Havuz

JEV-guided context compaction for OpenCode. Prune stale tool history instead of rewriting conversations with an LLM.

An unofficial, provider-neutral Unix client for Jev from TypeSafe AI. Written in Go.

Claude Code/Codex skills, evals, harnesses/context tools, and learning resources that make it easy to build with Jev, TypeSafe's System One

Jev judgment, not generation: prune long tool output, screen fetched pages for injected instructions, and gate completion claims inside Deep

Shepherd is an end-to-end AI Software Engineering platform that supervises the complete software delivery lifecycle—not just code generation

Runtime constraints for the pi coding agent: checks every side-effecting tool call against what you said, before it runs. Powered by TypeSaf

High-throughput, robust native DuckDB extension for batched and streaming TypeSafe/Jev classification, scoring, and semantic predicates from

A decision model that returns calibrated answers to yes/no, choice, score and multi-label questions.

Chrome extension that cleans up your LinkedIn feed: hides engagement bait, self-promo and off-topic posts using Jev, TypeSafe AI's typed cla

Agent skills for designing, training, evaluating and improving application-specific decision systems. Primitive/model selection, data assemb

Let Jev steer Claude Code: the right reasoning effort, subagent model and skill for every prompt. A Claude Code plugin powered by TypeSafe's

Per-turn model routing for the Pi coding agent: a small judge picks the cheap or the strong tier for each message, with multi-model failover

What can Jev actually do? Reproducible experiments and research reports exploring its capabilities and limits.

Community-maintained JEV runner integration for Midscene Test

A 4B decision model built for Korean.

One grounded Jev/Playwright core: typed SDK, persistent CLI, and MCP server with native browser operations and deterministic assertions.

Don't trust an Agent. Test it. Open-source lab for agent credit — exams → evidence → explainable, recomputable scores. Built for the day age

x402 risk-check provider: signed pre-payment verdicts — OFAC, phishing feeds, transaction simulation, drainer-kit code, and our own EIP-7702

Jev? Déjà vu. Browser agents that run on instinct, no Jev needed. One look at the page, one call to any open model, one action. Faster than

High-performance, high-precision, locally-fast-deployed Jev-class medical decision service on national medical exams — four readings, one ro

Research design for a Jev-native agent system:enable more options than jev provided with virtulization and paging, tool integration, externa

Jev Codex Router, Codex için her model çağrısında model kapasitesini, thinking depth ve route lease seçimini otomatik yapan bir yönlendirme katmanıdır.

Neden yükseliyor · Codex kullanımında model seçimi ve kota verimliliği gibi güncel bir ihtiyaca odaklanıyor; 11 günde 274 yıldız ve son 28 günde 69 commit dikkat çekiyor. Ancak iki farklı yazar olması ve yıldızların kısa sürede birikmesi, ilginin kalıcı benimsemeye dönüştüğünü tek başına kanıtlamıyor.

Jev-compatible API server for open decision models. Ollama for decision models: pull and serve 29 model families (Laya, pplx-decider, Clef,

AgentJev, metin üretmek yerine verilen state ve yapılandırılmış sorular için tek bir forward pass ile olasılık dağılımları döndüren küçük bir decision modelidir.

Neden yükseliyor · 13 günde 338 yıldıza ulaşması dikkat çekiyor; agent'lar için token decode etmeden karar dağılımı üretme yaklaşımı ilgi çekici olabilir. Ancak son 28 gün commit ve farklı yazar verisi ölçülmediğinden aktiviteyi ve ilginin nedenini doğrulamak mümkün değil.

Local gateway that routes coding-agent tool decisions through Jev and can attach a read-only advisor (GPT or Claude).