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

chat2jev

Workbench that turns OpenAI chat requests into Jev state and questions, compares both outputs, and publishes question sets as proxy routes.

ProxyprototypeMIT
View on GitHub ↗

Overview

Chat2Jev is an independent, experimental workbench for local or trusted-network use. You paste a Chat Completions request as JSON, a message array, plain text or a cURL command, convert it into Jev State and Questions, edit them, and compare the chat model's text output with Jev's typed judgments and response times side by side. Question packs and drafts are saved in the browser.

A proxy lab publishes reusable question sets as routes matched by slug, a jev: model name or a request fingerprint, returning JSON or tool calls in a Chat Completions-shaped response, so existing chat clients can get typed decisions. Built-in examples cover ticket triage and tool routing. The proxy is partially compatible and does not support streaming. The interface is mainly in Chinese.

What the decision model decides

TypeQuestionAnswers
ChoiceWhich department should handle this ticket?—
Yes / noDoes the customer want a refund?yes · no
ChoiceWhich tool should be called? (tool-router example)—

How it works

  1. Paste a chat request
  2. Convert to State and Questions and edit
  3. Compare chat and Jev results and timings
  4. Publish the question set as a proxy route

Strengths

  • Side-by-side chat vs typed comparison
  • Lets chat-shaped clients use Jev
  • Keys entered locally

Caveats

  • Experimental; no streaming
  • Interface mainly Chinese

Getting started

git clone https://github.com/Chandler-Sun/chat2jev.git, run it locally, and enter your OpenAI-compatible and TypeSafe keys in settings.

At a glance

Type
Proxy
Status
prototype
Runtime
local web app + proxy
Has code
yes
License
MIT

Use cases and integrations

How curated lists describe it

  • Convert OpenAI-compatible Chat Completions requests into TypeSafe System One (Jev) \*\*State / Questions\*\*, compare generated text with structured judgments, and publish reusable question sets as proxy routes.

    logicrw/awesome-jev-projects

  • Convert legacy chat completion API request to Typesafe jev API.

    hellogumbo/awesome-jev

Frequently asked questions

What does it convert?

Chat Completions requests into Jev State and Questions.

Can my chat client call Jev?

Yes, through proxy routes that return Chat Completions-shaped responses.

Does it stream?

No, streaming is not supported.

Related projects

  • decision-1-vs-jev

    C# System One client and benchmark comparing Microsoft-Decision-1, Jev and open-jev on the same labelled support tickets.

    LibraryprototypeC#
  • jev-workbench

    Local service to define, test and publish versioned Jev judgment functions, callable over HTTP by backends and over MCP by agents.

    MCP serverbetaTypeScript
  • judge_rails

    Ruby gem that stores Jev Noul, Choice and Score answers about a record's text as ordinary ActiveRecord columns you can query.

    LibrarybetaRuby
  • system-one-adapter-rust

    Rust port of TypeSafe's system-one-adapter that answers System One questions with OpenAI or Anthropic LLMs for comparison with Jev.

    ProjectbetaRust
  • systemone

    Unofficial Go client for TypeSafe System One with generics that bind each choice, score and noul answer to a declared Go type.

    ProjectbetaGo
  • here-we-go-jev

    One-page playground and test bench for Jev with TypeSafe, OpenRouter, LLM-baseline and mock backends behind one /v1/systemone contract.

    ServiceprototypeTypeScript