Skip to content

TypeSafe Showcase

Six browser demos of Jev's three primitives, each running the same judgment past Gemini or Claude for a side-by-side, with visitors pasting their own keys into localStorage instead of sharing a server-side key.

Stars
1
Language
TypeScript
Published by
Community
Author
Ashadeepa
Last verified
Oct 2026

Next.js UI showing off TypeSafe's System One model (Jev) — parallel Noul judgments and a Choice-based citation checker, deployable to Vercel

Forks
0
Last pushed
Updated Oct 2026

Repo stats from the GitHub API, cached Oct 2026.Homepage

About this project

Six live demos of TypeSafe's System One model (Jev) — each pairing a real interaction with a real judgment, and each one checkable against Gemini or Claude in the open.

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

Highlights

  • Routes cover all three primitives: 16-ticket noul batching, 8 three-way claim checks, and a 10-message escalation rubric.
  • No shared key on the server; every Server Action builds its own client from the key the visitor supplied.
  • The tone demo is the honest one: Jev and Gemini agree only about 60% of the time on escalation scoring.
  • One demo races you against Jev's measured latency, reported around 470ms on average, instead of a countdown.

Watch out

Nothing runs without the visitor pasting a key, so you cannot read the demos without a TypeSafe account and an OpenAI or Anthropic account for the side-by-side. Although the calls execute server-side, the key travels from the browser into each Server Action as a parameter and is only kept in localStorage client-side. The comparison strip costs real money: every demo also calls Gemini or Claude with the visitor's second key at that provider's published per-token rates, and the result is never used for what the demo itself does. The tone rubric is the one route where the models genuinely diverge, so its numbers describe a disagreement as much as a quality gap. 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

98GitHub stars
Self-hosted Docker service and web UI that packages Laya structured-decision inference behind a compatible System One endpoint.
Jev Demos & Experiments#community#jevbest#catalog-import
20GitHub stars
General-purpose System One agent where Jev chooses tools and arguments while Jot executes them and loops until it can reply.
Jev Demos & Experiments#community#awesomejev#catalog-import
Browser drone-fleet lab where TypeSafe Jev steers each drone as fast reflexes and an optional planner offers one-use strategy advice when confidence drops, without pausing control.
Jev Demos & Experiments#community#jevbest#catalog-import
23GitHub stars
Chrome extension that sends each X post to Jev with six typed questions in one call near the viewport, then renders a behavior chip under the post plus a session cost and latency panel.
Jev Demos & Experiments#community#jevbest#catalog-import
14GitHub stars
Manifest V3 Chrome extension in TypeScript that sends each LinkedIn post to Jev and collapses bait, self-promo, off-topic, and sponsored posts into a one-line bar with reasons and scores.
Jev Demos & Experiments#community#jevbest#catalog-import
11GitHub stars
Standalone Flue example that routes support chat through Cloudflare Workers AI, using gpt-4o-mini for chat and Jev for typed intent and urgency routing.
Jev Demos & Experiments#community#awesomejev#catalog-import

From the Jev guides

Original write-ups on what this entry is for.

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