31carlton7/realm/lecture-notes

Claude Code and agent skill by 31carlton7/realm

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Use during or after a class — answering questions from what the lecture just covered, cleaning up notes typed in class, folding in a recording's transcript, turning a lecture into flashcards and a study guide. Works on the dated Markdown files under lectures/ in a course space; reads the course folder with docs_search and never invents what the lecture said.

Install

npx vskill@latest install 31carlton7/realm/lecture-notes

SKILL.md

lecture-notes — before, during and after a class

A lecture is a Markdown file at lectures/YYYY-MM-DD-<topic>.md in the course's space folder, with front-matter (course, title, date) and the sections ## Notes, ## Questions, ## Follow-ups. A recording imported from Plynn adds ## Transcript (verbatim) after a rule. Realm's Documents pane shows the file live: edit it with your normal file tools and the user sees the change.

The rule that governs everything here: the lecture is the record. Your job is to organise and connect what was said, not to write the lecture you would have given. Mark inference as inference.

During class (the lecture session)

The user is in class, typing under ## Notes and dropping questions under ## Questions. When asked something:

  1. Read the lecture file first; the last few lines are the context.
  2. docs_search the course folder (lectures/, slides/, guides/) for the term — the answer is

usually in an earlier lecture or the slide deck, and should cite it by path.

  1. Answer briefly. Two paragraphs, then offer to go deeper. The user is listening to someone else.
  2. If the question is in ## Questions, write the answer under it in the file (indented under the

bullet) rather than only in chat, so it survives.

Do not rewrite the notes while class is running; the user is typing in the same file.

After class (the wrap-up)

When asked to wrap up a lecture (Realm's "Wrap up a lecture" sends this request with the file path):

  1. Read the whole file. If there is a ## Transcript, it is what was said; ## Notes is what the

user found worth typing — the emphasis. Without a transcript, work from the notes and the course materials and say so at the top of the notes.

  1. Connect: docs_search for the lecture's main terms across lectures/ and guides/, then the

whole folder (slides are PDFs; their text is searchable). Note which earlier lecture introduced each idea.

  1. Rewrite `## Notes` into structured notes: headings per idea in the order taught, the user's

own phrasing kept where it is theirs, transcript material added where the notes are thin, and anything inferred marked (inferred). Definitions get their own line. Worked examples stay worked. Keep it under roughly two screens.

  1. Answer `## Questions` in place, under each bullet, citing the transcript or source file.

A question the material does not answer says so and goes to ## Follow-ups.

  1. Keep `## Follow-ups` as a checklist (- [ ]), adding what the lecturer said to do.
  2. Append `## Flashcards`: 8–15 **Q:** … / **A:** … pairs on the core ideas, no trivia.
  3. Leave the transcript untouched. It is evidence.
  4. Build or extend the study guide for the topic under guides/ — follow the study-guide skill

if it is available (self-contained HTML, rg-quiz per concept with data-topic, rg-steps for procedures, KaTeX for math, a sources footer naming this lecture file). Extend an existing guide on the same topic rather than creating a second. Then docs_open it.

  1. Reply with five lines: what changed, what to review first, and any question left open.

Reviewing before an exam

docs_progress on a guide reports best/last per topic and which are weak. Build the review from the weak topics: re-read those lectures, add targeted questions to the guide's existing quizzes (keep the data-topic slugs so history survives), and list the three things to redo by hand.

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
85
Repository forks
5
7-day trend score
22

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/31carlton7/realm/lecture-notes/badge)](https://verified-skill.com/skills/31carlton7/realm/lecture-notes)

Something wrong with this skill or its page? Tell us.