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

jev-cli

Thin, zero-dependency Python CLI for Jev's models and evaluate endpoints, built for use with a Grok-based Jev router bot.

CLIprototype
View on GitHub ↗

Overview

jev-cli by badry-dev is an unofficial thin command-line client for TypeSafe System One (Jev) that calls the HTTP API directly using only the Python standard library, so agents and scripts can call Jev without installing the official SDK. The author built it mainly for use with a Jev router bot on Grok.

It lists models via GET /v1/models and asks questions via POST /v1/systemone, either from a full request file or with state and questions passed separately, including structured JSON state. It reads TYPESAFE_API_KEY or JEV_API_KEY from the environment or a permission-restricted local file, and recommends pinning a model version instead of the moving jev-latest alias once thresholds are tuned. A bundled skill folder carries its own MIT licence. Model listing is supported.

What the decision model decides

TypeQuestionAnswers
OtherSends user-defined Noul, Choice or Score questions to Jev—

How it works

  1. Set TYPESAFE_API_KEY or a key file
  2. jev models to list models
  3. jev ask with a request file or state and questions
  4. Pin a model version for stable thresholds

Strengths

  • No dependencies
  • Simple for agents and scripts
  • Encourages version pinning

Caveats

  • Unofficial
  • Minimal feature set

Getting started

`python3 -m pip install -e .` then `jev ask --request examples/ticket.request.json`.

At a glance

Type
CLI
Status
prototype
Language
Python
Runtime
CLI
Has code
yes

Use cases and integrations

How curated lists describe it

Frequently asked questions

Why not use the official SDK?

This CLI is dependency-free so agents can call Jev without installing it.

Which endpoints does it use?

/v1/systemone for evaluation and /v1/models for listing.

Should I pin a model?

The README recommends pinning once thresholds are tuned.

Related projects

  • ai

    Hack Club's lightweight LLM proxy, listed for adding a Jev forwarding endpoint behind its auth, limits and usage logging.

    MCP serverproductionTypeScript
  • 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
  • jev-mcp (freepik-company)

    Go MCP server giving any client typed Jev decisions via OpenRouter or TypeSafe, with validated answers and no paid retries.

    MCP serverproductionGo
  • llm-typesafe

    Simon Willison's LLM plugin that adds Jev as a model for yes/no, choice and score questions from the command line.

    PluginproductionPython