From Look Tongji Notes
Generates structured Markdown study notes from lecture transcripts and slides. Runs transcript and slide download in parallel, then writes a timeline outline and detailed notes.
How this skill is triggered — by the user, by Claude, or both
Slash command
/look-tongji-notes:noteThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Generate detailed study notes from lecture transcript and slides.
Generate detailed study notes from lecture transcript and slides.
/note or "write notes for this lecture".python "<SKILL_DIR>/../../scripts/look_tongji.py" note --course-id "<ID>" --sub-id "<ID>"
python "<SKILL_DIR>/../../scripts/look_tongji.py" note \
--course-id "<ID>" --sub-id "<ID>" \
--material "slides-from-teacher=/path/to/file.pdf"
note Re-Runs Transcriptionnote always re-downloads the video and re-runs ASR, even if the lecture was already transcribed. This is by design — note = transcribe + slide + note in one combined operation.
If you already ran transcribe separately, do NOT run note. Instead:
slide to download slides (if not already done)This saves ~4 minutes of re-downloading and re-transcribing per lecture.
When scripting or working non-interactively:
| Flag | Description |
|---|---|
--no-slide | Skip slide download (transcript only) |
--no-material-prompt | Skip "any supplementary materials?" prompt |
For --no-material-prompt, use with --material to supply materials non-interactively:
python "<SKILL_DIR>/../../scripts/look_tongji.py" note \
--course-id "<ID>" --sub-id "<ID>" \
--no-material-prompt \
--material "slides=/path/to/file.pdf"
After the CLI finishes, the agent must:
MM:SS-MM:SS:课程阶段内容python "<SKILL_DIR>/../../scripts/timeline_tools.py" timeline-normalizeAfter the CLI finishes, ask the user for their preferred note style (unless they already stated one). Options:
### headings.Write notes into the session raw-data folder:
<course_id>_<sub_id>_notes.md<course_id>_<sub_id>_dialogue.mdAlso read supplementary material conversions when available:
materials/<name>/converted.mdCRITICAL: Notes must be PURE knowledge content. No task-oriented meta descriptions.
FORBIDDEN phrases and patterns:
Self-check: After writing notes, the agent MUST re-read the output and remove any sentence that describes HOW the notes were made rather than WHAT was learned.
Read transcript TXT (not JSON) and slide images directly.
Use LaTeX $...$ / $$...$$ for formulas.
If slides and transcript conflict, prefer slide text.
Use the appropriate note prompt based on the chosen style:
If standard style (default):
You are a professional course TA. Based on the provided ASR transcript and lecture slide snapshots, write detailed study notes in Markdown (notes content in Simplified Chinese).
Requirements:
1) Output notes directly. No polite preface. No "here is the summary" opener.
2) Make the text fluent and logically structured. Fix obvious ASR errors and repetitions, but do not fabricate content not present in transcript/slides.
3) Markdown formatting: only use headings starting from ### (allow ###/####/#####). Do not use # or ##. Use bold/lists/tables when appropriate; avoid overly fragmented bullet-only output.
4) If the lecture mentions assignments/exams/attendance/grouping, put a short "### Course Reminders" section at the very top.
5) Use LaTeX for variables and formulas: inline $...$, block $$...$$. Do not put non-ASCII characters inside LaTeX.
6) Be faithful to transcript/slides and include enough details so that a student can learn from the notes (not just an outline).
7) If transcript and slides conflict, prefer slide text for terminology/spelling and briefly note uncertainty.
Now write the notes from transcript + slides:
If dialogue style:
You are a professional course TA. Based on the provided ASR transcript and lecture slide snapshots, write study notes in **dialogue format** in Simplified Chinese.
CRITICAL OUTPUT RULES:
1) Output notes directly. No polite preface. No "here is the summary" opener.
2) Write in a natural Q&A conversation between "**老师**" (teacher) and "**学生**" (student):
- "学生" raises questions that a real learner would have at each stage
- "老师" answers using the lecture content and slides
- Switch roles naturally as the lecture progresses through topics
3) Strict format for each exchange (one blank line between turns):
**学生**:[学生提出的问题或疑惑]
**老师**:[老师的解答和讲解,包含知识点、公式、示例]
4) Cover ALL major topics from the lecture in order. Do not skip sections.
5) Use LaTeX for variables and formulas: inline $...$, block $$...$$. Do not put non-ASCII characters inside LaTeX.
6) Be faithful to transcript/slides. Fix obvious ASR errors and repetitions, but do not fabricate content not present in transcript/slides.
7) If transcript and slides conflict, prefer slide text for terminology/spelling and briefly note uncertainty.
8) If the lecture mentions assignments/exams/attendance/grouping, include it naturally in the dialogue.
Now write the dialogue notes from transcript + slides:
python "<SKILL_DIR>/../../scripts/look_tongji.py" index
python "<SKILL_DIR>/../../scripts/look_tongji.py" build
<course_id>_<sub_id>_timeline.txt — timeline outline<course_id>_<sub_id>_notes.md — study notes (standard style)<course_id>_<sub_id>_dialogue.md — study notes (dialogue style)<SKILL_DIR> Points<SKILL_DIR> is the directory containing this SKILL.md. Shared scripts (look_tongji.py, timeline_tools.py, tongji_backend/) and references live two levels up in the repository root (<SKILL_DIR>/../../scripts/ and <SKILL_DIR>/../../references/).
npx claudepluginhub walkerkiller/look-tongji-notes --plugin look-tongji-notesCrawls online course pages to extract lecture schedules, slides, and videos, then synthesizes structured vault notes per lecture. Also refines existing course notes.
Transforms messy lecture transcripts, OCR notes, screenshots, and scattered learning materials into structured WPS study notes for review and revision.
Applies Cornell Notes method to capture lecture or reading notes in a structured three-section format that builds review and retrieval practice.