jev-lab
Spanish-language progressive learning lab for Jev built with spec-driven development, starting with a hexagonal skeleton.
Overview
jev-lab (mberliner) is a progressive, Spanish-language learning project for Jev's Choice, Score and Noul primitives, from basics to production techniques. Each delivery is a runnable vertical slice (a CLI, an API or a report) with a lesson on what it teaches, how to run it and what can fail.
Its guiding ideas are that the model judges and code decides (arithmetic, dates, permissions and policy stay in code), that every Choice has an escape option and only answers above a risk-based confidence threshold are automated, and that evidence comes before claims (offline tests without keys by default, with cost-capped live tests). It is built with GitHub's Spec Kit through specification, plan, tasks and implementation. The current state is Spec 000, foundations: a reproducible environment, a hexagonal skeleton, pre-commit and CI quality gates and a doctor command. It runs on Linux and Windows with Claude Code or Antigravity.
What the decision model decides
| Type | Question | Answers |
|---|---|---|
| Choice | Example Choice with an escape option(inferred) | — |
How it works
- Clone and uv sync
- Run tests and jevlab doctor
- Follow lesson documents per spec
- Add live tests with cost caps
Strengths
- Rigorous, spec-driven structure
- Offline tests by default
- Clear design principles
Caveats
- Only the foundation spec is complete
- Spanish documentation
Getting started
`git clone <repo> jev-lab && cd jev-lab && uv sync && uv run poe test && uv run jevlab doctor`.At a glance
- Type
- App
- Status
- prototype
- Language
- Python
- Runtime
- CLI (uv)
- Has code
- yes
- Has tests
- yes
Use cases and integrations
How curated lists describe it
Laboratorio progresivo para aprender y demostrar Jev, el modelo de decisión…
wh000wh000/awesome-jev-live
Frequently asked questions
What is it?
A step-by-step Spanish learning lab for Jev.
How far along is it?
Spec 000, the foundations, is done; further specs are on the roadmap.
Do tests need a key?
No, default tests run without network or keys.
Related projects
Korean course code for attaching Jev to Claude Code, Codex and OpenClaw: skills, hooks, routers and open-model comparisons.
ProjectproductionJavaScriptClaude skill that gives a ten-line gut-check on whether Jev could help with what you are building, citing tested patterns.
Agent skillbetaMarkdownThirty worked Jev uses in ten levels, from a smart if statement to a pi coding agent that writes its own Jev questions.
ProjectbetaTypeScriptAgent skill that makes a coding agent try Jev first for bounded judgments like classify, rank or gate, and log each attempt.
Agent skillbetaPythonField guide site on decision models built from two Jev projects: a code, decision model, LLM, human cascade from ELI5 to engineering.
ArticleproductionPythonAwesome list and guide to System One models (Jev, Laya and others) organized around a five-stage agent lifecycle, with comparison tables.
Projectproduction