← Back to Skills Marketplace
17
Downloads
1
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install daily-learning-card
Description
Daily Learning Card
README (SKILL.md)
Daily Learning Card
每日学习卡片。每天10:00扫描前一天对话生成学习卡片，每周日20:00汇总本周学习卡片生成Word文档。支持手动触发。
触发时机
每天10:00扫描前一天对话生成学习卡片（通过HEARTBEAT.md配置）
每周日20:00汇总本周学习卡片生成文档
也支持手动触发：用户说"汇总本周学习"、"生成本周总结"、"记录今天学的"等。
核心功能
每日任务：扫描前一天对话生成学习卡片（10:00）
-
跨会话扫描前一天对话历史
- 使用
sessions_list获取所有会话 - 使用
sessions_history获取每个会话的前一天对话 - 时间范围：今天 00:00 到现在
- 识别学习模式：用户提问 → AI 解答 → 用户确认学会
- 使用
-
提取六类信息（增强版）
- 学习活动：做了什么
- 决策过程：为什么选择A而非B（选项对比）
- 关键数据：数字 + 意义（不只是数值）
- 踩坑记录：问题 + 解决方案 + 教训
- 新概念：术语 + 定义 + 应用场景
- 金句/洞见：一句话总结
-
生成学习卡片（增强版格式）
- 位置：
memory/learning-cards/YYYY-MM-DD.md - 单文件多主题：一天只生成1个文件，包含当天所有学习主题
- 没有学习内容则跳过
- 标注来源渠道：卡片开头注明（飞书/WebUI/Telegram 等）
- 标注可出题点：每个主题末尾标注可用于出题的内容
- 位置：
-
增强版格式要求
### 主题1：[名称] **场景/背景：** [一句话说明] **决策过程：** | 选项 | 考虑因素 | 结果 | |------|---------|------| | A | ... | 选/不选 | **关键数据：** | 数据 | 数值 | 意义 | |------|------|------| | X | 数值 | 说明 | **踩坑记录：** - **问题：** ... - **解决方案：** ... - **教训：** ... **新概念：** - **术语：** ... - **定义：** ... - **应用场景：** ... **金句/洞见：** > "..." **可出题点：** - [ ] 决策逻辑题 - [ ] 数据应用题 -
合并去重
- 同一会话的多个学习点 → 合并到一张卡片
- 不同会话的学习点 → 分别生成卡片
每周任务：汇总本周学习卡片（周日 20:00）
-
扫描本周学习卡片
- 位置：
memory/learning-cards/ - 筛选条件：
- 文件名格式：
YYYY-MM-DD-*.md - 日期范围：本周一到本周日
- 排除：TEMPLATE.md
- 文件名格式：
- 位置：
-
提取卡片内容（增强版） 从每张卡片中提取：
- 主题（文件名中的主题部分）
- 决策过程（为什么选择A而非B）
- 关键数据（数字+意义）
- 踩坑记录（问题+方案+教训）
- 新概念（术语+定义+场景）
- 金句/洞见
- 可出题点（用于生成考题）
- 耗时
-
生成汇总文档
- 输出位置：
memory/learning-summaries/YYYY-Www-summary.md - 例如：
memory/learning-summaries/2026-W12-summary.md - 文档结构：
- 本周学习概览
- 学习主题详情（每个主题一小节）
- 踩坑汇总
- 关键命令速查
- 下周学习计划（可选）
- 输出位置：
-
转换为 Word（可选） 如果用户需要 Word 格式，调用 Python 脚本转换：
python3 scripts/convert-to-docx.py memory/learning-summaries/2026-W12-summary.md输出：
memory/learning-summaries/2026-W12-summary.docx
文档模板
见 references/summary-template.md
与导师 Agent 配合
- 导师负责日常创建学习卡片
- 秘书负责定期汇总成文档
- 两者共享
memory/learning-cards/文件夹
手动触发
用户说以下任意一句时触发：
- "汇总本周学习"
- "生成本周总结"
- "把这周学的整理一下"
- "生成学习周报"
参考资源
references/summary-template.md：汇总文档模板scripts/convert-to-docx.py：Markdown 转 Word 脚本memory/handover/learning-exam-system.md：学习考试系统Handover（含增强版格式详细说明）
质量检查清单
生成学习卡片时自检：
- 每个主题都有"场景/背景"
- 有决策/选择的都有"决策过程"表格
- 每个数字都有"意义"说明
- 每个坑都有"问题+方案+教训"
- 每个新概念都有"术语+定义+场景"
- 标注了"可出题点"
与考试系统配合
学习卡片是出题的主要来源：
- 周五17:30（或配置的时间）出题时，扫描本周学习卡片
- 提取"可出题点"生成考题
- 决策过程 → 出推理题
- 关键数据 → 出应用题
- 踩坑记录 → 出情景题
- 新概念 → 出概念题
Usage Guidance
Review this carefully before installing. It may be useful if you want an automated learning diary, but you should only enable it if you are comfortable with broad chat-history scanning and persistent memory files. Consider requiring manual approval, limiting which sessions or channels can be scanned, and removing the unrelated SOUL.md behavior rules.
