dsh-jev-verify
为 DeepSeek Harness 端到端接入 TypeSafe Jev（System One 决策模型）：jev_decision（choice/score/noul 并行判定，类型化答案+置信度）、「设置 → 插件」配置卡片（API Key 与开关免命令行）、对话内联工具视图（每次判定直接显示答案、置信度、延迟与成本）、自动护栏（确定性规则 + Jev 风险/循环检测，失败时回退权限确认）、本地 /jev 仪表盘与在线验证基准（2026-09-21 实测 27 题准确率 96.3%、中位延迟约 283 ms）。诚实设计：只走真实 API，无 Mock 回退。
TypeSafe Jev (System One decision model) for DeepSeek Harness, end to end: jev_decision (choice/score/noul, parallel, typed answers + confidence), a Settings > Plugins card for the API key and switches, inline tool views showing every decision in the conversation (answer, confidence, latency, cost), an auto-guard (deterministic + Jev risk/loop checks), a local /jev dashboard, and an online verification benchmark (96.3% on 27 labeled questions, median ~283 ms on jev-latest, 2026-09-21). Honest by design: real API only, no mock fallback.
怎么安装
dsh plugin add dsh-jev-verify 简介
dsh-jev-verify Jev — TypeSafe AI's **System One** decision model — as a first-class plugin for DeepSeek Harness (dsh). Jev does not generate text. Given a \state\ plus typed questions it returns **typed answers with calibrated probabilities** in one parallel API call (~70–500 ms published). This plugin exposes that as agent tools, adds an opt-in **auto-guard** (risk + loop checks), and makes sure the claims are *verified, not trusted blindly*: | Tool | What it does | | --- | --- | | \jev_decision\ | Choice / Score / Noul questions against a \state\ in ONE call. Returns typed answers, confidenc…
推荐参考
信息
- 协议
- MIT
- 语言
- JavaScript
- GitHub 星标
- 1
- 月下载
- –
- 最近更新
- 2026-09-23
- 创建于
- 2026-09-21
基础安全检查
- 检查结果
- 无
- 收录来源
- curated:awesome-dsh-plugin.com, curated:awesome-dsh-plugin/awesome-dsh-plugin