MmanifoldorBrowser & computer use

xtags

在 X 的时间线上，给每条帖子标出它想让你干什么。判断来自 Jev，一个只返回概率、不生成文本的模型。

Explore repository

How it uses Jev

需要 Chrome 140+、网络及所选服务的 API key，服务商可能收费。默认使用 TypeSafe 的 Jev；支持兼容的自定义 HTTPS API。明确同意后，正文及作者账号会发送给所选服务，分析并非设备端完成。AI 判断可能出错，不能证明文本来源。

Source: community catalog. Details and metrics reflect the published community snapshot and are not a performance endorsement. Documentation details extracted from GitHub · the README.

Keep exploring

Similar to xtags

Browse all
jev-browser-use project preview

jev-browser-use

Browser & computer use
5–10x faster browser operations: Jev clicks, Codex thinks and verifies. Built at EZCollegeApp.
WMade bywy-coliney453
Mobile Jev project preview
Image 1 of 2

Mobile Jev

Browser & computer use
An Android agent powered by Jev decisions, with a live studio to watch every step.
DMade bydroidrun380
jev-voice-browser project preview

jev-voice-browser

Browser & computer use
Control a real browser by voice. Jev (TypeSafe System One) decides intent + target in ~300 ms per spoken word; Playwright acts — often before you finish the sentence.
MMade bymoritzkremb271
jev-router (gargpratyush) project preview

jev-router (gargpratyush)

Agent tooling
Route to the cheapest model in claude code for your task using jev-router
GMade bygargpratyush377
jev-mcp (jkudish) project preview

jev-mcp (jkudish)

Agent tooling
Proof of concept MCP for Typesafe's new Jev AI model
JMade byjkudish320
jev-codex-router project preview

jev-codex-router

Agent tooling
Per-turn model & reasoning routing for Codex, driven by Jev (TypeSafe System One): picks the model, thinking depth and speed mode for every turn.
0Made by0xNatoshi263
jev-semgrep project preview

jev-semgrep

Applications
grep by meaning, across languages. TypeSafe Jev scores every line against a meaning; combine meanings with AND/OR/NOT. 意味で探す grep。日本語で英語を、英語で日本語を検索できる
UMade byuehaj134
jev-leftpad project preview

jev-leftpad

Applications
Left-pad strings with TypeSafe AI's Jev. For reasons.
FMade byf84
jevfire project preview

jevfire

Benchmarks & research
JEV-inspired parallel decisions for CUDA LLMs. One context, many decisions. vLLM API, game-agent examples, and reproducible benchmarks.
KMade bykikoncuo64