目录包含 1,334 条仓库或集成记录、61 条文章与讨论、33 条其他资源。仓库链接按地址去重后对应 1,332 个 GitHub 仓库；仅凭公开链接不能判定其许可证为开源。
精选 10 个关注度较高的项目，覆盖浏览器自动化、上下文压缩、交易、官方技能、电脑操作、代码审查、Agent 管理与数据查询。
可以按项目名称、作者、用途或标签搜索，也可以按类别和活跃度筛选。点击项目详情可查看 Jev 的具体作用和源码链接。
Anthony Maio's essay on what a model that cannot generate text is for.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
Anthony Maio's essay on what a model that cannot generate text is for.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
Hardel-DW
Minecraft mod where Jev tries to finish the game from scratch without a scripted route.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
Minecraft mod where Jev tries to finish the game from scratch without a scripted route.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
Wing9897
Laya 是選用本機模型，權重不在這個 repo。需要時再裝，第一次判斷才會下載：
Jev 对每批 Telegram 消息逐条判断是否符合用户自然语言条件，并对整批进行类型分类。
把选择和打分接进现有程序；暂无可核验的性能对照。
xinyao27
Local OpenAI-compatible proxy that sits between a coding agent and its providers and asks Jev, per turn, which routed model and thinking level should serve it.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
Local OpenAI-compatible proxy that sits between a coding agent and its providers and asks Jev, per turn, which routed model and thinking level should serve it.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
ndolinschi
TypeSafe Jev playground — custom Choice/Score/Noul builder with live distributions
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
TypeSafe Jev playground — custom Choice/Score/Noul builder with live distributions
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
CorieW
Bounded exploratory browser testing with Jev, deterministic assertions, and replayable evidence.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
Bounded exploratory browser testing with Jev, deterministic assertions, and replayable evidence.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
ishantanu
OpenTelemetry Collector processor that uses Jev to assess span operation metadata and annotate spans with diagnostic value, business criticality, and retention probabilities without dropping traces.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
OpenTelemetry Collector processor that uses Jev to assess span operation metadata and annotate spans with diagnostic value, business criticality, and retention probabilities without dropping traces.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
WeSecureYou
Job Risk Analyzer: CLI and REST API that uses Jev to score an occupation's exposure to AI-driven layoffs and its resilience.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
Job Risk Analyzer: CLI and REST API that uses Jev to score an occupation's exposure to AI-driven layoffs and its resilience.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
Jev trades 15-minute and 1-hour BTC, ETH, and SOL markets on Kalshi.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
Jev trades 15-minute and 1-hour BTC, ETH, and SOL markets on Kalshi.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
kunobi-ninja
Rust client for the TypeSafe System One API (Jev)
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
Rust client for the TypeSafe System One API (Jev)
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
ndolinschi
LaneBreak — support ticket priority+routing via TypeSafe Jev
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
LaneBreak — support ticket priority+routing via TypeSafe Jev
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
0x963D
A cyberpunk border encounter powered by TypeSafe Jev. Bluff the guard. Inspect the receipts.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
A cyberpunk border encounter powered by TypeSafe Jev. Bluff the guard. Inspect the receipts.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
TypeSafe's founder on why RLCD-trained decision models are a shorter path to value than chat models.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
TypeSafe's founder on why RLCD-trained decision models are a shorter path to value than chat models.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
mikecann
A 3D Magic 8 Ball you shake and let go, where Jev answers your question as a Choice over the 20 classic answers and the page shows its probability for each one, running on Convex through the Convex AI Gateway.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
A 3D Magic 8 Ball you shake and let go, where Jev answers your question as a Choice over the 20 classic answers and the page shows its probability for each one, running on Convex through the Convex AI Gateway.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
社区项目
The case for machine-native intelligence built for software, not conversation.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
The case for machine-native intelligence built for software, not conversation.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
ndolinschi
Match user goals to MCP catalog (two-stage) via TypeSafe Jev
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
Match user goals to MCP catalog (two-stage) via TypeSafe Jev
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
jxucoder
Rewrite AI drafts in your own voice with a bounded TypeSafe feedback loop.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
Rewrite AI drafts in your own voice with a bounded TypeSafe feedback loop.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
Every's Mike Taylor runs his whole archive through Jev.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
Every's Mike Taylor runs his whole archive through Jev.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
myokoym
Ongoing Japanese research deck on Jev and System One models, maintained as Markdown slides.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
Ongoing Japanese research deck on Jev and System One models, maintained as Markdown slides.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
Ephraim Duncan's demo where Jev decides which model should serve a request.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
Ephraim Duncan's demo where Jev decides which model should serve a request.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
社区项目
Task plus subscription list in, Jev picks which model or agent should handle it.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
Task plus subscription list in, Jev picks which model or agent should handle it.
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
DomMonte
n8n community node for the TypeSafe AI System One API — typed yes/no, choice and score questions with calibrated probabilities
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
n8n community node for the TypeSafe AI System One API — typed yes/no, choice and score questions with calibrated probabilities
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
STRML
Jev-powered model-judged permission gate for OMP (TypeSafe System One)
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
Jev-powered model-judged permission gate for OMP (TypeSafe System One)
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
One direct Jev question per row against 12–14 Jev-scored dimensions with locally fitted weights on three classification tasks: 5,477 test rows, 25,174 Jev calls, $1.43. Decomposition wins on Japanese NLI (0.9076 vs 0.8373) but flags about 25× more hard benign rows as attacks (37.2% vs 1.5%).
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。
打开项目链接查看原始资料与实现。
One direct Jev question per row against 12–14 Jev-scored dimensions with locally fitted weights on three classification tasks: 5,477 test rows, 25,174 Jev calls, $1.43. Decomposition wins on Japanese NLI (0.9076 vs 0.8373) but flags about 25× more hard benign rows as attacks (37.2% vs 1.5%).
当前收录包含项目简介与链接，尚无可核对的 Jev 决策点。