Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 5 additions & 0 deletions CHANGELOG.md
Original file line number Diff line number Diff line change
Expand Up @@ -22,6 +22,11 @@

## [Unreleased]

### 2026-07-23 — `docs` — 优化 README 首屏定位与 GitHub About
- 重写中英文 README 首屏介绍,统一用规范学习闭环表述:预习、学习与过程练习、错题与错因修复、知识迁移验证、间隔复习、考前计划、期末试卷生成;去掉首屏的 FSRS / HTML-PDF 等技术名词,突出“证据驱动学习闭环”与“动态下一步决策”。保留 V1/V2/V3 描述与核心标语「Explanation is not evidence.」。
- 补充 GitHub 仓库 About:写入描述、留空 Website、新增 9 个 topics(ai-agent、claude-code、study-assistant、learning-science、spaced-repetition、fsrs、knowledge-tracing、education、python)。
- 涉及:`README.md`、`README_EN.md`、`CHANGELOG.md`,以及 GitHub 仓库元信息(描述 / topics)。

### 2026-07-23 — `docs` — 新增 V3 项目文档与 Agent 引导版本说明
- 在中英文 README 的版本能力列表中保留 V1/V2,并新增 V3:README 视觉与导航、双语文档、Agent 新手引导协议,以及贡献、安全、Issue、PR 和 CI 入口。
- 涉及:`README.md`、`README_EN.md`、`CHANGELOG.md`。
Expand Down
6 changes: 4 additions & 2 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -20,9 +20,11 @@

---

`study-loop` 是一个面向大学课程的本地优先、长期有状态的持续学习 Agent(Claude Code Skill)。它记录的不只是“答对没有”,还记录为什么会错、在什么条件下会错、能否迁移、依赖多少提示、多久会忘,以及下一步最值得学什么
`study-loop` 是一个面向大学课程的本地优先、证据驱动的长期学习 Agent(Claude Code Skill)。它不是聊天式问答助手,而是把**预习、学习与过程练习、错题与错因修复、知识迁移验证、间隔复习、考前计划、期末试卷生成**串成一个持续学习闭环。每一次答题、提示、错误和复习都会沉淀为学习证据,系统据此判断知识状态,并动态安排下一步最值得完成的任务

> **Explanation is not evidence.** 听懂不是掌握证据,独立完成才是。
它不把“听懂解释”当作“掌握”:解释、独立作答、错因修复和迁移题验证会被分别记录。

> **Explanation is not evidence.** 听懂不是掌握,独立完成才是。

## 快速开始

Expand Down
4 changes: 3 additions & 1 deletion README_EN.md
Original file line number Diff line number Diff line change
Expand Up @@ -20,7 +20,9 @@

---

`study-loop` is a local-first, stateful learning agent for university courses. It tracks more than whether an answer was right: it records why you made a mistake, when the mistake appears, whether you can transfer the skill, how much help you needed, when you may forget it, and what to learn next.
`study-loop` is a local-first, evidence-driven long-term learning agent for university courses (Claude Code Skill). Rather than a chat-style Q&A assistant, it chains **pre-study, study and practice, mistake and misconception repair, transfer validation, spaced review, exam-prep planning, and final-exam paper generation** into one continuous learning loop. Every answer, hint, mistake, and review becomes learning evidence; the agent uses it to judge knowledge state and dynamically schedule the next most valuable task.

It does not treat “understanding an explanation” as mastery: explanation, independent work, misconception repair, and transfer validation are recorded separately.

> **Explanation is not evidence.** Understanding a solution is not the same as proving mastery.

Expand Down
Loading