ContextGem
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
Structured extraction (结构化抽取), Classification (分类)
noul, choice, score: 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: contextgem/internal/base/decision_models.py
Summary · 摘要
Python framework that extracts structured data from documents with LLMs and decision models. Given a decision model such as Jev, it asks a yes/no question per paragraph to find each aspect's clauses and turns Boolean, label and rating concepts into noul, choice and score questions.written by a model
用 LLM 和决策模型从文档中抽取结构化数据的 Python 框架。配合 Jev 这类决策模型时，它对每个段落提一个是非题，找出每个方面（aspect）对应的条款，并把布尔、标签和评级概念转成 noul、choice 和 score 问题。模型撰写机翻
Notes · 备注
Requests go through LiteLLM's Decisions API (litellm.adecisions) in OpenAI's format, where a yes/no question is a predicate; the README's example names typesafe/jev-latest. The same code also reaches OpenAI's Decisions API and Microsoft Foundry, which are not Jev. Open questions still go to an LLM. Row drafted from the discovery queue (issue #14) by an AI assistant that read the README and the cited file on 2026-11-06; a maintainer has not yet read the call site.
请求通过 LiteLLM 的 Decisions API（litellm.adecisions）以 OpenAI 格式发出，其中是非题称为 predicate；README 示例使用 typesafe/jev-latest。同一代码也可调用 OpenAI 的 Decisions API 和 Microsoft Foundry，它们不是 Jev。开放式问题仍交给 LLM。本条目由 AI 助手根据发现队列（issue #14）起草，于 2026-11-06 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/contextgem.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
jev4pg
Natural-language SQL and 41 semantic operators for PostgreSQL. Filtering, extraction, ranking, matching and verification go to Jev in bounded batches with validated answers; a choice option needs 0.55 and a noul 0.8 to count. The authors report 98% accuracy against 92% for pg-jev's request path.
jev-design
Experiment in having Jev design a dashboard: from a one-sentence brief like "an 18th century printed book", two staged requests of choice, score and yes/no questions pick every visual token (colour, type, radius, icons) from a catalog a person wrote, and an engine applies them to a Next.js console.
smart-paste
Fills form fields from pasted text: the form's heading, labels and your text go to TypeSafe, and it inserts the values it matches for you to review before submitting.
jeveryword
Text extraction with Jev: field extraction, PII detection and exact quotes, built on TypeSafe's Jev.
jev-macos-loop
Computer-use agent for native macOS apps on Apple silicon: OmniParser CoreML, Apple Vision OCR and accessibility labels find a window's controls locally, and Jev picks the next action with a choice from the observed text, never pixels. Input is guarded, and benchmarks check results, not Jev's DONE.