[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"pack-detail-ai-personal-tutor-learning-en":3,"seo:pack:ai-personal-tutor-learning:en":98},{"code":4,"message":5,"data":6},200,"操作成功",{"pack":7},{"slug":8,"icon":9,"tone":10,"status":11,"status_label":12,"title":13,"description":14,"items":15,"install_cmd":97},"ai-personal-tutor-learning","🎓","#7C3AED","new","New · this week","AI Personal Tutor + Self-Learning","Ten picks for the adult learner using AI to study on their own — coding, language, math, exam prep. Mr. Ranedeer prompt + document-grounded tutor + Socratic \u002F active-recall agents + structured learning path + reflection + Fabric patterns + NotebookLM + second brain + persistent memory. Install in order: pick topic → tutor prompt → flashcard \u002F active-recall → quiz → review.",[16,28,36,43,50,58,66,75,82,89],{"id":17,"uuid":18,"slug":19,"title":20,"description":21,"author_name":22,"view_count":23,"vote_count":24,"lang_type":25,"type":26,"type_label":27},24,"08be0198-969e-4daa-be30-2fec2a868e87","mr-ranedeer-personalized-ai-tutor-prompt-08be0198","Mr. Ranedeer — Personalized AI Tutor Prompt","GPT-4 system prompt that creates a personalized AI tutor. Adapts to your learning level, language, and style. Used by 30K+ learners.","TokRepo精选",438,0,"en","skill","Skill",{"id":29,"uuid":30,"slug":31,"title":32,"description":33,"author_name":34,"view_count":35,"vote_count":24,"lang_type":25,"type":26,"type_label":27},4178,"bab2efd8-5293-11f1-9bc6-00163e2b0d79","deeptutor-ai-powered-personalized-learning-assistant-bab2efd8","DeepTutor — AI-Powered Personalized Learning Assistant","An agent-native learning platform that uses multi-agent systems and deep research to provide interactive, personalized tutoring from any document or paper.","AI Open Source",77,{"id":37,"uuid":38,"slug":39,"title":40,"description":41,"author_name":22,"view_count":42,"vote_count":24,"lang_type":25,"type":26,"type_label":27},4434,"2cf35906-3ad2-4486-9b1c-f1d756166134","claude-code-agent-microsoft-study-mode-2cf35906","Claude Code Agent: Microsoft Study Mode","Activate your personal Microsoft\u002FAzure tutor - learn through guided discovery, not just answers.",44,{"id":44,"uuid":45,"slug":46,"title":47,"description":48,"author_name":22,"view_count":49,"vote_count":24,"lang_type":25,"type":26,"type_label":27},4322,"5e3e7a37-0303-48d1-b1bb-e05e32cc0c6b","claude-code-agent-demonstrate-understanding-5e3e7a37","Claude Code Agent: Demonstrate Understanding","Validate user understanding of code, design patterns, and implementation details through guided questioning.",31,{"id":51,"uuid":52,"slug":53,"title":54,"description":55,"author_name":56,"view_count":57,"vote_count":24,"lang_type":25,"type":26,"type_label":27},3164,"286cadd2-b509-4e99-8746-64cc27c6a2c1","claude-howto-guided-claude-code-learning-path","Claude Howto — Guided Claude Code Learning Path","Claude Howto is a visual, example-driven roadmap for Claude Code that ships copy-paste `.claude\u002F` templates, quizzes, and a 15-minute quick start.","Prompt Lab",103,{"id":59,"uuid":60,"slug":61,"title":62,"description":63,"author_name":64,"view_count":65,"vote_count":24,"lang_type":25,"type":26,"type_label":27},486,"47d7c959-8340-4390-a163-39cf2e41f7b3","claude-reflect-self-learning-claude-code-47d7c959","Claude Reflect — Self-Learning for Claude Code","A Claude Code plugin that captures corrections and preferences during development, then syncs learnings to CLAUDE.md automatically. Makes Claude remember across sessions. 870+ stars.","Skill Factory",154,{"id":67,"uuid":68,"slug":69,"title":70,"description":71,"author_name":56,"view_count":72,"vote_count":24,"lang_type":25,"type":73,"type_label":74},166,"6692e725-30db-45bd-93c5-65d28891bf2a","fabric-ai-prompt-patterns-everything-6692e725","Fabric — AI Prompt Patterns for Everything","Collection of 100+ AI prompt patterns for real-world tasks. Summarize articles, extract wisdom, analyze code, write essays, create presentations, and more.",188,"prompt","Prompt",{"id":76,"uuid":77,"slug":78,"title":79,"description":80,"author_name":64,"view_count":81,"vote_count":24,"lang_type":25,"type":26,"type_label":27},3635,"7a9502e3-58ab-59a1-b9d5-f7e80b83a4b3","notebooklm-py-notebooklm-cli-python-api-skill","notebooklm-py — NotebookLM CLI + Python API Skill","notebooklm-py provides a CLI + Python API for NotebookLM with agent hooks; verified 13,142★ and uses Playwright Chromium for browser login.",86,{"id":83,"uuid":84,"slug":85,"title":86,"description":87,"author_name":34,"view_count":88,"vote_count":24,"lang_type":25,"type":26,"type_label":27},323,"4cbd3b7b-5251-4a16-a4ef-d7c1f9600d52","khoj-your-ai-second-brain-4cbd3b7b","Khoj — Your AI Second Brain","Khoj is a personal AI app for chat, search, and knowledge management. 