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

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 skillbetaMIT
View on GitHub ↗

Overview

jev-anything is an independent agent skill for adding a bounded Jev decision layer, such as a classifier, router, scorer or guardrail, to a real software workflow. It requires the coding agent to define the decision contract, criteria, error behaviour, fallback and evaluation plan before writing code, and per the MrJev list it writes a client scaffold for you.

The repository root is the skill folder and is cloned into the agent's skills directory for Codex or Claude Code, or attached as project instructions for other agents. Access can be through TypeSafe's direct API or OpenRouter, with reference files documenting endpoints and model aliases and a reminder to recheck live docs since they change. It states it is not an official TypeSafe or OpenRouter product and points to TypeSafe's own MIT-licensed official skill. It is MIT licensed.

What the decision model decides

TypeQuestionAnswers
OtherGuides the agent to define Choice, Score or Noul decision contracts with criteria, fallbacks and evals(inferred)—

How it works

  1. Install the skill into the agent's skills folder
  2. Agent defines the decision contract, criteria, errors, fallback and eval plan
  3. Agent scaffolds the client and implements the decision
  4. Test and tune against the evaluation plan

Strengths

  • Forces design and evaluation before code
  • Works with TypeSafe direct or OpenRouter
  • Reference docs for API details

Caveats

  • Independent, not the official skill
  • API details may drift; recheck live docs

Getting started

`git clone https://github.com/grandamenium/jev-anything.git ~/.claude/skills/jev-anything` (or ~/.codex/skills), then set TYPESAFE_API_KEY or OPENROUTER_API_KEY.

At a glance

Type
Agent skill
Status
beta
Runtime
agent skill
Has code
yes
License
MIT

Use cases and integrations

How curated lists describe it

  • Agent skill for designing, building, testing, and tuning bounded JEV decision layers

    logicrw/awesome-jev-projects

  • Agent skill for designing, building, testing and tuning a bounded decision layer, with a client scaffold it writes for you.

    MrJev/awesome-jev

Frequently asked questions

What does the skill make the agent do?

Define the decision contract, criteria, error handling, fallback and evaluation plan before coding.

Is it the official TypeSafe skill?

No; the README points to typesafe-ai/skills for the official one.

Which access paths are supported?

TypeSafe direct API or OpenRouter.

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-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
  • 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
  • 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
  • typesafe-ai/skills★ 2,685

    Official TypeSafe agent skill and Claude Code plugin that teaches coding agents to design System One workflows with typed judgments.

    Agent skillproduction