Mindket Learn
Practical guides, clear explanations and workflows you can use today.
How AI models work, how to learn with them and spot their fakes, how to prompt them, and which AI tool to pick for each job.
Use ChatGPT, Claude, Gemini and NotebookLM for everyday work, write with AI in your own voice and connect it to your apps.
Make images, designs, videos, voices and music from a prompt, then build a working app by describing it.
From your first Claude Code session to skills, settings, CI and the Agent SDK.
Choose an open model, run it on your own computer, find and share models on Hugging Face, then fine-tune one on your data.
Write MCP servers, call the Claude and OpenAI APIs with tools and give an agent your own data.
Go past the basics: examples, chains and templates, prompts for reasoning models, and prompts that AI helps you improve and test.
Engineer what an agent sees, the harness around it, its defences against prompt injection and the evals that show whether it works.
Build n8n workflows that handle errors and scale, build AI apps in Dify and hand browser tasks to an AI agent.
Make or edit pictures from words
Turn a prompt or a picture into a clip
Read text aloud, transcribe, clone or change a voice
Songs, music beds, sound effects and cleanup
Answers from the web or your own files, with sources
Hand off a task, or build an agent of your own
Set up Claude Cowork step by step: install the app, add a plugin and a skill, turn on computer use and hand Claude a first task you can check.
Study mode in ChatGPT, the Learning style in Claude and Guided Learning in Gemini make an AI ask and hint before it answers. See how each works.
How to tell if an image is AI-generated: five checks to run today, what a reverse image search finds, and why the old tells such as hands no longer settle it.
Learn how to draft with AI: start from your own notes and a writing sample, agree an outline, write one section at a time and check every fact yourself.
A free AI basics course for beginners: four lessons on how large language models work, which AI to choose, AI hallucinations and AI privacy. No code.
A free course on learning with AI: use study modes in ChatGPT, Claude and Gemini, write tutor prompts, make flashcards and quizzes, and practice a language.
A free course on AI fakes: check a suspicious image, handle a deepfake call, read Content Credentials and SynthID, and avoid AI scams.
A free beginner guide to picking AI tools by job: image, video, voice, music, research and agent platforms compared on price, free plans and limits.
A free prompt engineering course for beginners: four lessons on prompting basics, system prompts, context windows and AI agents, from vendor guides.
A free advanced prompt engineering course in four lessons: few-shot examples, chain of thought, prompt chaining and reusable prompt templates.
Free course on reasoning models: how they think before they answer, how to prompt them, how to set effort on OpenAI, Claude and Gemini, and where they pay off.
A free prompt optimization course in four lessons: generate a first prompt with AI, improve it, test the versions on a small test set and tune it with DSPy.
A free AI course for office workers: write emails, research topics, summarize documents and analyze spreadsheets with ChatGPT, Claude, Gemini or Copilot.
A free beginner course on AI assistants: use ChatGPT, Claude Projects, NotebookLM and Gemini for everyday work, with the menus from each help center.
A free course on writing with AI: draft from your own notes, edit without losing your voice, spot the signs of AI writing and see what AI detectors can show.
A free course on AI connectors: link ChatGPT and Claude to Gmail and Drive, add your own MCP server and check what an app can reach before you connect it.
A free course on AI images and video: write image prompts, edit photos with Nano Banana and ChatGPT, make clips in Gemini and Flow, and try ComfyUI.
A free AI design course for beginners: make a logo, get readable text into images, keep one brand look across social graphics and build product and app mockups.
A free AI video course in four lessons: what ended with Sora, a first Veo clip, how to write a video prompt, and how to run open models like Wan2.2 yourself.
A free AI audio course for beginners: text to speech, voice cloning with consent, making a song and dubbing a video, with each model's license spelled out.
A free vibe coding course for beginners: build and publish an app by describing it, with Lovable, Cursor and GitHub Copilot, and check what the AI made.
A free Claude Code tutorial for beginners in four lessons: install it, plan before it edits, use and write slash commands, and add hooks, all from the docs.
A free four-lesson Agent Skills course: write a description that triggers, split a skill into scripts and references, test it, then install it in Claude Code.
