev

Frequently asked questions

Why an OpenAI-compatible bridge?

So tools that only speak the OpenAI API can call Jev.

What happens on low confidence?

The CLI prints the probabilities and exits with code 3.

Is it official?

No, it is a community project.

Related projects

  • TanStack AI

    Provider-agnostic TypeScript AI SDK; per list sources, an @tanstack/ai-typesafe adapter exposes Jev decisions through decide().

    ServiceproductionTypeScript
  • claude-jev-mod

    Claude Code mod that adds $.jev to the engine interface so other plugins and hooks can ask Jev typed questions through six providers.

    ServicebetaTypeScript
  • Ax

    Ax, a DSPy-style multi-language LLM framework, listed with a native TypeSafe provider for Boolean, Choice, Score and raw Jev questions.

    PluginproductionTypeScript
  • Effect

    Effect TypeScript library whose @effect/ai-typesafe package maps classify, probability and rating operations to Jev questions.

    ProjectproductionTypeScript
  • instructor-php

    PHP monorepo whose Polyglot module adds a TypeSafe driver for typed Noul, Choice and Score decisions alongside LLM inference.

    ProjectproductionPHP
  • req\_llm

    Elixir library for unified LLM calls across providers, with an evaluate interface that calls Jev via TypeSafe or OpenRouter.

    ProjectproductionElixir