study-mate: An AI Study Partner for Learning by Doing in DeepSeek Harness

miaotofu01/study-mate

Tools & CapabilitiesVerified
Listed on 2026-09-24
Page last updated 2026-10-04

StudyMate is a DSH "learning mode" preset and skill set that builds study roadmaps, lesson pages and practice assessments for math and CS subjects, with cross-subject shared memory.

study-mate is a learning mode preset plus a set of skills for DeepSeek Harness that turns a conversational AI into a partner able to carry you through an entire subject. It focuses on mathematics and computer science, dispatching five sub-roles on demand for material collection, image sourcing, curriculum design, explanation, and exercise assessment, and produces a full set of illustrated HTML courseware. Knowledge points are arranged by prerequisite dependencies and colored by status, while progress and preferences settle into workspace files so a new conversation resumes from the last node and your level and preferred explanation style carry across subjects.

How to Install

install
dsh plugin --profile web add github:miaotofu01/study-mate
Category
Tools & Capabilities
Platform
DSH Plugin
Author
miaotofu01
Distribution
Plugin

study-mate Key Features

Dependency-based knowledge roadmapHTML lesson pages with illustrationsOn-the-spot quizzes with evidence checksCross-subject shared memoryResumable learning progress
Listed on dsh-plugin.org

study-mate Repo Summary

What Does It Do?

StudyMate is a DSH plugin for DeepSeek Harness that ships a "learning mode" preset plus a set of SKILLs, turning the harness into an AI study partner that can carry you through an entire subject. It focuses on mathematics and computer science topics and follows a "learn with doing" principle, dispatching five sub-roles on demand: material collection, image sourcing, curriculum design, explanation, and exercise assessment. Its output is a complete set of HTML courseware, solving the problem of chat-based learning where context runs out, nothing is retained, and every session starts from zero.

Core Features

  • Curriculum roadmap: knowledge points are arranged by prerequisite dependencies, labeled as concept, hands-on, or lab lessons, colored by status, and expandable into lesson sub-cards in place.
  • Textbook-style courseware: lessons are generated as illustrated HTML pages with exercises and project goals tailored to your level, and the courseware is the main carrier of learning.
  • Cross-subject shared memory: it remembers your current level, which explanation styles work, and your common sticking points, so the next subject needs no reintroduction.
  • Inheritable progress: learning progress is stored in workspace files, so each new conversation continues from the last node.
  • On-the-spot tutoring and quizzing: paste confusing material and it explains a short segment while logging a record; ask to be tested and it generates questions and verifies them against runnable evidence.
  • Engineering validation scripts: home page generation, lesson rendering, and four checks keep the curriculum, lessons, image pool, and skill definitions structurally consistent.

How to Use This Plugin?

After enabling it in DSH, start a new session, choose the learning mode preset, and say you want to learn a subject. Once the first course is generated, the course home page lives at index.html inside the learning workspace and serves as the entry point for all future lessons; learning data is stored in the workspace directory by default, so to move machines copy the whole project including the workspace and rerun the install script as described in the official instructions. To relocate the workspace, change the workspace field in the config file or set the environment variable and rerun the script. In daily use you can simply say things like "I want to learn C++ for competitive programming," "I don't get this part," "quiz me," or "too easy," and the plugin will respectively interview you and build a roadmap, explain on the spot, generate and verify questions, or switch explanation style and record the preference in shared memory. The examples directory in the repository is a complete sample workspace covering linear algebra and computer networks for previewing real output.

This page is an independent rewrite of the plugin's official README — for authoritative documentation and the latest changes, refer to the source: miaotofu01/study-mate. The plugin is third-party code that runs on your machine once installed; inclusion does not imply endorsement — please review the source before installing.

Install, update, and uninstall this plugin in the Plugin Market of DeepSeek Harness's DSH Plugin Hub

DSH Plugin FAQ