Which production placements do the panels cover?

Support triage, RAG relevance gating, agent action firewall, inline moderation, CI eval judging, plus a raw console for custom state and questions.

How is the API key kept out of the browser?

The server proxies /api/jev with TYPESAFE_API_KEY from the environment or shell secrets file, so the browser never holds the key directly.

Do I burn an API call every time I move a threshold?

Workflows 02, 03, and 05 re-render from the cached response as you drag, including flipping blocked to hold for approval when the gate exceeds confidence.

Can a single score be trusted for pass or fail?

Yes. The digest notes a correct terse answer and a correct complete answer landing on opposite sides of a 1.5 bar, documented as Jev 1.13 jaggedness.

How do I run the playground?

Run node server.mjs in the repository directory, serving http://localhost:7171. It needs Node 18+ for global fetch and has no dependencies or install step; set a different port with PORT=8080 node server.mjs.

How do I add my own workflow?

Append one object to WORKFLOWS in public/app.js, giving it a build(vals) that returns {state, questions} and either decide(answers, gate) for the default card grid or custom: true with your own render(data, vals, gate).

Which panels re-render from cache on threshold moves?

Workflows 02 (RAG relevance gate), 03 (agent action firewall), and 05 (CI eval judge) have a threshold slider that re-renders from the cached response without refetching. On 03, dragging the gate above the model's confidence flips BLOCKED to HOLD FOR APPROVAL.

CommunityJev Demos & Experiments

Jev Playground — AviroopPaul

Zero-dependency local playground showing six Jev workflows from triage to eval judging with shippable snippets.

See highlights

Open github.com

Stars: 0
Language: JavaScript
Published by: Community
Author: AviroopPaul
Last verified: Oct 2026

AviroopPaul/jev-playground

User repository on GitHub

View on GitHub

A playground for TypeSafe AI's Jev (System One model), built around five real production workflows: support triage, RAG relevance gating, agent action firewall, inline moderation, and CI eval judging.

Forks: 0
Last pushed: Updated Sep 2026

Repo stats from the GitHub API, cached Oct 2026.

This is one of several Jev listings for the same project. The canonical entry is Jev Playground — Little-Planet-Labs.

About this project

A local playground for Jev, TypeSafe AI's System One model, built around the places it actually fits in a production system rather than around its API surface. Jev is not a chat model. You hand it state plus a set of typed questions, and it answers all of them in one non-autoregressive pass, returning typed values with calibrated probabilities instead of prose you have to parse. That property is what each of the five workflows below is built to show.

Quoted from the project's README. Read the full README.

Tagscommunity github

Works forJev for Evaluation Jev for Tool calling Jev for Classification Jev for Moderation

Used inJev for Software & DevTools

Languages: JavaScript

Highlights

Runs node server.mjs with no install step on Node 18+ and configurable PORT.
Decides five ticket fields, twelve RAG candidates, or whole eval suites in one round trip.
Renders pipeline strips plus downstream actions plus copyable integration snippets.
Caches responses for threshold sliders and documents sharp retrieval yet jagged grading.

Watch out

Demo panels with author-measured timing on a home connection; grading can be jagged so calibrate bars on trusted labels. Copy reviewed against the repository README on 2026-10-04, not from running it.

Questions about this entry

Short answers drawn from the linked project or page.

More Jev resources like this

tev1-50-use-cases

Fifty runnable tev1 decision-model examples through local Ollama, from ticket triage and guardrails to six playable games, with a live request viewer and scoreboard.

Jev Demos & Experiments#community#github

Community

PromptEngineer48/tev1-50-use-casesPython

jev-browse

Browser agent where Jev makes every navigation decision and agent-browser performs clicks and typing, stopping when the named goal page is reached or blocked.

Jev Demos & Experiments#community#jevbest#catalog-import

Community

kyrylosyzonenko/jev-browseJavaScript

jev-browser

A browser MCP server where Jev picks the next page action from locally ranked candidates in roughly 300 ms without LLM tokens, deterministic state diffs verify the step, and the driving agent steps in only on escalation.

Jev Demos & Experiments#community#jevbest#catalog-import

Community

MahmoudAdelbghany/jev-browserJavaScript

computer-use-cache

Drop-in OpenAI-compatible cache proxy for repeatable agent workflows: exact hits stay local, and with a TypeSafe key Jev judges whether a cached response is semantically reusable for the current request.

Jev Demos & Experiments#community#jevbest#catalog-import

Community

rohanarun/computer-use-cacheJavaScript

jev-kit

Claude Code kit that guards PreToolUse calls with a free pre-filter plus one Jev question for ambiguous cases, and adds sub-agent sizing, rechecks, search, daemon speedups, and monitoring.

Jev Demos & Experiments#community#awesomejev#catalog-import

Community

jonathanavis96/jev-kitPython

robodiag-harness

Terminal harness that checks ROS 2 robots with deterministic health tests, capability descriptions, SQLite history, and optional Jev evidence routing.

Jev Demos & Experiments#community#jevbest#catalog-import

Community

YueBit/robodiag-harnessPython

From the Jev guides

Original write-ups on what this entry is for.

Per-Tick Decisions in Games

Sep 28, 2026

A game tick is a state plus a legal action list. How Clash Royale, Minecraft, and the earlier Mario, Pokémon, StarCraft, and drone loops ask Jev one step at a time.

Jev Demos & Experiments#gaming#agents#control

Read article

Back to all resources

From the Jev community

Posts from builders shipping with Jev right now.

Follow @typesafeai