Skip to content
JevDirectory.org
CommunityTools & Integrations3 starsVerified 2026-09-22

jev-skill-gate

Claude Code hook that scores installed skills against your prompt with Jev and writes skillOverrides so irrelevant skills stay out of context; a local TF-IDF ranker runs when no API key is set.

Open github.comShivamPansuriya/jev-skill-gate
Category
Tools & Integrations
Published by
Community
Author
ShivamPansuriya
Added
2026-09-22
Tagscommunityjavascriptclaude-codeskillscli

Highlights

  • Scores every installed skill in one parallel Jev pass and writes skillOverrides so irrelevant descriptions stay out of context.
  • A committed eval covers 20 labeled cases over 217 installed skills; on 4 live Jev cases all 19 expected skills survived gating.
  • Measured pairwise AUC was 1.000 on every Jev case, with irrelevant skills at ranks 149-217 and scores of 0.01-0.03.
  • The eval moved the nameOnly threshold from 0.25 to 0.15 after relevant skills scored as low as 0.15.
  • Without an API key a local TF-IDF ranker runs instead and produces ranks rather than probabilities, so the planner uses a top-N slice.

Quickstart

bash
node bin/jev-skill-gate.mjs install
node eval/run-eval.mjs --local --fresh

Watch out

MIT-licensed. Needs Node.js and a TypeSafe API key for the Jev arm; it is verified against Claude Code v2.1.274, and the fallback ranker returns ranks, not probabilities.

More like this

jev-use
17GitHub stars
Claude Code, Codex, and pi plugin plus MCP server and npm library that hands steps needing no text output to Jev: one batched call answers typed questions about a state, and an opt-in PreToolUse gate can only deny or ask.
Tools & Integrations#community#javascript#claude-code
Communityjev-use
Hermes Jev Skills
425GitHub stars
A Python toolkit that installs nine plain SKILL.md files for Hermes, Claude Code, and Codex, letting Jev route models, filter retrieved passages, select skills, and choose bounded computer or browser actions.
Tools & Integrations#community#python#agents
Communityhermes-jev-skills
Jev Browser Use
354GitHub stars
A Codex skill (with a plugin option) that hands clicks, toggles, navigation, and scrolling to Jev over accessibility text while Codex types, interprets visuals, and verifies the outcome.
Tools & Integrations#community#javascript#skills
Communityjev-browser-use
Back to all resources

From the community

Posts from builders shipping with Jev right now.

Follow @typesafeai

Navigating Neo4j with Jev

思维怪怪
思维怪怪
@0xLogicrw
·

Jev 这个 waitlist 还是很给力的，昨天申请，今天就能用上。 给已经拿到 API、但还不知道怎么玩的人整理了一份 Awesome Jev，目前我能确认到的 Jev 项目基本都在这里： 1. jev-ultrafast Browser Use 做的高速浏览器 Agent。Jev Show more

Image
思维怪怪
思维怪怪
@0xLogicrw

前 OpenAI 研究员 Diogo Almeida 创办的 TypeSafe AI 推出新模型 Jev。它有点像一个能读懂自然语言的超级分类器，不生成文本，只返回选项、分数和概率，专门给软件做判断。 普通大模型需要一个 token 一个 token 往外生成，Jev 则可以并行给出多个结果。TypeSafe 还用新的 RLCD

Reply
Read 35 replies

Reranking 33,047 catalog entries

Jason Zhu
Jason Zhu
@GoSailGlobal
·

拿 Jev 做搜索重排，我先泼一盆冷水：单独用，它没打赢向量检索 TypeSafe 的 Jev 这阵子很火，一堆项目拿它做重排。我们在 Agent Skills Hub 的 33,047 条目录上认真测了一次，164 条中英文真实查询，9,831 对分级标注，整套只花了 2.6 美元 三个结论 01｜单独重排，约等于没赢 Jev 重排 bge-m3 Show more

Watch on X
Jason Zhu
Jason Zhu
@GoSailGlobal

有美团、阿里的老哥嘛？ 试试加一路召回、重排（离线、近实时实现），我觉得有奇效 他在文本理解上 跟之前机器学习、llm很不一样 还能自动打标签做特征

Reply
Read 6 replies

Six uses that stuck after 60 days

Security decisions that fit Jev

Kostas
Kostas
@Kostastsale
·

This made me rethink where AI actually fits into security engineering. For purely engineering work, forget about ChatGPT or Claude. TypeSafe AI just released Jev, and I think it’s going to change how we build AI into security workflows. Instead of asking an LLM to “investigate Show more

TypeSafe AI
TypeSafe AI
@typesafeai

we are officially out of stealth! join the frontier and get access to Jev on our website (link on profile)

Reply
Read 14 replies

A million judged questions

Inferring Jev's internals from 1,000 calls

iwashi / Yoshimasa Iwase
iwashi / Yoshimasa Iwase
@iwashi86
·

Jevの内部アーキテクチャを推測している技術記事(Jev’s Architecture Unmasked)からメモ。 ・本記事はJevのAPIを約1万回の呼び出して、内部構造を推測したもの ・従来の言語モデルを用いた分類やルーティングでは、トークンを1文字ずつ逐次生成するために膨大な無駄な計算コストが発生していた。 Show more

Reply
Read 2 replies