{"id":"3893b380-f72d-4a0e-9f7d-e27d17b8b8f0","slug":"learning-loop-coach","kind":"skill","name":"learning-loop-coach","summary":"Evidence-based AI learning coach | 循证学习教练：学习计划、完整课件、对话教学、阶段测验、进度打卡与错题复习。MIT-0.","owner":"HoLauv","repoUrl":"https://github.com/HoLauv/learning-loop-coach","license":"MIT-0","licenseTier":"permissive","trust":"community","stars":1,"match":null,"fewStars":true,"format":"skill","createdAt":"2026-09-11T15:11:55.000Z","addedAt":"2026-10-21T11:35:03.706Z","logoUrl":"https://avatars.githubusercontent.com/u/49048240?v=4","linkOnly":false,"collection":null,"skillPath":null,"categories":["files-documents"],"tags":["active-recall","education","exam-preparation","learning-coach","learning-science","markdown","progress-tracking","self-directed-learning"],"runtime":"binary","local":true,"cloudCapable":false,"cloudCapability":"unknown","cloudReadyOnPlatform":false,"needsApiKey":false,"reviewed":false,"syncedAt":"2026-10-22T10:37:02.941Z","sample":false,"url":"https://conduits.dev/tools/learning-loop-coach","language":"Python","transports":["stdio"],"install":{},"requirements":{"env":[],"upstreamMayCharge":false},"tools":null,"readme":"# learning-loop-coach\n\n<img src=\"assets/logo.png\" alt=\"Learning Loop Coach: book, growth and feedback loop\" width=\"220\">\n\n一个可复用的循证学习教练skill，用于持续学习知识、专业技能和备考。技能名称与显示名称均为 `learning-loop-coach`。\n\n## 能做什么\n\n- 明确目标与基础，按先修关系制定可调整的学习计划。\n- 生成完整Markdown课件，通过对话、示例、反例和练习教学。\n- 按已学范围生成原创阶段题集，分开试卷、解析与答题记录。\n- 准确批改部分交卷，区分课程完成、打卡和能力证据。\n- 记录错题、主动收藏题、延迟复习和后续迁移表现。\n\n核心方法为检索练习、分散复习、示例到独立练习、反馈及迁移。研究依据和边界见[学习科学](references/learning-science.md)。不承诺保证通过考试或快速掌握任何技能。\n\n## 安装\n\n仓库根目录就是skill目录，`SKILL.md` 直接位于根目录，没有额外的 `skills/learning-loop-coach/` 包装层。克隆 https://github.com/HoLauv/learning-loop-coach 到目标环境支持的skills目录，或下载源码后将文件夹命名为 `learning-loop-coach`。保留 `agents/`、`assets/`、`references/`、`scripts/` 和 `tests/` 子目录。同名已存在时先备份/比较，不覆盖。安装后重新加载技能列表。\n\n## 使用示例\n\n```text\n使用 $learning-loop-coach 帮我系统学习Python数据处理。\n我了解变量和循环，每天45分钟，希望6周后独立完成CSV清洗项目。\n用Markdown课件，每5课检查一次，工具操作计入实践评价。\n```\n\n后续可以直接说“继续下一课”“完善这课的教材”“按已学范围出30题”“更新打卡”“把这些错题和收藏题加入复习集”。语言、时长、题量与工具实践是否计分都可配置。\n\n## 可执行工具\n\n主skill不依赖Python，也不依赖第三方study-planner。可选的单选题批改脚本使用Python 3标准库，仅读key并输出JSON，不访问网络或写学习文件。使用格式见[考核参考](references/assessment.md)。\n\n```text\npython -m unittest discover -s tests -v\n```\n\n## 数据与授权\n\n实际学习数据保存在用户工作区，不保存在skill安装目录或此公开仓库。定时提醒需用户明确请求和当前环境可用的调度工具；仅安装skill不会自动建立提醒。未提供文件工具时不会假称已保存，未执行测试时不会假称已验证能力。\n\n本项目为新写的工作流、模板与脚本，采用 [MIT-0](LICENSE) 许可，允许使用、修改及商业再分发，无需署名；软件按原样提供，不作担保。不分发用户成绩、私人教材或第三方skill源码；引用资料的权利仍归各自权利人。\n"}