33.8K+ stars. Multi-LLM, docs, Obsidian, WhatsApp, custom agents. AGPL-3.0.",162,{"id":90,"uuid":91,"slug":92,"title":93,"description":94,"author_name":95,"view_count":96,"vote_count":24,"lang_type":25,"type":26,"type_label":27},703,"96da1f40-1823-4d87-a84f-7d8269edeb24","mem0-memory-layer-ai-applications-96da1f40","Mem0 — Memory Layer for AI Applications","Add persistent, personalized memory to AI agents and assistants. Mem0 stores user preferences, past interactions, and learned context across sessions.","Mem0",720,"tokrepo install pack\u002Fai-personal-tutor-learning",{"pageType":99,"pageKey":8,"locale":25,"title":100,"metaDescription":101,"h1":102,"tldr":103,"bodyMarkdown":104,"faq":105,"schema":121,"internalLinks":126,"citations":139,"wordCount":152,"generatedAt":153},"pack","AI Personal Tutor + Self-Learning — 10 Tools for the Adult Learner","Mr. Ranedeer, DeepTutor, Microsoft Study Mode, Demonstrate Understanding, Claude Howto, Claude Reflect, Fabric, NotebookLM, Khoj, Mem0 — the AI stack an adult uses to actually learn coding, a language, math, or exam material on their own. Install in order. Active recall + spaced review + persistent memory.","AI Personal Tutor + Self-Learning — From Topic to Mastery, Solo","Ten picks in deliberate order for the adult learner who is the teacher *and* the student: a tutor prompt to talk to, a document-grounded tutor for the textbook you actually own, a Socratic agent and an active-recall agent to stop passive reading, a structured path so you don't wander, a reflection skill so lessons stick, plus Fabric \u002F NotebookLM \u002F Khoj \u002F Mem0 as the supporting cast. Loop: pick topic → tutor prompt → flashcard \u002F active recall → quiz → review.","## What's in this pack\n\nThis is the stack for an adult who decided to learn something hard on their own — a new programming language, conversational Spanish, calculus they avoided in college, the AWS Solutions Architect exam. You are the student *and* the teacher, and the bottleneck isn't access to information (it's never been easier). The bottleneck is **structure**, **active engagement**, and **memory** — turning passive reading into something that actually sticks six months later.\n\nThis pack does **not** assume you have a course or a class. It assumes you have a topic, a chunk of free hours per week, and the willingness to be honest with an AI about what you don't understand. Every pick is open-source or has a clear free tier, and each one is in here because it solves a specific failure mode of solo learning — drifting off-topic, fooling yourself that you understood, forgetting in two weeks what you learned today.\n\nDifferent from the [Teacher \u002F Educator pack](\u002Fen\u002Ftopics\u002Fteacher-educator-ai-lesson-kit): no LMS, no quiz form builder, no slide tool. This is for the learner, not the instructor.\n\n## Install in this order (pick topic → tutor prompt → flashcard → quiz → review)\n\n1. **Mr. Ranedeer — Personalized AI Tutor Prompt** — start here. It's a single GPT-4 \u002F Claude system prompt that turns any chat AI into a tutor that adapts to your level, language, and pace. Zero install. Paste it into Claude or ChatGPT and tell it *I want to learn linear algebra; I'm comfortable with Python; treat me like a smart adult who never took the class.* You've just hired a 24\u002F7 tutor. Spend a week here before adding anything else.\n2. **DeepTutor — AI-Powered Personalized Learning Assistant** — Mr. Ranedeer hallucinates because it doesn't know your textbook. DeepTutor grounds answers in a document *you upload* — the PDF of the chapter, the API docs, the practice exam. Now when you ask \"what does this equation mean,\" the tutor cites your book instead of inventing a similar one.