词元之母TOK.MOM - 平台充值汇率 1:1 即 1 人民币充值到账 1 美元,支持一个 Key 调用近 600+ 海内外模型,限时特价模型低至 1 折,欢迎上岸!
| 来源 | 可选 — 使用 hermes skills install official/productivity/memento-flashcards 安装 |
| 路径 | optional-skills/productivity/memento-flashcards |
| 版本 | 1.0.0 |
| 作者 | Memento AI |
| 许可证 | MIT |
| 平台 | macos, linux |
| 标签 | Education, Flashcards, Spaced Repetition, Learning, Quiz, YouTube |
| 用户意图 | 操作 |
|---|---|
| "记住 X" / "将此保存为闪卡" | 生成问答卡片,调用 memento_cards.py add |
| 发送事实但未提及闪卡 | 询问"要将此保存为 Memento 闪卡吗?" — 仅在确认后创建 |
| "创建一张闪卡" | 询问问题、答案、集合;调用 memento_cards.py add |
| "复习我的卡片" | 调用 memento_cards.py due,逐张呈现卡片 |
| "用 [YouTube URL] 测验我" | 调用 youtube_quiz.py fetch VIDEO_ID,生成 5 道题,调用 memento_cards.py add-quiz |
| "导出我的卡片" | 调用 memento_cards.py export --output PATH |
| "从 CSV 导入卡片" | 调用 memento_cards.py import --file PATH --collection NAME |
| "显示我的统计" | 调用 memento_cards.py stats |
| "删除一张卡片" | 调用 memento_cards.py delete --id ID |
| "删除一个集合" | 调用 memento_cards.py delete-collection --collection NAME |
~/.hermes/skills/productivity/memento-flashcards/data/cards.jsonmemento_cards.py 子命令。该脚本通过原子写入(先写入临时文件,再重命名)来防止数据损坏。Turn the factual statement into a front-back pair.
Return exactly two lines:
Q: <question text>
A: <answer text>
Statement: "{statement}""General" 作为默认值。"General")memento_cards.py add。next_review_at <= now 的卡片 JSON 数组。如需集合过滤:Agent: 柏林墙是哪年倒塌的? 用户: 1991 Agent: 不太对。柏林墙倒塌于 1989 年。下次复习是明天。
(agent 调用:memento_cards.py rate --id ABC --rating hard --user-answer "1991")下一题:第一个登上月球的人是谁?
--rating retire。| 评级 | 间隔 | ease_streak | 状态变化 |
|---|---|---|---|
| hard | +1 天 | 重置为 0 | 保持 learning |
| good | +3 天 | 重置为 0 | 保持 learning |
| easy | +7 天 | +1 | 若 ease_streak >= 3 → retired |
| retire | 永久 | 重置为 0 | → retired |
https://www.youtube.com/watch?v=dQw4w9WgXcQ 中提取 dQw4w9WgXcQ)。{"title": "...", "transcript": "..."} 或错误信息。missing_dependency,告知用户安装:You are creating a 5-question quiz for a podcast episode.
Return ONLY a JSON array with exactly 5 objects.
Each object must contain keys 'question' and 'answer'.
Selection criteria:
- Prioritize important, surprising, or foundational facts.
- Skip filler, obvious details, and facts that require heavy context.
- Never return true/false questions.
- Never ask only for a date.
Question rules:
- Each question must test exactly one discrete fact.
- Use clear, unambiguous wording.
- Prefer What, Who, How many, Which.
- Avoid open-ended Describe or Explain prompts.
Answer rules:
- Each answer must be under 240 characters.
- Lead with the answer itself, not preamble.
- Add only minimal clarifying detail if needed.question 和 answer 字符串。如果验证失败,重试一次。video_id 去重 — 如果该视频的卡片已存在,则跳过创建并报告现有卡片。question,answer,collection(无标题行)。--collection 参数值。total:卡片总数learning:活跃轮换中的卡片retired:已掌握的卡片due_now:当前到期待复习的卡片collections:按集合名称的细分统计cards.json — 始终使用脚本子命令以避免数据损坏youtube_quiz.py 需要 youtube-transcript-api;如果缺失,告知用户运行 pip install youtube-transcript-apiyoutube.com/watch?v=ID 和 youtu.be/ID 两种 URL 格式