A free GitHub project directory for the TypeSafe Jev / System One model.Explore official SDKs, agent tools, open-source alternatives and Jev news.

Data updated

  • Hermes plugin: drive and read a visible terminal-browser tab with Jev typed decisions. No screenshots, fractions of a cent per step.

    nedzen/jev-terminal-browser-driver · Python

    10
    browser-automationcdphermes-agenthermes-pluginterminal-browserpython
  • Pre-registered benchmark: can a 2B local model (Gemma 4 E2B) answer web questions without making things up when a decision model (TypeSafe Jev) makes every call? SearXNG for search, MemPalace for verbatim memory, seven arms including open local judges. Spec and thresholds fixed before any run.

    clduab11/jev-test · Python

    10
    benchmarkgemmahallucinationmempalaceragsearxngsmall-language-models
  • Run only the tests your change touched: import graph + ownership rules first, Jev decides the ambiguous rest, fail-safe to the full suite. Replay benchmark with real ground truth.

    criguex/jev-test-impact · TypeScript

    10
    ciplaywrightqa-automationtest-impact-analysistypescriptvitest
  • Jev (TypeSafe System One) 테스트베드 — 클라우드 API와 로컬 셀프호스팅(jeff/GLiFormer) 양쪽 실행 예제 및 실측 결과

    hulryung/jev-testbed · Python

    10
    python
  • A lightweight, auditable Jev testbench built through Agentic Engineering, with frozen evidence, a preregistered null result, and a bilingual technical blog. 基于 Agentic Engineering 的轻量、可审计 Jev 实验台,包含冻结证据、预注册负结果、双语技术博客与认知负债反思。

    Charlie-Wang-03/jev-testbench · Python

    10
    agentic-engineeringai-agentsllmmodel-evaluationpreregistrationprobabilistic-aipython
  • A visual TypeSafe demo where Jev chooses verified Tetris placements.

    MachineLearning-Nerd/jev-tetris · Python

    10
    python
  • A janitor for markdown vaults, powered by TypeSafe Jev. Redacts locally before anything is sent, shows the bill before the run, and meters it against a ceiling. Jev votes, your code files, you review the low-confidence pile.

    kylehovance-ai/jev-the-janitor · Python

    10
    python
  • MCP-first toolkit for TypeSafe/Jev — the System One decision model. One stdio server (jev mcp) serves any MCP-capable harness, backed by one local event log and Prometheus impact metrics you can scrape into your own Grafana.

    jbt95/jev-toolkit · TypeScript

    10
    ai-agentseffecteffect-tseffectsmcp
  • Millisecond-class test-failure triage for coding agents: RETRY / FIX_CODE / FIX_ENV, powered by TypeSafe Jev.

    ccai40359-wq/jev-triage · Python

    10
    ai-agentsclaude-codecoding-agentsdeveloper-toolspytesttest-automation
  • Jev, TypeSafe AI's decision model, triages every message before your OpenClaw agent's main model runs.

    CodeCave0000/jev-triage · JavaScript

    10
    javascript
  • .Net SDK for the TypeSafe Jev API

    mehdihadeli/jev-typesafe-dotnet · C#

    10
    dotnetc#
  • jev-typesafeai: TypeSafe Jev ecosystem repository.

    vishivishvish/jev-typesafeai · TypeScript

    10
    typescript
  • ⚡ Ultrafast browser agent with dynamic, indexed action spaces. Zurich → London on Google Flights in 7.1s.

    authrain-cloud-abdullahformuli/jev-ultrafast · Python

    10
    python
  • A Jev powered hook that detects overloaded multi-topic multi-question messages and forces the agent to present issues one at a time, adding relevant context and missing info.

    lucaminudel/jev-unpack-hook · Python

    10
    agent-harnessagent-hookclaude-codecodexjev-api
  • Jev (TypeSafe AI's System One decision model) use-case catalog: real-world builds, cost math, design patterns, and a reality check on vendor claims.

    vamsikrishna2421/jev-usecases

    10
  • The doorman for Claude’s context window. JEV-powered model routing and recoverable context filtering for Claude Code.

    cvsgireesh/jev-usher · TypeScript

    10
    ai-agentsclaudeclaude-codeclaude-code-plugincontext-engineeringcontext-windowllm
  • Talk to your browser: Claude plans the command, Jev carries it out in your own Chromium profile. A side-panel extension with no chrome.debugger and no remote-debugging port.

    emoubarak/jev-voice · Python

    10
    ai-agentsbrowser-agentbrowser-automationchrome-extensionclaude-codenative-messagingspeech-to-text
  • A realtime voice agent with no LLM in the loop: Deepgram Flux ends the turn, TypeSafe's Jev picks the reply, Deepgram Flux TTS speaks it.

    dg-coreylweathers/jev-voice-agent · TypeScript

    11
    typescript
  • Voice-controlled browser via Jev, native CDP (port 9229), Doubao Voice Bridge and Web Audio streaming

    winter-loo/jev-voice-browser · JavaScript

    10
    javascript
  • jev-voice-browser-zh: TypeSafe Jev ecosystem repository.