\n3. **Claude Code Agent: Microsoft Study Mode** — a Claude Code skill that teaches through **guided discovery**: it refuses to give you the answer. It asks you what you think the answer is, then nudges. Even outside Microsoft topics, the *pattern* is the antidote to \"AI wrote the code, I didn't learn anything.\"\n4. **Claude Code Agent: Demonstrate Understanding** — the active-recall agent. Instead of you reading and *feeling* like you understood, this skill makes you explain back. It asks you to teach the concept in your own words, then catches your hand-waves. Pair with #3: discovery agent for new material, demonstrate-understanding agent before you move on.\n5. **Claude Howto — Guided Claude Code Learning Path** — a structured curriculum (not just a doc dump) for learning Claude Code and the broader agent toolchain. Use it as the *template* for any \"how do I become competent at X\" path — module structure, exercises, progress markers. If you're learning coding through AI, start the actual learning here.\n6. **Claude Reflect — Self-Learning for Claude Code** — a reflection skill: at the end of a study session, it walks you through what you learned, what's still fuzzy, what to attack next. This is the closest open-source equivalent to a study journal that asks the right questions back. Run it nightly. Cheap habit, compounds for years.\n7. **Fabric — AI Prompt Patterns for Everything** — the 100+ prompt-pattern library. The patterns you'll live in as a learner: `extract_wisdom` (textbook chapter → clean outline), `summarize_paper`, `create_quiz` (turn a chapter into 15 questions for self-test), `analyze_claims` (when something in the material smells wrong). One CLI, consistent shape.\n8. **notebooklm-py — NotebookLM CLI + Python API Skill** — NotebookLM's killer feature for learners is **audio overviews**: upload 4 PDFs from your reading list, get a 12-minute podcast-style discussion you can listen to on a commute. The CLI lets you script that across an entire syllabus instead of clicking through the web UI.\n9. **Khoj — Your AI Second Brain** — semantic search over your own notes, PDFs, and chat history. After 6 weeks of study you'll have 200 markdown notes, 30 PDFs, 50 chat transcripts with your tutor. Khoj is how you ask \"didn't I cover Bayes' rule somewhere\" and actually find it — without forcing you into the rigid Anki review schedule (use Khoj-as-retrieval *instead of* spaced-repetition flashcards, or alongside them).\n10. **Mem0 — Memory Layer for AI Applications** — the piece that makes #1, #2, #3, #4 actually personalized. Without persistent memory, every tutor session restarts: \"what's your level? what do you already know?\" With Mem0, the tutor remembers you got stuck on closures last Tuesday and opens with that on Thursday. The single biggest unlock for solo learning.\n\n## How they fit together — the daily \u002F weekly loop\n\n```\n  Pick topic\n      │\n      ▼\n  Mr. Ranedeer  ◄────── Mem0 (remembers your level)\n      │\n      ▼  asks: what do you know?\n  DeepTutor (your textbook PDF) ──► answers grounded in source\n      │\n      ▼\n  Microsoft Study Mode ──► guided discovery (you do the thinking)\n      │\n      ▼\n  Demonstrate Understanding ──► you teach it back, agent catches gaps\n      │\n      ▼\n  Fabric: create_quiz on chapter ──► 15 self-test questions\n      │\n      ▼\n  Claude Reflect (end of session) ──► what's solid? what's fuzzy?\n      │\n      ▼\n  Khoj (next week) ──► search your notes when something feels familiar\n```\n\nThe four-tool spine is **Ranedeer + DeepTutor + Demonstrate Understanding + Mem0**: prompt to talk to, source to ground, active recall to check, memory to remember. Everything else is amplification. NotebookLM is for when you have 4 PDFs and a commute. Khoj is for month two when your notes get unwieldy.\n\n## Tradeoffs you'll hit\n\n- **Mr. Ranedeer vs DeepTutor** — Ranedeer is a *prompt* (zero infrastructure, works in any LLM, hallucinates freely). DeepTutor is a *platform* (you deploy or use the hosted version, requires you to actually upload your textbook, but cites it). Use Ranedeer for early exploration of a topic; switch to DeepTutor once you have specific source material.