A free advanced Claude Code course in four lessons: settings.json permissions, headless mode, GitHub Actions and the Agent SDK, taken from the official docs.
A free course on open-weight models: what they are, how Apache 2.0, MIT and Llama licenses differ, how to choose one by size and context, and quantization.
A local AI course in four lessons: install Ollama, call its API from code, run models in LM Studio and work out the memory and VRAM a local LLM needs.
A free Hugging Face course in four lessons: find and download open models, run them with Transformers, call hosted models and share a Gradio demo.
Learn fine-tuning in four free lessons: decide if you need it, build a dataset, train a LoRA adapter with TRL, then run the model in Ollama.
A free n8n course for beginners in four lessons: build your first workflow, start one from a webhook, build an AI agent and self-host n8n with Docker.
A free n8n advanced course in four lessons: error workflows, sub-workflows, the Code node and queue mode with Redis and workers, for people who know the basics.
A free Dify tutorial in four lessons: run Dify on Cloud or Docker, build workflows, add a knowledge base, then build an agent and publish it.
A free five-lesson course on browser agents: ChatGPT agent, the computer use API, prompt injection and Claude Cowork with computer use and skills.
Learn MCP in four free lessons: build an MCP server in Python or TypeScript, test it with MCP Inspector, then connect it to Claude Desktop, Cursor and VS Code.
A Claude API course for developers in four lessons: your first Messages API call, tool use, prompt caching and an AI agent loop you write from scratch.
An OpenAI API course for developers in four lessons: the Responses API, function calling, Structured Outputs and an agent built with the OpenAI Agents SDK.
A free RAG course for developers in four lessons: build a pipeline with citations, learn embeddings and pgvector, chunk documents well and evaluate with Ragas.
A free context engineering course in four lessons: what an agent should see, why long context rots, how agents remember and how to load context just in time.
A free harness engineering course in four lessons: what an agent harness is, how to write tools for agents, agents that run for hours and their guardrails.
A free AI security course in four lessons: prompt injection, jailbreaks and system prompt leaks, the OWASP Top 10 for LLM apps and red teaming your own app.
A free AI evals course in four lessons: build an eval set from real failures, grade with an LLM judge, evaluate agents and run evals in CI on every change.
147
lessons in 36 free courses
588
quiz questions to check yourself
1,332
sources cited, each with the day it was read
251
posts and videos from the people who make and use the tools
Mindket Learn is a set of free AI courses for beginners and for builders. Its hands-on lessons cover AI assistants, writing, design, video and voice, Claude Code, open models, MCP, the Claude and OpenAI APIs, RAG, AI security, n8n and Dify. Steps and code come from the vendors' own docs. Every lesson is in English and Chinese, ends with a short quiz and needs no sign-up.
Yes. Every course and lesson is free to read, with no account and no sign-up. You only need an account to get or sell listings in the store.
New to AI? Take the AI fundamentals path: AI basics, learning with AI, spotting AI fakes, then prompt engineering. For everyday office work, take Work faster with AI. To make pictures, designs, videos, voices or an app without code, take Create with AI. If you already code, start the Code with AI agents path with the Claude Code course. To automate work, start with the n8n course.
Not for most of them. The AI fundamentals, Work faster with AI and Create with AI paths need no code, and nor does the first n8n course. Local AI, Hugging Face, fine-tuning, Claude Code, agent skills, MCP, the two API courses, RAG and the AI engineering path ask you to use a terminal and edit code.
Press Mark as done at the end of a lesson. Your browser keeps the list, so the course outline ticks the lesson and each learning path shows how far you are. Nothing is sent to Mindket, and clearing this site's data clears it.
From the vendors' own docs, specs and repositories, and from what the people who build and use the tools posted on X, YouTube and GitHub. Each lesson lists its sources at the end, with the day each was read.
Each lesson shows the day it was published and last updated, and lists its sources with the day each was read. What's new on this page lists the latest changes the lessons cover, and Coming up lists the dates vendors have announced.
No. Each lesson ends with a short Check yourself quiz instead, and the store holds skills, MCP servers and n8n templates to use next.