Capability Assessment
Purpose & Capability
The learning-card purpose is coherent, but the artifacts require broad cross-session conversation scanning and persistent reuse of derived notes, which is high-impact access without clear exclusions, retention limits, or approval boundaries.
Instruction Scope
SOUL.md contains broad role, style, and auto-trigger rules that appear unrelated to daily learning-card generation and could affect general agent behavior beyond the invoked skill.
Install Mechanism
There is no install spec and the static scan is clean, but the documentation references manual installation from a GitHub repository and a HEARTBEAT.md schedule file that is not included in the manifest.
Credentials
The skill asks to use session-listing and session-history access across all sessions, then write summaries into memory folders; this is sensitive and not clearly bounded to user-selected chats.
Persistence & Privilege
The skill describes recurring daily and weekly automation plus persistent memory files that are shared with other learning/exam workflows, without clear opt-out, deletion, or containment guidance.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install daily-learning-card - After installation, invoke the skill by name or use
/daily-learning-card - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.2
- Added detailed documentation outlining daily and weekly triggers, core features, output structure, and usage scenarios.
- Enhanced learning card format to include scene/background, decision tables, key data with meaning, pitfalls with lessons, concepts with definitions, and quotable insights.
- Weekly summary workflow now covers learning overview, theme-by-theme details, pitfall aggregation, cheat sheet, and an optional next-week plan.
- Supports both scheduled and manual triggers for daily card generation and weekly summary.
- Integrated checks to ensure completeness and quality of generated learning cards.
v1.0.1
- Updated skill metadata: added English "title" and expanded the English description.
- Improved documentation language by adding parallel English headings and summary at the top.
- No changes to core functionality; all workflows and usage instructions remain the same.
- Enhanced readability for broader audiences while preserving all detailed instructions.
v1.0.0
- 首次发布：daily-learning-card 1.0.0。
- 每天10:00自动扫描前一天对话，智能提取6类核心学习信息，生成结构化学习卡片。
- 每周日20:00自动汇总本周学习卡片，输出周总结文档，可按需转换为Word格式。
- 支持指定关键词手动触发日卡生成与学习周报汇总操作。
- 卡片与周报均采用增强版模板，详细记录决策过程、关键数据、踩坑经验、新概念与金句，适配后续考试出题流程。
- 内置质量检查清单，确保学习卡片结构完整与可出题性。
Metadata
Frequently Asked Questions
What is Learning Secretary Skill?
Daily Learning Card. It is an AI Agent Skill for Claude Code / OpenClaw, with 17 downloads so far.
How do I install Learning Secretary Skill?
Run "/install daily-learning-card" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Learning Secretary Skill free?
Yes, Learning Secretary Skill is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Learning Secretary Skill support?
Learning Secretary Skill is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Learning Secretary Skill?
It is built and maintained by vlarkspur (@vlarkspur); the current version is v1.0.2.
More Skills