An agent-native teaching OS — a classroom that remembers, for you and your AI. MCP server + self-hosted web app.
View on GitHub ↗Your AI is becoming the one who knows you best. Learn Shell gives it a classroom that remembers — an agent-native teaching OS that turns your AI's teaching intent into things a learner can actually study, practice against, and get graded on.
Codename: learn-shell · Status: working system, actively dogfooded since 2026-07-02. Not authenticated. MIT-licensed; v1 is a single-machine, no-auth trust model.
Setting this up from a code snapshot? Follow SETUP.md — from-zero bring-up (Docker Postgres → migrate → seed → run → connect your own AI over MCP), every command verified against this repo's real config. Once it's running, the product tutorial (Chinese-first: cold-start pitfalls, the recommended lesson path, the agent's onboarding letter) is docs/TUTORIAL.md.
If you already have an AI you talk to daily — a work assistant, a long-running companion — it is quietly becoming the person who knows you best: it has read your every answer and seen your every hesitation. What it's missing is not knowledge of you. It's a floor. A chat window has no floor: the evidence of teaching evaporates with the context window, every new session your teacher wakes up a stranger, and you introduce yourself, again, to the one who knows you best.
Learn Shell is the floor. It doesn't replace your AI — it gives its teaching somewhere to live: courses, lessons, flashcards (FSRS-scheduled), exercises with grading, a mindmap, a document reader, and a memory layer the agent reads back (getcontext / getlearnerbrief) instead of re-deriving you from scratch every session. The teacher can change sessions, ch
From the project README.
Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:
[](https://mcp.liqiwa.com/s/sf-shenfeng--learn-shell.html)
Self-hosted, multi-user spaced-repetition (SRS) service: FSRS v6 scheduling, 10 card types, Markdown/TeX, REST API and built-in MCP server. Written in Go.
rwnalds/engramSelf-hosted MCP server + dashboard giving Claude Code, Cursor & Hermes agents shared long-term memory over a git-backed markdown vault. Agent-native, Obsidian-compatible second brain — markdown RAG, n
motif-Labs/motifCoding agents are single-player: each session starts cold, in its own tool, blind to what the rest of the team already worked out. Motif makes your team's multiplayer — every Claude Code, Codex and Cu
docouno/notariumFile-first knowledge base for humans and AI agents — self-hosted Markdown notes with a web editor and a built-in MCP server.
flytomoon/al-buddy-memoryPortable, governed, model-agnostic memory for AI agents: facts invalidated never deleted, raw text as truth, one documented export format.
osaxyz/sepiaceLong-term memory for coding agents. The sepiace plugin for Claude Code and Codex.
The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.