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

jev-mode (tiffygk)

Claude Code and Codex skills plus a study course for building with Jev, including Jevaluate, which rates how well projects use Jev.

Projectbeta
View on GitHub ↗Live demo ↗

Overview

Jev Mode is a plugin marketplace for Claude Code and Codex with skills and learning material for building with TypeSafe's Jev, drawn from TypeSafe's cookbooks and documentation. Not every skill calls Jev. Jevaluate reads a Jev project's code and rates how well it uses Jev, with provenance for each finding; an LLM rates and code checks the rating. A private variant rates unpublished code, and a harness and eval skill tune and test the rubric.

Jev Lens turns images into a neutral JSON state that Jev can read, calling Jev only for an optional check. Jev Sources finds the relevant TypeSafe docs and cookbook sections by keyword and meaning and tells Claude to read them before making claims. A five-level study course summarises cookbooks and docs with quizzes, available as a Claude artifact or on GitHub Pages. A Ratings folder lists rated Jev projects and the evaluation methodology.

What the decision model decides

TypeQuestionAnswers
OtherHow well does this project use Jev (rubric rating by an LLM)?—
Yes / noOptional Jev check of a finished image state(inferred)yes · no

How it works

  1. Add the marketplace in Claude Code or Codex
  2. Install jevaluate, jev-lens or jev-sources
  3. Run a skill, for example /jevaluate-private on a folder
  4. Use the study course to learn Jev concepts

Strengths

  • Grounds agent answers in TypeSafe's own docs
  • Structured rating rubric with provenance
  • Learning path with quizzes

Caveats

  • Most skills do not call Jev; ratings come from an LLM
  • Jev Sources downloads documentation locally
  • GitHub Pages course uses an untested keyword check

Getting started

claude plugin marketplace add tiffygk/jev-mode, then claude plugin install jevaluate@jev-mode

At a glance

Type
Project
Status
beta
Runtime
Claude Code and Codex plugins / skills
Has code
yes

Use cases and integrations

How curated lists describe it

  • Claude Code/Codex skills and learning resources that make it easy to build with Jev, TypeSafe's System One model. All materials pull directly from their canonical cookbooks and documentation.

    hellogumbo/awesome-jev

Frequently asked questions

What is Jevaluate?

A skill that reads a Jev project's code and rates how well it uses Jev, with provenance for every finding.

Do these skills call Jev?

Mostly not; Jev Lens can optionally call Jev to check a finished state.

Where is the course?

As a Claude artifact (recommended) or on GitHub Pages.

Related projects

  • Building with Jev★ 155

    Agent skill for writing and improving programs that call Jev: question design, state structure, thresholds and diagnosing wrong answers.

    Agent skillproduction
  • jev-agent-toolkit

    Portable Agent Skill teaching coding agents to use Jev as a bounded judgment layer, plus an optional MCP bridge.

    MCP serverbetaTypeScript
  • jev-anything

    Independent agent skill that makes a coding agent design, build, test and tune a bounded Jev decision layer with a contract first.

    Agent skillbeta
  • jev-mcp (arunav25)

    MCP server exposing Jev as one evaluate tool, plus a harness comparing Jev's judgments with general LLMs on shared datasets.

    MCP serverbetaJavaScript
  • jev-skill (marcodicesare-dev)

    Community Jev skill for Claude Code and Codex: a guide, a dependency-free Python CLI and 5 measured question sets.

    Agent skillbetaPython
  • jevriel

    Skill and plugin for Codex, Claude Code and other agents that helps them use Jev for bounded judgments and measure the effect.

    Agent skillbetaJavaScript