152 guides· 16 topics· updated Nov 2026· free· no sign-up
Skip the docs.
Paste the command.
Short, numbered guides for Claude Code, ChatGPT, MCP and local models. Each one ends in something you can copy and run.
each dot is a guide·hover a node, click to open
01 Topics
The whole manual, in 16 topics
Every plate has an address: 05.02 is plate 02 in topic 05, MCP. Pick the topic you're working in.
- 01 Install & set up Get Claude Code installed, updated, and unstuck — Windows, macOS, Linux, VS Code. 7 plates
- 02 First steps Your first session, plan mode, the shortcuts that matter, and Gemini CLI. 4 plates
- 03 Claude Code commands Hooks, subagents, slash commands, CLAUDE.md, permissions, context, and costs. 14 plates
- 04 Workflows Reviewing PRs, working across a whole codebase, giving an agent the right context. 7 plates
- 05 MCP Install servers, connect GitHub, and the best servers to start with. 9 plates
- 06 Claude Projects and Artifacts in the Claude app. 2 plates
- 07 ChatGPT Custom GPTs, Projects, custom instructions, voice mode, and long PDFs. 6 plates
- 08 Cursor AI coding inside the Cursor editor. 4 plates
- 09 Local LLMs Ollama and LM Studio: run a model locally, pick a quantization, call the API. 19 plates
- 10 Prompting Better prompts, system prompts, chaining, RAG, and fewer hallucinations. 9 plates
- 11 Choosing tools Claude vs ChatGPT, Claude Code vs Cursor, which model, which API. 7 plates
- 12 AI media Generate and edit images, video and audio locally — FLUX, Midjourney, inpainting, background removal, upscaling, text to speech, transcription, OCR. 14 plates
- 13 Everyday tasks Practical jobs: spreadsheets, transcripts, summaries, data analysis with AI. 4 plates
- 14 Other coding agents Codex CLI, Aider, Copilot CLI, Windsurf — the first session with each. 11 plates
- 15 Building with the API Claude and OpenAI APIs: first request, tool use, structured output, caching, batch, streaming. 23 plates
- 16 Hugging Face The Hub in practice: download and cache models, gated repos, GGUF with Ollama, transformers, Inference Providers, datasets, uploads, Spaces. 12 plates
02 Start here
Start here — six plates, in order
Install, run your first session, give the agent memory, then learn to prompt and pick a model.
≈ 19 min total
- 01.01 Install Claude Code on Windows: Download, Setup, Run 3 min
- 01.02 How to Install Claude Code on macOS & Linux 3 min
- 02.01 Your First Claude Code Session in 5 Minutes 3 min
- 03.01 How to Set Up a CLAUDE.md That Actually Helps 4 min
- 10.01 How to Write Better AI Prompts (A Framework) 3 min
- 11.01 Which AI Model Should You Use? 3 min
03 Paths
Three more sequences
Ordered runs through one subject. Start at the first plate; every guide in a path links to the next one.
- 03.01 Make Claude Code yours Memory, slash commands, hooks, subagents, permissions, and keeping the bill in check. 6 steps → 03.06
- 09.01 Run models locally Ollama on Windows, LM Studio, the right quantization, the local API, and your own Modelfile. 5 steps → 09.02
- 10.01 Prompt like a pro Better prompts, a real system prompt, chaining, RAG, and fewer hallucinations. 5 steps → 10.03
04 Tools
Builders that write the config for you
Fill in a form, copy the file. Everything runs in your browser — nothing is uploaded.
- tool MCP config generator Pick servers, fill in commands, env and headers — get a valid .mcp.json, claude_desktop_config.json or `claude mcp add` commands. .mcp.json
- tool Claude Code settings & hooks builder Compose a settings.json: permission allow/deny rules, env, model, and hooks (PreToolUse, PostToolUse, Stop…) with a live preview. .claude/settings.json (project) or ~/.claude/settings.json (user)
- tool Ollama Modelfile builder Base model, system prompt, temperature, context size, stop sequences — get a Modelfile and the `ollama create` command. Modelfile
05 Latest plates
Newest first
- 05.09 Connect the guides-ai MCP Server to Your Agent Add the guides-ai MCP server to Claude Code, Cursor, or any agent with one URL. No install, no token. Then search the whole guide library from chat. MCP 3 min
- 16.12 Fix stuck or incomplete Hugging Face downloads Resume a stuck hf download, clear broken .incomplete partials with hf cache prune, and speed transfers with hf_xet when disk space allows. Hugging Face 3 min
- 15.23 Call functions (tools) with the OpenAI API Give the model your own functions with the OpenAI Responses API: define a tool, run the call it returns, and send the result back in a loop. Building with the API 3 min
- 15.22 Get structured JSON output from the Gemini API Make the Gemini API return JSON that matches a Pydantic schema with the google-genai SDK, then parse it into a typed object. Building with the API 3 min
- 15.21 Use Gemini API File Search (managed RAG) Create a managed File Search store with the google-genai SDK, import a file, query it with the file_search tool, and read citations. Building with the API 3 min
- 15.20 Count tokens and estimate API cost before you send Count tokens locally with tiktoken or Anthropic's count_tokens endpoint, then estimate cost from each provider's published per-token prices. Building with the API 3 min
06 Method
What this is, and what it isn't
-
One question, one plate. Every guide answers a single question and ends in a command or a concrete step you can copy. Each one carries a number — its topic, then its place inside that topic — so you can find it twice, or point someone else straight at it.
-
Two minutes, not twenty. Every plate states its reading time and most take two or three. No video, no account, no newsletter gate: press ⌘K, take the command, get back to work.
-
Not the documentation. This is a field manual, not a reference — the vendor's own docs stay the authority, and these tools change under all of us. So commands are written out in full rather than hidden behind a downloaded script, and a plate says when a step is specific to your platform or touches files on disk.
07 Questions
Questions people ask
Are these AI guides free?
Yes. Every guide on guides-ai.com is free to read, with no account, paywall, or sign-up. Some pages may contain affiliate links to tools we recommend, which never change the price you pay.
Do I need to know how to code?
No. The guides are written for beginners and busy people. Each one gives you the exact command or steps to copy and paste, with a short explanation of what it does — so you can get a result first and learn the details later.
Which AI tools do you cover?
Claude Code, ChatGPT, Claude Projects and Artifacts, Cursor, Gemini CLI, the Model Context Protocol (MCP), local models with Ollama and LM Studio, and AI image tools — plus practical topics like prompting, RAG, and picking the right model or API.
Are the commands safe to copy and paste?
Every command is written out in full, so you can read exactly what it does before you run it — nothing is hidden behind a script you have to download first. Each guide states which platform the steps apply to where it matters, and says up front when a step changes files or settings.
Do the guides work on both Windows and macOS?
Where a tool runs on both, we cover both — for example there are separate Claude Code install guides for Windows and for macOS & Linux. Where the steps differ by platform, the guide says which one it is written for.
How often are new guides added?
New guides are added as tools change and new ones appear. The full, always-current list lives on the guides page, and a machine-readable index is published at /llms.txt for AI assistants.