ONLINEAGENT_OPS 2026.Q3 HOME HUBSMODELSBUILDINGCOMPANIESFILMMAKINGSECURITYLIVING WITH ITSKILLSALL HUBS, ONE TERMINAL → ARTICLES CRAFT RECORD BLOG TOOLS SKILLS SEARCH /
HOME›ARTICLES›WHAT IS A SKILL IN AI?
ARTICLES · EVERGREEN EXPLAINERBEGINNER

What Is a Skill in AI? AI Skills Explained Simply

A skill is a folder of instructions an AI reads when a task calls for it. A tool is an action it can ask software to carry out. MCP is the open plug that connects both to real apps.

READ14 min
WORDS2,828
SECTIONS10
TYPEEXPLAINER
CHECKED1 OCT 26

A skill in AI is a folder of written instructions that an AI assistant loads when a task calls for it, so it does that job the same way every time. Anthropic, which created the format, defines skills as “folders of instructions, scripts, and resources that Claude loads dynamically to improve performance on specialized tasks.”1 A prompt is for one conversation; a skill is saved for every time the same job comes up.

TL;DR — THE SHORT VERSION

A skill in AI is a folder of written instructions that an AI assistant reads when a task calls for it, so it does that job your way every time. A prompt is what you type for one task; a skill is saved know-how the assistant picks up by itself whenever it is needed.

  • What is inside: a file called SKILL.md with a name, a short description and step-by-step instructions, plus optional reference files and small programs (scripts).
  • How it works: the assistant first sees only each skill’s name and description. It reads the full instructions only when your request matches.
  • Skill vs prompt: a prompt lasts for one conversation. A skill is reusable and loads on demand, so you stop repeating the same guidance.
  • Who uses them: the format was first developed by Anthropic for Claude and is now an open standard; OpenAI’s ChatGPT and Codex, GitHub Copilot, Google’s Gemini CLI and Cursor document support for it.
  • Tools and MCP are different things: a tool is an action the AI asks software to carry out; MCP is the plug that connects an AI to apps. A skill teaches it how to use them well.
  • The main risk: a skill is instructions and sometimes code. Install skills only from sources you trust, and read them first.

Looking for AI skills for your career? See AI skills for work. This page is about a different meaning: skills as a feature inside AI assistants.

Agents now come with “skills”, assistants “use tools” and everything “connects”. This page explains what those words actually mean, in plain English, with the vendors’ own documentation linked for every claim.

For installing skills, comparing them with MCP and plugins, and skills for video, see the Skills section.

What a skill is, and the problem it solves

A skill is a written method the AI can pick up when it needs it. AI assistants are good improvisers with no habits: ask twice and you may get two different methods. For real work you want the same good method every time, such as your company’s report format or the house style. A skill is that method written down and packaged.

Anthropic, which makes Claude and created the format, defines skills in one line: “Skills are folders of instructions, scripts, and resources that Claude loads dynamically to improve performance on specialized tasks.”1 At its heart, a skill is a folder holding one required file, SKILL.md. That file starts with a name and a description, then gives the instructions. The folder can also hold scripts (small programs), reference documents and templates.2

◈ IN PLAIN TERMS

Anthropic compares building a skill to writing an onboarding guide for a new member of staff.3 Another way to picture it: laminated recipe cards for a talented chef with no memory. The chef only pulls out the card for the dish that has been ordered.

It is not just a Claude feature any more. Anthropic published the format as an open standard, meaning anyone can build tools that read it.4 OpenAI says skills work in ChatGPT and Codex and build on that standard.5 GitHub describes skills for its Copilot assistant in almost the same words: “folders of instructions, scripts, and resources that Copilot can load when relevant”.6

Skill vs prompt vs agent: what is the difference?

A prompt is for this conversation. A skill is for every time the same job comes up. Anthropic’s documentation draws the line directly: unlike prompts, skills “load on demand, so you don’t have to repeat the same guidance across conversations.”7 The comparison: a prompt is telling a colleague what you want today; a skill is the written procedure they look up whenever that kind of task lands on their desk.

ThingWhat it isWhen the AI uses it
PromptWhat it is: the message you type for one task.Used: once, in the conversation where you typed it.
Custom instructionsWhat it is: standing preferences you set once in the app.Used: in every conversation, whatever the task.
SkillWhat it is: a saved folder of instructions for one kind of job.Used: only when your request matches that job.
Tool or MCP connectorWhat it is: an action the AI can ask software to carry out, or the plug to an app.Used: when the task needs an action, such as a search or a calendar entry.
AgentWhat it is: an AI system that decides its own steps, and which tools to use, to reach a goal.Used: it is the one doing the work; along the way it reads skills and calls tools.

