convez.ai
Skill
taliamekh

course-notes-skill

taliamekh/course-notes-skill

Ver no GitHub
InicialPluginScore · de 0 a 100

Interactive course notes skill for Claude Code — Deep Space dark theme, supports engineering, math, stats, CS, business, physics

Instalação

$npx convez-ai taliamekh/course-notes-skill

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
0
Forks
0
Acompanhando
0
Saúde
28%
Último commit
há 5 meses
Versão
—
Issues abertas
0
Contribuidores
1

README

A tradução para o português deste README ainda está na fila.

Course Notes Skill

A skill that turns lecture material into interactive, color-coded HTML study notes for university courses. Built around a single-page web app with chapters, worked examples, assignment solutions, a reference sheet, per-assessment prep sections, and a checkpoint-based progress tracker that persists locally.

Works with any AI assistant that supports skills or custom instructions. Cross-conversation memory is handled through a progress.md file in a GitHub repo — so a new chat can pick up exactly where the last one left off without re-explaining anything. NotebookLM can optionally supplement this for source material querying.

How it works

  1. One-time course setup. First time a new course is mentioned, the skill asks four questions: course name + code, course structure (tests / midterm / final / assignments / labs / quizzes), how content is organized (chapters / lectures / modules), and theme choice. The answers are saved in progress.md inside the course's GitHub repo.

  2. Material is added chapter by chapter — uploaded slides, textbook PDFs, handwritten notes, assignment sheets, past exams. The skill reads every page, inventories diagrams and figures, and decides what to embed vs. transcribe.

  3. Cross-chapter memory via progress.md. Before writing a new chapter, the skill pulls a concepts index from progress.md that lists every concept and formula already covered. Anything previously defined gets a back-reference ("recall from Ch2 that F=ma"), not a re-explanation. After writing, the index is updated so the next chapter does the same. NotebookLM can optionally supplement this for querying source material, but isn't required.

  4. HTML output. Notes are rendered as a styled single-page app with:

    • A dynamic menu that only shows sections relevant to the course (no "Midterm prep" if there's no midterm)
    • Color-coded content blocks for equations, definitions, tips, derivations, worked examples, and units
    • Full solutions with thought process → setup → step-by-step work → boxed answer → exam tips, built from scratch as a teaching artifact
    • A reference sheet where every equation/theorem/algorithm/framework is a mini-lesson (statement → plain-english meaning → origin → when to use → when NOT to use → examples)
    • Per-assessment prep sections generated from the course config (Test 1, Test 2, Midterm, Final, etc.) — each with scope banner, topic checklist, condensed review, must-know items, practice problems, and solved past exams
    • A progress tracker with hierarchical checkboxes and checkpoint dividers at each test/midterm/final, persisting via localStorage
  5. Subject-aware vocabulary. The same structure adapts to engineering (FBDs, unit analysis), math (proofs, lemmas), stats (distributions, hypothesis tests), CS (pseudocode, Big-O, trace tables), or business (frameworks, case studies). Workflow and theme don't change — only the content blocks do.

Smart task routing

Not every interaction needs the full workflow. The skill classifies tasks into tiers to avoid loading unnecessary context:

TierWhenWhat loads
1 — Targeted fix"Fix the sidebar color," "the popup is broken"Only the named file
2 — Scoped addition"Solve these problems for Ch3," "add Test 2 prep"Relevant fragment files
3 — New chapter"Write Chapter 6 notes from these slides"progress.md + chapter fragments
4 — New course"I'm starting a new course"Nothing — runs setup questions

Themes

30 themes are available, each with a complete color set (equations, definitions, tips, derivations, examples, units) and a paired Google Font. Users can pick by name or number, preview before committing, or save a default with "make this my default."

