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

typesafe-ai-playground

Community web playground with 110 runnable Jev examples in 22 categories, including A/B comparisons, games and dilemmas.

Appbeta
View on GitHub ↗

Overview

This is an independent community playground for exploring TypeSafe AI. It ships 110 runnable examples across 22 categories, including 41 A/B comparisons, practical use cases, party games, dilemmas and reasoning challenges. You pick an example, inspect and edit its input and questions, and run it through the API.

It runs with Python 3.10+ and a browser, with no packages to install or build step. The key is entered at a hidden terminal prompt and kept in the server process, not saved or sent to the browser; you can also browse and edit without a key. The layout is designed for screen sharing on Discord, and a phone layout splits the flow into browse, build and results stages. The server only listens on localhost.

What the decision model decides

TypeQuestionAnswers
ChoiceExample-specific Choice questions (games, dilemmas, use cases)—
OtherA/B comparisons of two question or input variants—

How it works

  1. Run python3 run.py and enter a key.
  2. Choose an example from a collection or category.
  3. Edit input and questions.
  4. Run it and compare results, including A/B variants.

Strengths

  • Large library of ready examples.
  • No dependencies or build step.
  • Key never reaches the browser.

Caveats

  • Needs a TypeSafe key for evaluations.
  • Community project, not official.

Getting started

git clone https://github.com/nickthompson480/typesafe-ai-playground.git && cd typesafe-ai-playground && python3 run.py

At a glance

Type
App
Status
beta
Language
Python
Runtime
local web server + browser
Has code
yes

Use cases and integrations

How curated lists describe it

  • Community TypeSafe AI playground: 110 use cases, games, dilemmas and model challenges, with editable prompts, A/B comparisons and a mobile-friendly UI.

    hellogumbo/awesome-jev

Frequently asked questions

How many examples are included?

110 examples in 22 categories, including 41 A/B comparisons.

Do I need to install anything?

Only Python 3.10+; there are no packages or build steps.

Where is the key stored?

In the running server process only, not in files or the browser.

Related projects

  • TypeSafe Playground

    Next.js community playground extending the 110-example TypeSafe library with A/B comparisons, routing, extraction, games and simulations.

    ServicebetaTypeScript
  • jev-broadcast-lab

    Public Jev lab centred on a chess match broadcast where Jev picks from legal moves, plus smaller judgment booths and a Fast Flag catalogue.

    ProjectprototypeJavaScript, Python
  • jev-buzzword-rush

    25-second Japanese IT-buzzword game that shows Jev's speed and cost: every typed term is classified live, with no LLM-generated text.

    ProjectprototypePython, JavaScript
  • jevstudio

    Visual workbench (in Portuguese) for wiring Jev decision funnels as node graphs and watching them play a game in real time.

    ProjectprototypePython
  • keeping-an-npc-on-quest

    Educational code where a local LLM voices a game NPC and Jev classifies player messages and checks every line for spoilers.

    ProjectprototypePython
  • jev-behavior-study

    Independent field guide to Jev 1.13.0 behaviour across 11,621 requests: framing sensitivity, Snake and 3D city driving experiments.

    ProjectresearchPython, TypeScript