Anthropic’s help page puts the custom-instructions difference the same way: custom instructions “apply broadly to all your conversations”, while skills “only load when relevant”.1 Prompt, context, skill or fine-tune helps you choose between them, and prompt meaning covers the basics. An agent is the worker rather than the instructions: Anthropic describes agents as systems that direct their own process and tool use.8 So a prompt is a one-off request, a skill is a saved method, and an agent is the AI that carries out the work, using skills when they match.

How skills work: read only when needed

Loading a skill just means the AI reads it. Nothing magic happens. The assistant keeps a short list of the skills it has: just each one’s name and description. When your request matches a description, it reads that skill’s full instructions and follows them. Extra files are opened only if the instructions point to them. Anthropic calls this “progressive disclosure”.7 OpenAI describes the same order: ChatGPT and Codex “start with each skill’s name and description”, then load the full instructions when they use it.5

Why bother? An AI can only hold so much text in mind at once (its “context window”; see context window meaning). Reading skills only when needed means you can install many without crowding it. It also explains why the description matters so much: Anthropic says it is what the AI matches your request against, so it must say what the skill does and when to use it.7

Because the AI simply reads the instructions, a skill is only as good as its writing: precise, in order, with worked examples. Skills are prompt engineering that got a filing cabinet.

SKILLS, TOOLS, CONNECTORS · HOW THEY FIT
A skill is text loaded into the model’s context. A tool call is a request the surrounding software carries out, with the result handed back. A connector is how that software reaches an app.
Skillpackaged know-howModelemits a requestSoftwareexecutes the callAppMCP serverLOADEDTOOL CALLRESULTCONNECTORSKILL · THE RECIPE CARDTOOL · THE HANDSCONNECTOR · THE SOCKET
Reasoning — summarises this page’s sections on what a skill is, how skills load, tools and connectors, page checked 1 Oct 2026.

What is inside a skill (SKILL.md)

The plain point: a skill is a short text file with a label on top and instructions underneath. Anthropic’s documentation sets out the format: every skill needs a SKILL.md file that opens with a name and a description, followed by the instructions.9 Here is a made-up example in that format:

---
name: weekly-report
description: Writes the team's weekly status report in our house
  format. Use when someone asks for a weekly report, a status
  update or an end-of-week summary.
---

# Weekly report

## Instructions
1. If this week's wins, problems or next steps are missing, ask for them.
2. Write three short sections: Done, Blocked, Next.
3. Keep it under 200 words. No jargon.

## Examples
See examples/good-report.md for a report the team liked.

What each part does:

  • The lines between the two --- markers are the label (called “frontmatter”). For Claude, the name can be at most 64 characters, using only lowercase letters, numbers and hyphens, and the description at most 1,024 characters.9
  • The description is the trigger. Until the skill is needed, it is all the AI sees, which Anthropic puts at about 100 tokens per skill; it must say what the skill does and when to use it.9
  • The instructions underneath are read only when a request matches, and Anthropic’s guide puts them at under 5,000 tokens.9
  • Extra files (examples, reference notes, small scripts) sit in the same folder. The AI opens them only if the instructions point to them, and when it runs a script only the result is read back, not the code.9

Other apps that follow the open standard use the same two required fields; Cursor’s documentation, for example, lists name and description, with the same lowercase-and-hyphens rule for the name.10

TAKEAWAY

A skill is a labelled set of instructions. Spend most of the effort on the description, because it decides when the skill is used, and on clear, numbered steps.

Which AI apps support skills?

The plain point: several assistants now read the same SKILL.md format, because Anthropic published it as an open standard on 18 December 2025.11 Other makers added support within weeks, according to their own changelogs. The table lists only apps whose own documentation this site read (5 Oct 2026, with start dates added from their changelogs on 13 Oct 2026). The standard’s own list of apps that read it held 46 products on 13 Oct 2026, so others support skills too.12

