LessonFolk

Allow your AI chat to access and teach you Lessonfolk courses.

Hosted MCP Server

npx add-mcp 'https://lessonfolk.com/mcp'

Installs into Claude Code, Codex, Cursor and more

Documentation

Learn AI from zero to advanced, with an AI as your tutor.

lessonfolk.com

https://github.com/user-attachments/assets/bc8eccbd-1f95-4d20-92f5-15d528b05d47

LessonFolk is a collection of open-source AI courses designed to be taught by an AI. Connect your AI chat (Claude Code, Codex, Claude Desktop, Cursor…) to LessonFolk, say "let's start", and it guides you through lessons in a conversation: explaining, asking questions, adapting to your level, and saving your progress.

How it works

LessonFolk is a website and an MCP server (Model Context Protocol, a standard way for AI apps to use tools). Your AI chat connects to it: it gets the courses and the tutor's instructions, and saves your progress there. The website shows your courses, your path and your progress.

The tutor asks about your level and interests, recommends a path of courses, and teaches one short lesson at a time. Courses are grouped by theme: Understanding AI, Using AI tools, Building with AI and AI and society. The full list, in the recommended order, is courses/en/index.yaml, or ask your tutor "what can I learn?".

Coming soon: Projects. Real projects to practise what the courses teach: a subject, a few milestones to reach, and your tutor as a coach who gives hints instead of the solution. Some will close a course, others will stand alone.

Quick start

Pick one of two ways:

Hosted (lessonfolk.com): sign in with GitHub or Google, open its Connect page and add LessonFolk to your AI chat (in Claude Code, a plugin does it for you). Your progress is saved with your account.

Self-hosted: private, on your computer, no data sent to us. With Docker running:

git clone https://github.com/CGSeb/lessonfolk.git
cd lessonfolk
docker compose up -d --build

Open http://127.0.0.1:4321 to see the dashboard. Then start your AI chat in this folder:

Claude Code: run claude and approve the lessonfolk MCP server when asked.
Codex: run codex mcp add lessonfolk --url http://localhost:4321/mcp once, then codex.
Other apps: follow the dashboard's Connect page.

Then say:

Let's start learning AI.

The learner guide has the details.

Contribute

LessonFolk grows with its courses, and writing one needs no code. Open your AI agent in this folder and say "create a course about…": it interviews you, proposes an outline, writes the lessons with you and checks them. You are credited as the author on the dashboard.

Spotted a mistake in a lesson, or have a course idea? Open an issue.
Want to write or translate a course, or work on the code? Start with Contributing.
A question? Ask in Discussions.

Documentation

Using LessonFolk: the learner guide (first session, what to say, the dashboard, Docker, troubleshooting)
Contributing: writing courses, development setup, workflow
Developing LessonFolk: repository map, scripts, database changes
Settings: every environment variable
Course format: how courses and lessons are written
Testing: automated tests and manual checks
Sign-in for developers: GitHub and Google sign-in
Privacy and personal data: what is stored, export and deletion
Brand: logo, colours and type
All docs

License

LessonFolk is developed by CG Seb.

Code (dashboard, tooling, tutor instructions): MIT
Course content (courses/): CC BY 4.0. You may share and adapt the courses, including commercially, as long as you give appropriate credit.