Launches on September 23, 2026
Repositories created on this date, ranked by GitHub stars. They launched the same day, so their stars compare directly.
- 1011
Browser-use agent that finds flights with Jev (TypeSafe AI). First project with TypeSafe — Jev picks, code acts.
- 1021
Turn an inbox into a short action queue with Jev (TypeSafe System One)
- 1031
TypeScript MCP server that finds TikTok and YouTube micro-KOLs, tags their niche, scores campaign fit with Jev, and drafts outreach emails. Results are cached locally in SQLite.
- 1041
Making TypeSafe's Jev talk. It can't generate text, only pick from options, so it writes replies one word at a time by choosing each next word from a 1,000-word list.
- 1051
MCP server that lets Claude Code, Codex & AI agents drive a real Chrome browser with JEV (TypeSafe System One): ~100 ms typed decisions, calibrated confidence, human-in-the-loop questions, safety gates and a debug trace.
- 1061
Agent Skill that finds where TypeSafe Jev decision models would pay off in a codebase. Works in Claude Code, Codex, Cursor and other Agent Skills hosts.
- 1071
Ask TypeSafe's Jev model how your Singapore TOTO numbers look against recent draws
- 1081
先回|消息优先级助手 Windows + Android双端工具 JEV规则引擎 + AI自动判定消息优先级
- 1091
TypeSafe Jev (typesafe.ai) 反向代理网关:原生 /v1/systemone 透传 + OpenAI 兼容层,含号池与控制台
- 1101
A 2B browser-agent decision model. Speaks the TypeSafe System One protocol, so jev-ultrafast is repointed with one line. Answers do not move with the page title or the form layout. 100% on the action suite.
- 1111
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.
- 1121
Jev code harness
- 1131
Decision model on a Qwen3-0.6B backbone with LoRA: fused decision endpoint (/v1/systemone) + OpenAI-compatible chat completions with thinking, SSE streaming and vLLM-style sampling, running on Apple Silicon MPS.
- 1141
When Jev meets LLM — Transparent proxy that makes AI coding agents 3× faster and 80% cheaper. Cut response times from 2.4s to 890ms with Line J architecture.
- 1151
An Obsidian-compatible knowledge base with reviewable Jev classifications, TypeSafe and Vercel access, and a documented build process.
- 1161
Independent Rust runtime for Kev decision models (pointer heads on Qwen bases): MLX on Apple Silicon and Candle CPU, parity-gated against frozen upstream goldens
- 1171
🎮 Play. Learn. Respawn. Turn LAYA & ModernBERT Decoder into tactical AI pilots in Cinder Station, an original Doom-inspired game. Explore RLCD + CE, TypeSafe Jev, interactive lessons & transparent benchmarks. Train locally or in Colab. Fork it. Train it. Play it.
- 1181
Ask a yes or no question. Let Jev, or a fair coin, decide.
- 1191
Turn plain-English engineering rules into pull-request checks
- 1201
Stop AI agents from executing MCP tools unchecked — policy, approval, JEV decisions, and one-use execution permits.
- 1211
极简现代 macOS 暗黑胶囊输入法:集成小狼毫 (Weasel) · 万象拼音海量词库与 400MB 语法模型 · TypeSafe Jev AI / Ollama · 图形化交互设置控制中心 (WebUI)
- 1221
A tic-tac-toe app where one player is Jev, TypeSafe AI's System One Model with live confidence scores and probabilities.
- 1231
a work in progress, built from the typesafe-ai-skill for building and analyzing jev opportunities
- 1241
Fast, calibrated System One decision platform powered by TypeSafe Jev via OpenRouter. Sub-second logprob scoring, transfer curves, zero hallucinations.
- 1251
A zot extension that wires Jev (TypeSafe System One) into the agent loop.
- 1260
GPT-6 Sol / Codex로 개발하고 Jev AI의 탄막 회피 능력을 실험한 Unity 2D 탄막 프로젝트
- 1270
Open verification layer for AI agents. The burden of proof is on the evidence.
- 1280
Adaptive RAG Router using TypeSafe Jev, with budget constrained evaluation against rule based and LLM routers.
- 1290
ADK Go v2 demo: TypeSafe Jev + Gemini combined triage pipeline vs an all-Gemini pipeline, side by side
- 1300
Give your agents typed AI decisions instead of parsing LLM text. Jev (TypeSafe) classifier with confidence-gated routing + regex baseline benchmark.
- 1310
Measures source code quality with TypeSafe's Jev as the judge — an AII OS WASM plugin (id.aiii.codequality)
- 1320
Official open-source companion repository for Analytics Made Simple (analyticsmadesimple.com) tutorials, SQL queries, and scripts.
- 1330
AstrBot 主动回复门控插件:用 TypeSafe Jev 模型给群聊消息打分,只在真值得接话时才触发一次主动回复
- 1340
Score any job posting against your resume, using your own Jev API key. No server, no account. The open-source build of autosham for Jobs.
- 1350
A benchmark harness for bias in fast decision models: causal counterfactuals on real text, with control floors, reported in the decision's own currency
- 1360
A Camunda connector to bring TypeSafe's Jev model into your Camunda 8 processes.
- 1370
Claude Code skill: Jev (TypeSafe via OpenRouter) picks the model tier and sorts piles; Claude writes. With a measured token comparison.
- 1380
macOS clipboard history and context-aware Smart Paste powered by TypeSafe Jev.
- 1390
Jev-powered selective context handoffs for Codex, with a GitHub Copilot portability plan.
- 1400
Model, reasoning effort, skill, and bounded worker routing for Codex with optional TypeSafe Jev.
- 1410
Local Codex task triage with human-reviewed archiving and optional Jev analysis
- 1420
Command-line reviewer for technical writing: reader fit, AI-writing signs, and audience literacy, judged by TypeSafe Jev
- 1430
/compact-jev: Jev-guided verbatim compaction for Claude Code through Vercel AI Gateway, separate from /compact.
- 1440
Kişisel kripto & hisse yatırım araştırma asistanı: backtest, walk-forward, istatistiksel anlamlılık, paper trading, TypeSafe Jev entegrasyonu. Yatırım tavsiyesi değildir.
- 1450
Checking whether a model's probability means what it says: TypeSafe Jev and open-weights Laya against 150,000 human ratings, with stdlib tools to run the same check on your own data.
- 1460
Anonymous, share-card-first dating-chat evaluator: upload a screenshot, ask Jev for the read, and get the tea. Built on Next.js, Neon (Lakebase Postgres + Object Storage), and Vercel.
- 1470
Security & calibration middleware for System 1 AI models like jev & laya
- 1480
Decision Lab: visual System One playground for building decisions, comparing Jev, OpenJEV, Laya, and exploring what-if changes.
- 1490
project
- 1500
Doc, Scott's sidekick: Jev (TypeSafe) wired into Claude Code — checks comments before they're published, keeps effort estimates out of replies, points each turn at the right skill, and routes easy work to a cheaper model.