AppMade byWhere skills work, per its own docs, and since when
ClaudeMade by: AnthropicWhere: the Claude apps (claude.ai), Claude Code and the Claude API, plus ready-made skills for Word, Excel, PowerPoint and PDF files.9 Since: 16 Oct 2025.13
ChatGPT and CodexMade by: OpenAIWhere: ChatGPT on the web, desktop and mobile, Codex in the ChatGPT desktop app, and the Codex command-line tool.14 Since: Codex from 19 Dec 2025.15
GitHub CopilotMade by: GitHubWhere: Copilot’s cloud agent, code review, command-line tool and app, and agent mode in VS Code and JetBrains editors.1617 Since: 18 Dec 2025, the day the standard was published; VS Code followed in its 8 Jan 2026 release.1819
Gemini CLIMade by: GoogleWhere: Google’s command-line coding assistant, which asks you to confirm before a skill is switched on.20 Since: January 2026 (versions 0.24 to 0.26).21
CursorMade by: CursorWhere: the Cursor code editor’s agent, which finds skills automatically when it starts.10 Since: 22 Jan 2026 (Cursor 2.4).22

Which Claude plans include skills? Claude’s help centre now lists all of them, Free included: “Skills are available for users on Free, Pro, Max, Team, and Enterprise plans.”23 At launch they were for paid plans only, and this site found no date for when Free was added.13 To use them, switch on code execution in Settings > Capabilities, then go to Customize > Skills to turn skills on or upload your own.23 For step-by-step installs in each app, see how to install AI skills.

TAKEAWAY

One skill folder can work in more than one app, because they follow the same open format. Check each app’s own documentation for where it looks for skills and whether it asks before using one.

Tools: the hands

A tool is an action the AI is allowed to ask for. Search the web, look something up in a database, add a calendar entry, run code. The AI does not do these things itself. It sends a structured request, the software around it carries the request out, and the result comes back for the AI to read. Anthropic’s documentation describes that round trip: Claude “returns a structured call that your application executes”.24 This loop (request, carry out, return) is the mechanism behind every “AI that takes actions”. Tool use meaning has a plain-English version.

The craft is in the tool’s description. Anthropic says a good one explains what the tool does and when it should, and should not, be used.25

TAKEAWAY

An AI that “takes actions” is asking software to act on its behalf. How well that goes depends heavily on how clearly each tool is described, including when not to use it.

Connectors and MCP: the sockets

MCP is a shared plug between AI assistants and apps. For an AI to reach your calendar, documents or code, something has to translate between “the AI wants this” and each app. The Model Context Protocol (MCP) is an open standard for that: an app offers an MCP server, and any assistant that speaks MCP can plug in and use its tools. Its own documentation says: “Think of MCP like a USB-C port for AI applications.”26 MCP meaning explains it further.

Skills and MCP work together rather than compete. Anthropic sums it up: MCP gives access, skills give know-how.1 Skills vs MCP vs plugins puts all of them side by side, with plugins, subagents and Projects.

The honest risks

A skill can carry bad instructions as easily as good ones. Skills and tools inherit a basic weakness of AI models: text the AI reads can steer it. A malicious document or web page can try to slip in new orders. This is called prompt injection, and an AI with tools has real “hands” to misuse.27

Anthropic’s own advice is to treat a skill like software you install.7 In practice:

  • Read a skill before you install it, including any scripts. Anthropic’s checklist says to look for network calls or file access that do not match what the skill claims to do.7
  • Keep a human check on anything consequential, such as sending, paying or deleting. OWASP recommends “human-in-the-loop controls for privileged operations”, meaning a person approves the risky steps.27
  • Be wary of skill marketplaces. Skills are a supply chain covers a security firm’s audit of thousands of public skills.
TAKEAWAY

Treat a third-party skill like third-party software and check its source before installing it. Once an agent has tools, keep a human confirmation step on anything consequential it can do.

Why this matters if you do not write code

Writing a good skill is mostly writing a good procedure. The instructions are plain text that the AI reads, and Anthropic likens the whole folder to an onboarding guide for a new team member.7 In the Claude app, Anthropic lists everyday uses such as applying brand guidelines to documents and structuring notes in a company format.28

This site’s view: people who can write a clear procedure (teachers, office managers, anyone senior in a job) already have the main ability that skill-writing needs. The hard part is not code. It is knowing how the work should be done, and writing it down well. Skills, repos and MCP explains all three words without assuming you write code, and the glossary holds the rest of the vocabulary.

TAKEAWAY

If you can write a clear procedure for how a job should be done, you already have the main ability that skill authoring needs. Start with the writing, not the code.

Questions people ask

Do I have to turn a skill on?

It depends on the app. In the Claude apps you switch on code execution in Settings, then turn individual skills on or off under Customize > Skills; after that Claude picks them up by itself when a request matches.29 ChatGPT and Codex can choose a skill automatically or let you call one by name, and Gemini CLI asks you to confirm when it activates one (see the table above).14

Are skills safe to install?