    CatJuly/jev-voice-browser-zh · JavaScript

    10
    javascript
  • A single web page for requesting Jev from TypeSafe AI, supporting TypeSafe official api or Vercel AI gateway.

    frankie137/jev-web-ui · HTML

    10
    html
  • Turn any word into probability distributions over taste, material, scent, and shape. Powered by Jev.

    spate141/jev-wordfeel · TypeScript

    10
    typescript
  • Hands-on workshop for Jev, TypeSafe AI's System One model: 5 live-executed notebooks, a 42-slide deck, and a fully cited reference. Decisions, not text.

    fnusatvik07/jev-workshop · Jupyter Notebook

    11
    classificationjupyter-notebooksmachine-learningworkshopjupyter-notebook
  • Play Chinese Chess (Xiangqi) against Jev - TypeSafe System One decision model as the AI. Score fan-out over legal moves.

    Zafer-Liu/jev-xiangqi · HTML

    10
    html
  • Jev (TypeSafe System One) plays Zork I: one Choice per move over Jericho's valid actions, with its confidence on display. French dashboard.

    Resadan-dev/jev-zork · Python

    10
    jerichollmpythontext-adventurezork
  • A .NET client library for the [TypeSafe AI](https://typesafe.ai) System One API (`https://api.typesafe.ai`).

    nothingmn/Jev.Sdk · C#

    11
    c#
  • An unofficial Zig client for the TypeSafe System One API (Jev)

    jakeknowlton/jev.zig · Zig

    10
    sdkzigzig-libraryzig-package
  • Let Jev (TypeSafeAI) solve 2048

    KyleKreuter/jev2048 · TypeScript

    10
    typescript
  • Independent benchmark data for TypeSafe's Jev (System One model) vs LLMs: accuracy, calibration, cost. Boards + per-decision logs, CC-BY-4.0

    Jevals/jevals-data

    12
    benchmarkcalibrationdatasetleaderboardllm-evaluation
  • 1380JevApi
    JevApi — .NET tooling for the TypeSafe AI Jev model, which evaluates typed questions (noul/choice/score) against a state and returns structured answers with probabilities and confidence.

    JawzoD3TH/JevApi · C#

    11
    c#
  • Preregistered, bias-corrected test of TypeSafe Jev's calibration under human disagreement (ChaosNLI, 100 labels per item)

    GautamTalksDev/jevbench · Python

    10
    calibrationchaosnlievaluationpreregistrationpython
  • 1382jevcam
    Tell a Raspberry Pi camera what to look for in plain English: live 20 fps video, TypeSafe Jev detection, radio-DJ alarms, 3D map dashboard and a PPE monitor.

    TheAIMogul/jevcam · HTML

    10
    computer-visionelevenlabsflaskhackathonkioskmaplibreobject-detection
  • JevCode — Jev (TypeSafe System One) 技术解决方案与最佳实践 · https://www.jevcode.ai

    miounet11/jevcode · TypeScript

    10
    astrodecision-modeldocumentationtypescript
  • JevCore Agent — local coding harness (Jev decisions, Guard hard policy, MCP + CLI)

    carter1111/jevcore · JavaScript

    10
    javascript
  • JEV core — the judgement primitive and harness for TypeSafe System One (Jev). Closed-set, fail-open, stdlib-only.

    litshing/jevcore · Python

    10
    ai-agentsharnessjev-harnessjudge-modelllm-judgepythonstructured-output
  • jevditor: TypeSafe Jev ecosystem repository.

    nibzard/jevditor · TypeScript

    10
    typescript
  • Grounded information extraction that cannot hallucinate: code proposes spans, Jev decides. An open-source alternative to LangExtract, with a bilingual benchmark and paper.

    gabazureus/jevextract · Python

    11
    benchmarkgroundinghallucinationinformation-extractionlangextractllmnamed-entity-recognition
  • Typed judgments for Java: declare questions as an interface and let Jev answer them.

    rbrisuda/jevface · Java

    10
    javaspring-aispring-boot
  • Jev picks, you watch. A hybrid movie recommender: fast semantic + keyword search narrows 4,800 films to a shortlist, then TypeSafe Jev reads your constraints and picks the one film that fits - with a confidence score that decides whether to answer instantly or ask a follow-up.

    ArielBubis/Jevflix · Python

    10
    python

1351–1400 of 3057