Agent 工具

jev-mode

@ddfeyes2PythonMIT更新于 2026-09-18

jev-mode 将简单决策转为类型化判断模型,减少 token 使用和输入工作量,提高准确率,支持 Python 无依赖,MIT 许可。

英文原文

I kept watching coding agents burn context on decisions that aren't hard - triage 400 tickets, tag 600 files, route to one of six teams. jev-mode moves those verdicts to a typed-judgment model. I A/B'd it: 78% fewer tokens, 16x less work-attributable input, accuracy 96.1% vs 93.7%. Python, no deps, MIT.

ddfeyes/jev-mode

它在哪儿调用了 Jev

DEFAULT_ENDPOINT = "https://api.typesafe.ai/v1/systemone"

src/jev_mode/client.py:24

链接指向我们抓取当天的那个 commit,行号是准的。

这个项目没法在这儿跑

它的 question 组合是运行时拼出来的,或者代码里没有直接写出来,所以没法原样搬过来。源码链接在上面,可以自己去看。

同类的其他项目