repowisehkuds/deeptutor

hkuds / deeptutor Architecture and code health

DeepTutor: Lifelong Personalized Tutoring. https://deeptutor.info/.

Python★ 51,322
Use in your agentAsk this codebase

github.com/hkuds/deeptutorIndexed at 1ea7df1Up to date with upstream

What is deeptutor?

DeepTutor is an agent-native tutoring system for learners who study from their own questions, uploaded learning materials and knowledge bases. It runs these inputs through capability-driven agents and retrieval pipelines to produce tutoring answers, quizzes, visualizations, research notes and mastery-practice artifacts. Learners use it through a web learning workspace, and other agents can drive it through a CLI with an interactive REPL and structured JSON output. The project is written mainly in Python, with a TypeScript web frontend.

Written by a model from the README and the source

3,967 files and 37,003 symbols in 11 modules, led by Python (2,069 files) and TypeScript (1,468).

Start reading

Code health

6.7out of 10Fair

Hotspots average 4.9, the place fixes pay off most.

Full health report

On the leaderboards:#36 of 56 Python repos

Use hkuds/deeptutor in your agent

Point Claude Code, Cursor, Codex or VS Code at this repository. Your agent gets the index this page is built from: cited answers, callers, history and health for any file.

  1. 1.In Claude, open Customize → Connectors, click +, then Add custom connector. On a Team or Enterprise plan an owner adds it first under Organization settings → Connectors.
  2. 2.Paste the server URL below as the Remote MCP server URL, then click Add.
  3. 3.Click Connect. A Repowise sign-in window opens. Approve access.
  4. 4.Repowise's tools are now available in your Claude chats.

Connect with one click. Sign in with Repowise (OAuth), no API key to copy. Works in Claude on the web, desktop, and mobile.

Docs
bash
https://api.repowise.dev/mcp/hkuds/deeptutor
Editors need an API key: sign in to create one.Sign in to connect

This repo is public, so anyone with a Repowise account and an API key can connect.

Try asking

  • get_risk(["deeptutor/capabilities/mastery/loop.py"])
  • get_answer("how is deeptutor put together?")
  • get_why("why is deeptutor structured this way?")

Explore deeptutor

Key features of deeptutor

  • One runtime and session context serve Chat, Ask Questions, Quiz, Research, Visualize, Solve, Mastery Path, Immersive Reading and Audio Overview while each keeps its own loops and pipelines.
  • Task Board tracks study tasks in To do, In progress and Done, with notes and an archive you can restore from.
  • Multi-engine knowledge supports versioned RAG libraries across LlamaIndex, PageIndex, GraphRAG, LightRAG and other sources, with pluggable document parsing.
  • Immersive video learning adds YouTube or Bilibili links to Reading for native playback, transcript search, timestamp-grounded tutoring and saved notes.
  • Inspectable memory records L1 traces, L2 surface summaries and L3 synthesis, and a Memory Graph links them so personalization can be viewed and edited.

How hkuds/deeptutor works

Input enters through two front doors: the web frontend in web, which renders screens and sends typed requests to the server, and the CLI in deeptutor_cli/__main__.py, which turns typed commands into requests and prints rich text or JSON. Requests reach the API server started from deeptutor/api/main.py. The deeptutor package assembles the application: it reads runtime settings, picks a coordination backend, and builds one container that holds the turn services, the capability registry and per-session runtime bookkeeping. The model-driven workers in deeptutor/agents take a learner's request and its settings and handle features such as chat, research, question generation, visualization and math animation. These workers return streamed results and structured output, which become the answers, quizzes and other artifacts the learner sees.

The codebase has 3,967 files and 37,003 symbols in 11 modules, led by Python, TypeScript and JavaScript. Code health is 6.7 out of 10, rated fair.

hkuds/deeptutor FAQ

What is DeepTutor?

DeepTutor is an agent-native tutoring system that takes learners' questions, uploaded materials and knowledge bases and produces tutoring answers, quizzes, visualizations and research notes. It also supports mastery-practice artifacts.

How does DeepTutor work?

DeepTutor runs learner inputs through capability-driven agents and retrieval pipelines. Learners use a web learning workspace, while other agents can drive it through a CLI that returns structured JSON output.

What language is DeepTutor written in?

DeepTutor is written mainly in Python, with a TypeScript web frontend. Its code also includes some JavaScript, a Dockerfile and a shell script.

How is the DeepTutor code organized?

The deeptutor package assembles the application and holds the capability registry. The deeptutor/agents directory contains the model-driven workers, deeptutor_cli provides the terminal interface and web contains the browser screens and shared components.

Who is DeepTutor for?

DeepTutor is for learners who work through a web learning workspace. Other agents can also use its CLI to drive tutoring sessions and receive structured JSON output.

What license does DeepTutor use?

DeepTutor uses the Apache-2.0 license, as identified by the SPDX id Apache-2.0.

The summary, features and answers on this page are written by a model from the README and the source, and refreshed with each index.

About this page

Get a page like this for your own codebase

Repowise built this map of hkuds/deeptutor from its source and git history. It parses every symbol, draws the dependency graph, scores each file's health from complexity, duplication, coverage and churn, finds hotspots and owners in the history, and writes documentation for every module. Your coding agent reads the same index over MCP, so its answers cite real files.

Index your repo freeSee how Repowise worksFree for 2 public repos, with a page like this one.

Prefer local? pip install repowise && repowise init indexes your own checkout, no account needed.