Jev Graph Search
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: src/jev.js
Summary · 摘要
Finds notes and passages in a local Obsidian vault or Logseq folder for an AI agent: it builds a shortlist on your machine, then asks Jev a relevance noul per candidate and ranks by it, returning the original passage with its source. Comes as a CLI and an agent skill.written by a model
为 AI 智能体在本地 Obsidian 库或 Logseq 文件夹中查找笔记和段落：先在本机生成候选短名单，再就每个候选向 Jev 提一个相关性是非题并据此排序，返回原文段落及其出处。提供命令行工具和智能体技能两种形式。模型撰写机翻
Notes · 备注
On npm. The author reports Jev reranking raised top-five source recall from 63.9% to 81.8% on one 8,851-node tax-code graph. An offline mode ranks lexically without Jev. Row drafted from the discovery queue (issue #14) by an AI assistant that read the README and the cited file on 2026-11-03; a maintainer has not yet read the call site.
已发布在 npm 上。作者报告，在一个 8,851 节点的税法图谱上，Jev 重排序把前五条的来源召回率从 63.9% 提高到 81.8%。离线模式不调用 Jev，仅按词汇匹配排序。本条目由 AI 助手根据发现队列（issue #14）起草，于 2026-11-03 阅读了 README 和所引文件；维护者尚未亲自阅读调用点。
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/jev-graph-search.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
AutoGPT TypeSafe blocks
Seven production blocks — choice, score, yes/no, ask-many, route, pick-best, filter — with a UTF-8 byte budget, verbatim wire capture and eleven test files.
OpenViking: retrieval reranking
One Noul per candidate document in a single batched request, with the yes-probability used directly as the relevance score.
oh-my-pi
⌥ Coding agent with the IDE wired in
FastMCP jev_search transform
Two-stage MCP tool search: a wide Choice coarse-ranks the whole catalogue, then a shortlist gets full descriptions plus one Noul each to decide whether it does the job at all.
jcode: memory recall without embeddings
Replaces the whole retrieval stack for memory recall — no embeddings, no BM25, no reranker — with one batched Noul per candidate memory.