Only as safe as their source. A skill can include code as well as instructions, and Anthropic says to use skills only from trusted sources and to treat installing one like installing software.9 Skills are a supply chain covers what security researchers found in public skill collections.

What is the difference between a skill and MCP?

MCP connects an AI to an app or a data source; a skill teaches it how to do a task, including how to use those connections well. They are designed to be used together.1 MCP meaning explains the connection side.

Can I write my own skill?

Yes. It is a folder with a SKILL.md file in it, like the example above. In Claude Code, for instance, you place the folder in a skills directory and Claude finds it without any upload.9 Skills, repos and MCP walks through it without assuming you write code.

Related on this site: Best AI Video Skills for Claude? · Check a Skill Before Installing · We Checked the Five ‘Claude Video Skills’ From a Viral Reel · What Is SkillSpector?

SOURCES
  1. 1Claude Help Center, What are Skills?, read at source 1 Oct 2026“Skills are folders of instructions, scripts, and resources that Claude loads dynamically to improve performance on specialized tasks.” ↑a“Skills are folders of instructions, scripts, and resources that Claude loads dynamically to improve performance on specialized tasks.” ↑b“Custom instructions apply broadly to all your conversations. Skills are task-specific and only load when relevant, making them better for specialized workflows.” ↑c“MCP connects Claude to external services and data sources. Skills provide procedural knowledge—instructions for how to complete specific tasks or workflows. You can use both together: MCP connections give Claude access to tools, while skills teach Claude how to use those tools effectively.” ↑d“MCP connects Claude to external services and data sources. Skills provide procedural knowledge—instructions for how to complete specific tasks or workflows.” ↑e
  2. 2Agent Skills, Agent Skills overview (agentskills.io), read at source 1 Oct 2026“At its core, a skill is a folder containing a SKILL.md file. This file includes metadata (name and description, at minimum) and instructions that tell an agent how to perform a specific task. Skills can also bundle scripts, reference materials, templates, and other resources.” ↑
  3. 3Anthropic Engineering, Equipping agents for the real world with Agent Skills, 16 Oct 2025, read at source 1 Oct 2026“Building a skill for an agent is like putting together an onboarding guide for a new hire.” ↑
  4. 4Agent Skills, Agent Skills overview, read at source 1 Oct 2026“The Agent Skills format was originally developed by Anthropic, released as an open standard, and has been adopted by a growing number of agent products.” Anthropic’s engineering post carries the update: “We’ve published Agent Skills as an open standard for cross-platform portability. (December 18, 2025)”. ↑
  5. 5OpenAI, Build skills, read at source 1 Oct 2026“A skill packages instructions, resources, and optional scripts so either product can follow a workflow reliably.” and “Skills build on the open agent skills standard.” ↑a“ChatGPT and Codex start with each skill’s name and description, then load the full SKILL.md instructions when they decide to use that skill.” ↑b
  6. 6GitHub Docs, About agent skills, read at source 1 Oct 2026agent skills are “folders of instructions, scripts, and resources that Copilot can load when relevant to improve its performance in specialized tasks.” ↑
  7. 7Anthropic, Agent Skills (Claude developer docs), read at source 1 Oct 2026“Unlike prompts (conversation-level instructions for one-off tasks), Skills load on demand, so you don’t have to repeat the same guidance across conversations.” ↑a“Claude loads information in stages as needed, rather than consuming context upfront.” and “until a Skill is triggered, only its name and description occupy context.” ↑b“The description is what Claude matches your request against when determining whether to trigger the Skill, so it must say both what the Skill does and when to use it.” ↑c“Use Skills only from trusted sources: those you created yourself or obtained from Anthropic. Skills give Claude new capabilities through instructions and code, which also means a malicious Skill can direct Claude to invoke tools or execute code in ways that don’t match the Skill’s stated purpose.” ↑d“Look for unusual patterns such as unexpected network calls, file access patterns, or operations that don’t match the Skill’s stated purpose”. ↑eskills exist as directories “containing instructions, executable code, and reference materials, organized like an onboarding guide you’d create for a new team member.” ↑f
