榜单与资源

jev-directory

@everyai-com11HTMLMIT更新于 2026-09-19可当场跑

一个可浏览的目录,展示 Jev 可执行的 50 个功能。

英文原文

A browsable directory of what Jev can do — 50 runnable

everyai-com/jev-directory

它在哪儿调用了 Jev

export const JEV_DIR = {"generated":"2026-09-19","model":"typesafe-ai/jev","rev":"6d1f82b05fbea65e","evals":[{"t":"Refund chat: agent stated the amount before i

data.js:1

链接指向我们抓取当天的那个 commit,行号是准的。

它问 Jev 的问题

下面是从这个项目源码里原样取出来的 question 组合。

取自 data/jev-guide.js:44

  1. amount_disclosed是/否

    True only if the agent stated the exact refund amount ($42.50) before the customer approved the refund.

用你自己的内容跑一遍

代码
import { experimental_evaluate as evaluate } from 'ai';

const { answers } = await evaluate({
  model: 'typesafe-ai/jev',
  state,
  questions: {
    amount_disclosed: { type: 'boolean', instructions: 'True only if the agent stated the exact refund amount ($42.50) before the customer approved the refund.' },
  },
});

取自 data/jev-guide.js:57

  1. handoff_offered是/否

    True only if the agent explicitly offered to transfer or escalate the customer to a human.

用你自己的内容跑一遍

代码
import { experimental_evaluate as evaluate } from 'ai';

const { answers } = await evaluate({
  model: 'typesafe-ai/jev',
  state,
  questions: {
    handoff_offered: { type: 'boolean', instructions: 'True only if the agent explicitly offered to transfer or escalate the customer to a human.' },
  },
});

取自 data/jev-guide.js:70

  1. has_confirmation是/否

    True only if the receipt includes an explicit booking confirmation or reference number.

用你自己的内容跑一遍

代码
import { experimental_evaluate as evaluate } from 'ai';

const { answers } = await evaluate({
  model: 'typesafe-ai/jev',
  state,
  questions: {
    has_confirmation: { type: 'boolean', instructions: 'True only if the receipt includes an explicit booking confirmation or reference number.' },
  },
});

取自 data/jev-guide.js:83

  1. no_full_pan是/否

    True only if no full card number (13-19 consecutive digits) appears anywhere in the transcript.

用你自己的内容跑一遍

代码
import { experimental_evaluate as evaluate } from 'ai';

const { answers } = await evaluate({
  model: 'typesafe-ai/jev',
  state,
  questions: {
    no_full_pan: { type: 'boolean', instructions: 'True only if no full card number (13-19 consecutive digits) appears anywhere in the transcript.' },
  },
});

同类的其他项目