PairLearn

pending

by Boyan Lin

AI-copiloted learning — build curriculums, track progress, and continue sessions with any LLM.

Updated 11d agoMITDiscovered via Obsidian Unofficial Plugins
View on GitHub

PairLearn

AI-copiloted learning for Obsidian. Build curriculums, track progress, and continue sessions with any LLM — Claude, ChatGPT, Codex, or a local model.

PairLearn pairs you with AI to learn anything. Everything is stored as plain Markdown, so you're never locked into one tool.

Installation

From Obsidian Community Plugins

  1. Open Settings → Community Plugins → Browse
  2. Search for PairLearn
  3. Click Install, then Enable

Manual Install

  1. Download main.js, manifest.json, and styles.css from the latest release
  2. Create a folder .obsidian/plugins/pairlearn/ in your vault
  3. Copy the three files into it
  4. Enable PairLearn in Settings → Community Plugins

Getting Started

  1. Click the graduation cap icon in the left ribbon (or use command palette: "Open PairLearn")
  2. Click + New Topic to create your first learning topic
  3. Use an AI to build your curriculum:
    • Claude Code: Use /learn and say "I want to learn Rust"
    • ChatGPT / any LLM: Copy Learning/CONTEXT.md + your topic's progress.md into a chat and say "Help me build a curriculum"
  4. Come back to PairLearn to see your progress dashboard

How It Works

Learning/                     # In your Obsidian vault
├── INDEX.md                  # Master index of all topics
├── CONTEXT.md                # Instructions for any LLM (portability layer)
├── <topic-slug>/             # One folder per learning topic
│   ├── plan.md               # Versioned curriculum
│   ├── progress.md           # Current state — THE key file
│   └── lessons/              # Session logs (append-only)
│       └── YYYY-MM-DD-HH-<slug>.md

The Key Files

  • progress.md — Source of truth for where you are. Any LLM can read this and know exactly where to pick up.
  • plan.md — Your curriculum. Versioned — old versions archived inline under <details> blocks.
  • CONTEXT.md — Paste this + progress.md into any AI to continue learning anywhere.

Plugin Features

  • Learning Hub — Dashboard of all topics with status, progress bars, and filters
  • Topic Dashboard — Current position, chapter list, confidence ratings, lesson history
  • Interactive confidence stars — Click to rate your understanding (1-5) per chapter
  • Chapter checkboxes — Mark chapters complete directly from the dashboard
  • Status management — Pause, archive, and restore topics
  • Auto-refresh — Dashboard updates when files change (e.g., after an AI session)
  • Open files — Click to open plan, progress, or lesson files in Obsidian

Using with AI Tools

Claude Code (recommended)

/learn                     # Opens the learning skill
"I want to learn Rust"     # Starts a new topic
"Continue Solana"          # Picks up where you left off
"How's my progress?"       # Shows all topics

ChatGPT / Any Chat LLM

  1. Copy Learning/CONTEXT.md
  2. Copy your topic's progress.md
  3. Paste both into a new chat
  4. Say "Let's continue" — the AI knows exactly where you are

Codex / Agentic Tools

Point the agent at Learning/. CONTEXT.md has all instructions needed.

Concepts

Confidence Scale (1-5)

RatingMeaning
1Lost — need to start over
2Shaky — get the idea, can't apply it
3Okay — can do it with references
4Solid — can do it independently
5Mastery — can teach it

Topic Status

StatusMeaning
activeCurrently studying
pausedOn hold, will resume
completedFinished the curriculum
archivedNo longer pursuing

Plan Versioning

  • Minor (v1.1, v1.2): Reordering, adding sub-topics, adjusting scope
  • Major (v2.0): Fundamental restructure, different learning approach

Design Principles

  1. Plain Markdown — no databases, no proprietary formats
  2. AI-portable — works with any LLM via CONTEXT.md + progress.md
  3. Append-only lessons — sessions are never overwritten
  4. Versioned plans — curricula evolve, old versions are kept
  5. Human-readable — browse everything in Obsidian, a text editor, or on GitHub

Development

npm install
npm run dev    # Watch mode
npm run build  # Production build

License

MIT

For plugin developers

Search results and similarity scores are powered by semantic analysis of your plugin's README. If your plugin isn't appearing for searches you'd expect, try updating your README to clearly describe your plugin's purpose, features, and use cases.