\n- **Khoj vs Anki (spaced repetition)** — Anki is the legendary flashcard tool with the empirically-validated review schedule. It's not in this pack because Anki is *not AI-native* — making good cards is the entire skill, and AI hasn't fully cracked that. Khoj is the AI-native alternative: instead of forcing yourself to review 50 cards\u002Fday, you search semantically when curious. Both work. Anki is more rigorous; Khoj is more humane. If you've tried Anki before and bounced, try Khoj.\n- **Microsoft Study Mode (Socratic) vs just asking ChatGPT** — asking ChatGPT for the answer is fast and you learn nothing. Socratic agents are slow and you learn something. Use the Socratic pattern for the *core concept*; use plain chat for the *syntax lookup*.\n- **Reflection skill vs writing a journal yourself** — both work, but the friction of staring at a blank journal kills most people in week two. An agent that asks \"what was the hardest part of today's session?\" gets answers a blank page doesn't.\n- **Fabric `create_quiz` vs a real practice test** — generated quizzes are great for *exposing* what you don't know (and free, and infinite). Real past papers (for certification exams especially) are the ground truth. Generate quizzes daily; do one official practice test per week.\n\n## Common pitfalls (passive reading + over-trusting the AI)\n\n- **Letting the AI write the code \u002F essay \u002F proof and calling it learning** — this is the #1 failure mode of AI-assisted study. The moment you stop typing answers yourself, learning stops. The Demonstrate Understanding agent (#4) is in the pack specifically to enforce \"you say it first, I'll critique.\"\n- **Skipping the active recall step** — re-reading feels like studying and isn't. Generate a quiz with Fabric `create_quiz` *before* re-reading the chapter — if you score 12\u002F15 cold, you didn't need to re-read. If you score 4\u002F15, you know exactly where to focus.\n- **Hallucinated facts in a domain you can't yet verify** — when you're 3 days into a topic, you can't tell if the AI is wrong. Always run DeepTutor or NotebookLM on a real source you trust for *factual* claims; use the chat tutor for *intuition* and *worked examples*.\n- **No persistent memory = no real personalization** — without Mem0 (or equivalent), the tutor treats you like a stranger every session. By week three you'll be bored. Wire memory in day one, even if you have to fake it (export the chat, paste a summary back).\n- **No reflection = no consolidation** — what you don't review in some form within 24 hours, you mostly lose. Claude Reflect at the end of every study block is the cheapest unlock in this pack. Five minutes, every night.\n- **Trying to install all ten on day one** — minimum viable kit is three: Mr. Ranedeer (paste a prompt, 30 seconds), DeepTutor (upload one PDF), and Demonstrate Understanding (one Claude Code agent). That's enough for the first two weeks. Add the rest as you hit the failure modes they fix.",[106,109,112,115,118],{"q":107,"a":108},"I'm learning to code from scratch — does this pack assume I'm already a developer?","No. Mr. Ranedeer, DeepTutor, Microsoft Study Mode, and NotebookLM all run as a chat — paste a prompt, upload a PDF, ask a question. The only picks that assume some terminal familiarity are Fabric (CLI, but `pip install` and you're done) and Khoj (a local app you run). Even Claude Reflect and the Claude Code agents work as chat-style interactions inside Claude Code's UI. If you can install Claude Code or use Claude on the web, you can use 8 out of 10 picks in this pack.",{"q":110,"a":111},"Why no Anki \u002F spaced repetition flashcard app in here?","Anki is the empirically-validated gold standard for raw retention, and it's an honest omission. The reason it's not in the pack: making *good* Anki cards is a learnable skill that takes weeks, AI-generated cards are still mediocre (they over-test trivia and under-test concepts), and most adults who try Anki bounce in three weeks. This pack picks AI-native alternatives — `Fabric create_quiz` for active recall and Khoj for semantic recall — which