Agent 工具
BeLazy167/typesafe-mod 是一个 Claude Code mod，将决策路由到 TypeSafe 的 Jev 模型，根据 prompt 排序已安装技能，并在有信心时回答 agent 的选择题。
Claude Code mod that routes decisions to TypeSafe's Jev model: ranks installed skills per prompt, and answers the agent's own this-or-that questions when confident.
endpoint: 'https://api.typesafe.ai/v1/systemone',链接指向我们抓取当天的那个 commit，行号是准的。
下面是从这个项目源码里原样取出来的 question 组合。
needs_skill是/否Does answering `state.user_request` call for following a written procedure,
import { experimental_evaluate as evaluate } from 'ai';
const { answers } = await evaluate({
model: 'typesafe-ai/jev',
state,
questions: {
needs_skill: { type: 'boolean', instructions: 'Does answering `state.user_request` call for following a written procedure,' },
},
});