Live preview: taliamekh.github.io/course-notes-skill/references/theme-previews.html

#ThemeVibe
1Deep SpaceDark purple cosmos with teal and gold pops
2Pastel DreamSoft dreamy pastels on warm white
3Pink CloudBlush pink with rose, orchid, coral
4Hacker TerminalGreen-on-black retro terminal
5VS Code DarkDark+ with official syntax colors
6Earth & StoneWarm terracotta, moss green, sandstone
7Notebook ClassicLined paper with blue ink and red margin
8Grid PaperEngineering grid with pencil-grey tones
9Botanical GardenSage greens, soft pinks, cream
10Ocean DepthsDeep sea blues with bioluminescent accents
11Sunset GradientWarm oranges and pinks into purple
12Forest CanopyDeep woodland greens with autumn gold
13Arctic FrostIce whites and pale blues, crisp and clean
14Lavender FieldsSoft purples and lilacs on light mauve
15Midnight NavyClassic navy with blue and cream accents
16Engineering BlueprintBlueprint blue-on-white with drafting feel
17Med SchoolClinical white with anatomy accents
18Business FormalCharcoal and navy with gold accents
19CS TerminalDark IDE with syntax rainbow highlights
20Law ReviewCream parchment with burgundy and navy
21Chemistry LabPeriodic-table with element colors
22Architecture StudioMinimalist concrete with precise lines
23Math ChalkboardDark green chalkboard with chalk text
24Art StudioHandpainted bold expressive on canvas
25Neon CyberpunkElectric neons on pitch black
26Vintage LibraryAged paper with warm leather-brown
27Mint FreshCool mint green with clean white
28Cherry BlossomJapanese-inspired pink and white
29Coffee ShopWarm espresso browns and creamy latte
30Northern LightsAurora — green, violet, blue on arctic sky

Installation

This is an AI skill — it needs to live inside an AI product that supports custom instructions or skills.

1. Download the skill

Option A — Clone with git:

git clone https://github.com/taliamekh/course-notes-skill.git

Option B — Download as ZIP:

Download main.zip, then extract it and rename the resulting folder from course-notes-skill-main to course-notes.

2. Install in your AI tool

Place the skill folder where your AI tool reads custom skills or instructions. This varies by platform:

PlatformTypical location
Claude Code / Desktop~/.claude/skills/course-notes/
Claude.ai (web)Settings → Capabilities → Skills → upload
Other AI toolsCheck your tool's custom instructions or skills documentation

After install, open a new chat so the skill gets picked up.

3. NotebookLM (optional)

NotebookLM adds the ability to query source material (textbook PDFs, lecture slides) directly from within the workflow. The core skill works fully without it — progress.md handles all cross-session memory and cross-chapter referencing.

If you want the supplement:

pip install notebooklm-mcp-cli
nlm login

Then grab the .mcpb extension from jacob-bd/notebooklm-mcp-cli/releases and install it in your AI tool's MCP/extension settings.

4. Use it

Just ask. The skill triggers whenever you mention course notes, lecture notes, an assignment, exam prep, or a specific class — e.g.:

  • "Write notes for Chapter 3 of my Dynamics class"
  • "Help me prep for my MECH 2005 midterm"
  • "Fix the sidebar toggle on my notes"

Repository contents

The skill uses a multi-file architecture to keep context lean. The main SKILL.md contains the core workflow (~380 lines), while detailed specs live in reference files that only load when the task requires them.

SKILL.md                              Core workflow, task routing, setup, structure
references/
  themes.md                           30-theme color table, font options, UI specs
  component-styles.md                 Card, sidebar, pill, and chapter title CSS
  formula-sheet-spec.md               KaTeX popups, 3-view toggle, variable coloring
  diagram-rules.md                    SVG generation, label collision math
  solution-pipeline.md                7-step solution format, assessment prep, progress tracker
  summary-spec.md                     Concept clusters, What/Why/How format, quick ref table
  theme.md                            Deep Space component templates (work with any theme)
  theme-previews.html                 Static preview of all 30 themes side-by-side
scripts/
  push_to_github.py                   GitHub Contents API push (base64 encode + PUT)

License

MIT — use it, fork it, theme it.

Ver o original no GitHub

Bom para

Relacionados