are softer-edged but more sustainable for the typical self-learner. If you already love Anki, keep using it; this pack complements rather than replaces it.",{"q":113,"a":114},"Will this work for learning a human language (Spanish, Japanese) or just coding?","It works for languages with adjustments. Mr. Ranedeer is great for conversational practice (\"correct me as I speak\"). DeepTutor + a grammar reference PDF beats most language-learning apps for grammar questions. NotebookLM audio overviews are a surprisingly good way to consume news in a target language. The gaps: no built-in voice\u002Fpronunciation tool, no native spaced-repetition vocab trainer. For those, pair the pack with a TTS service (ElevenLabs \u002F native OS voice) and either Anki or the language-app of your choice. The tutor + memory + reflection loop carries over cleanly; the recall layer needs domain-specific help.",{"q":116,"a":117},"How is this different from the Teacher \u002F Educator's AI Lesson Kit?","Teacher pack assumes you have 130 students and a syllabus, and the work is *delivery*: plan lessons, build slides, host an LMS, write quizzes that 30 people will take. This pack assumes you are one adult with a topic and zero classroom infrastructure, and the work is *self-management*: stay on topic, do active recall instead of passive reading, remember last week's lesson without a teacher to remind you. There's overlap (Mr. Ranedeer, DeepTutor, Mem0, Fabric, NotebookLM are in both — they're general-purpose AI tutoring infrastructure). The teacher pack adds Moodle \u002F Canvas \u002F Heyform \u002F Marp for delivering to others; this pack adds Demonstrate Understanding \u002F Claude Howto \u002F Claude Reflect \u002F Khoj for managing your own study.",{"q":119,"a":120},"What's the smallest version of this pack I can run tonight?","Three picks, about 90 minutes. (1) Paste the Mr. Ranedeer prompt into Claude or ChatGPT and have a 30-minute tutor session on something you've been meaning to learn. (2) Install Claude Code if you haven't, then enable the Demonstrate Understanding agent and try to teach the concept back to it — notice every time it catches a hand-wave. (3) End the session by running the Claude Reflect skill (or just journaling the same three questions: what did I learn, what's still unclear, what's next). One evening. Add DeepTutor + NotebookLM next weekend when you have a real PDF to upload. The other five layer in over weeks as you outgrow the basics.",{"@context":122,"@type":123,"name":13,"description":124,"numberOfItems":125,"inLanguage":25},"https:\u002F\u002Fschema.org","ItemList","Ten AI tools for the adult learner using AI to study on their own — covering personalized tutor prompts, document-grounded answers, Socratic and active-recall agents, structured learning paths, reflection, semantic note search, and persistent memory.",10,[127,131,135],{"url":128,"anchor":129,"reason":130},"\u002Fen\u002Ftopics\u002Fteacher-educator-ai-lesson-kit","Teacher \u002F Educator's AI Lesson Kit","The companion pack for instructors — same infrastructure, different role",{"url":132,"anchor":133,"reason":134},"\u002Fen\u002Ftopics","Browse other topic packs","Discover packs for AI agents, MCP servers, coding toolchains, and more",{"url":136,"anchor":137,"reason":138},"\u002Fen\u002Fai-tools-for\u002Fwriting","AI writing tools for learners","Note-taking, summarization, and reflection writing overlap with the AI writing catalog",[140,144,148],{"claim":141,"source_name":142,"source_url":143},"Mr. Ranedeer is a GPT-4 system prompt that adapts to learner level, language, and style","Mr. Ranedeer AI Tutor on GitHub","https:\u002F\u002Fgithub.com\u002FJushBJJ\u002FMr.-Ranedeer-AI-Tutor",{"claim":145,"source_name":146,"source_url":147},"NotebookLM grounds answers in user-uploaded sources with inline citations and offers an audio-overview feature","Google NotebookLM","https:\u002F\u002Fnotebooklm.google.com\u002F",{"claim":149,"source_name":150,"source_url":151},"Khoj is an open-source AI second brain that performs semantic search across personal notes, documents, and conversations","Khoj on GitHub","https:\u002F\u002Fgithub.com\u002Fkhoj-ai\u002Fkhoj",950,"2026-05-22T16:00:00Z"]