agentgateway: Jev and OpenAI guardrail
Project / 项目 · In the catalogue / 在目录中查看
What Jev decides here · Jev 在这里做什么决定
noul: read in the code by a model, not yet by a person 由模型在代码里读到，尚无人复核
a model read the call site; no person yet 模型读过调用点，尚无人复核 · · Cited file / 引用的文件: examples/llm-guardrail-jev-openai/guardrail.ts
Summary · 摘要
Three yes-or-no questions (jailbreak, harmful content, secret disclosure) on each request and each response, rejected with HTTP 403 at a probability of 0.5 or more and blocked on any evaluation error; the gateway sends half the evaluations to Jev as Nouls and half to OpenAI.written by a model
对每个请求和每条回复提出三个是非问题（越狱、有害内容、泄露机密），概率达到 0.5 即以 HTTP 403 拒绝，评估出错时同样拦截；网关把一半评估以 Noul 形式发给 Jev，另一半发给 OpenAI。模型撰写机翻
Notes · 备注
Failing closed is the right default for a gate and is easy to get wrong. Note the contrast with the trading examples in this catalog, which deliberately fail open.
闸门失败时默认关闭是正确的默认值，也很容易做错。注意与本目录里那些交易类例子的对比 —— 它们刻意选择失败时开放。
Launches · 发布记录
Not launched yet. 还没有发布过。 Launch it / 发布它 ↗ · How a product launches / 怎样发布一个产品
Makers · Maker
Made this? Confirm or correct its entry / 这是你做的？确认或更正条目 ↗
The same form takes the product's one line and a demo link, in its maker's own words. 同一个表单也可以补上产品的一句话和演示链接，用 Maker 自己的话。
Badge for its README · README 徽章
[](https://kydlikebtc.github.io/awesome-jev/products/agentgateway-guardrail.html)One line for the product's README, linked to this page. The picture says that it is listed; while it is on the launch board, that it is launching; after a week in the top three, its place. It is rebuilt with the site, so the line never needs changing. 放进产品 README 的一行，链接到本页。图片显示已收录；在发布榜上时显示正在发布；进过某周前三后显示名次。它随网站一起重建，这一行不用再改。
Also under this pattern · 同一模式的其他产品
sub2api: Jev as a moderation endpoint
Drops in as a moderation API by asking many parallel Noul questions in one request, one per hazard category, with an anti-injection prefix on every instruction.
claude-code-templates: three Jev plugins
Three independently installable Claude Code plugins — guardrails, model router and skill suggestion — each with its own hooks and tests.
@langchain/typesafe
The JavaScript counterpart of the LangChain integration, with the same classifier and middleware shapes.
DeepChat: agent tool-permission review
Reviews each tool call on three axes — risk level, whether the user authorised it, and an explicit prompt-injection pressure check.
atomic
Coding-agent runtime in which workflows are versioned TypeScript stages with executable checks, review gates and human approvals. Its AI package lists Jev from models.dev as a System One decision model on five providers, apart from the chat models; the README does not say where a workflow calls it.