  8. 8Anthropic, Building effective agents, 19 Dec 2024, read at source 5 Oct 2026“Agents, on the other hand, are systems where LLMs dynamically direct their own processes and tool usage, maintaining control over how they accomplish tasks.” ↑
  9. 9Anthropic, Agent Skills (Claude developer docs), read at source 5 Oct 2026“Every Skill requires a SKILL.md file with YAML frontmatter” and “Required fields: name and description”. ↑aname: “Maximum 64 characters”, “Must contain only lowercase letters, numbers, and hyphens”; description: “Must be non-empty”, “Maximum 1024 characters”. ↑bLevel 1, metadata: loaded “Always (at startup)”, token cost “~100 tokens per Skill”; “The description must include both what the Skill does and when Claude should use it.” ↑cLevel 2, instructions: loaded “When Skill is triggered”, token cost “Under 5k tokens”. ↑d“Claude accesses these files only when referenced.” and “When instructions mention executable scripts, Claude runs them through bash and receives only the output (the script code itself never enters context).” ↑eCustom Skills “are available across Claude’s products: create them in Claude Code, upload them through the Claude API, or add them in claude.ai settings.” and “Anthropic provides pre-built Agent Skills for common document tasks (PowerPoint, Excel, Word, PDF)”. ↑f“Use Skills only from trusted sources: those you created yourself or obtained from Anthropic.” and “Treat like installing software: Be especially careful when integrating Skills into production systems with access to sensitive data or critical operations”. ↑g“Custom Skills in Claude Code are filesystem-based and don’t require API uploads: place them in ~/.claude/skills/ (personal) or .claude/skills/ (project).” ↑h
  10. 10Cursor, Agent Skills (Cursor docs), read at source 5 Oct 2026name: “Skill identifier. Lowercase letters, numbers, and hyphens only.”; description: “Describes what the skill does and when to use it.” ↑a“Agent Skills is an open standard for extending AI agents with specialized capabilities.” and Cursor “automatically discovers skills from skill directories and makes them available to Agent.” ↑b
  11. 11Anthropic, Skills for organizations, partners, the ecosystem, 18 Dec 2025, read at source 13 Oct 2026“We’re also publishing Agent Skills as an open standard. Like MCP, we believe skills should be portable across tools and platforms” ↑
  12. 12Agent Skills, Client Showcase (agentskills.io), read at source 13 Oct 2026, subtitle “Agent products that support the Agent Skills format.”: 46 products listed. This is the standard’s own list, not each vendor’s claim.↑
  13. 13Anthropic, Introducing Agent Skills, 16 Oct 2025, read at source 13 Oct 2026“Claude can now use Skills to improve how it performs specific tasks.” ↑a“Skills are available to Pro, Max, Team and Enterprise users.” ↑b
  14. 14OpenAI, Build skills, read at source 5 Oct 2026“Skills work in Chat and Work across ChatGPT on the web, desktop, and mobile, in Codex in the ChatGPT desktop app, and through Codex CLI.” ↑askills can be triggered by explicit invocation (using @ in ChatGPT or $ in Codex) or implicitly, when ChatGPT or Codex selects a skill that matches the task. ↑b
  15. 15OpenAI, Codex changelog, entry “Agent skills in Codex”, 2025-12-19, read at source 13 Oct 2026“Codex now supports agent skills: reusable bundles of instructions (plus optional scripts and resources) that help Codex reliably complete specific tasks.” ↑
  16. 16GitHub Docs, About agent skills, read at source 5 Oct 2026skills work with “Copilot cloud agent, Copilot code review, the GitHub Copilot CLI, the GitHub Copilot app, and agent mode in Visual Studio Code and JetBrains IDEs”. ↑
  17. 17Visual Studio Code, Use Agent Skills in VS Code, read at source 5 Oct 2026“Agent Skills are folders of instructions, scripts, and resources that AI agents can load when relevant to perform specialized tasks.” ↑
  18. 18GitHub, GitHub Copilot now supports Agent Skills, 18 Dec 2025, read at source 13 Oct 2026“You can now create Agent Skills to teach Copilot how to perform specialized tasks in a specific, repeatable way.” ↑
  19. 19Visual Studio Code, December 2025 (version 1.108), release date 8 Jan 2026, read at source 13 Oct 2026“VS Code now supports Agent Skills, allowing you to teach the coding agent new capabilities and provide domain-specific knowledge.” ↑
  20. 20Gemini CLI, Agent Skills (Gemini CLI docs), read at source 5 Oct 2026“Agent Skills let you extend Gemini CLI with specialized expertise, procedural workflows, and task-specific resources. Based on the Agent Skills open standard, a ‘skill’ is a self-contained directory that packages instructions and assets into a discoverable capability.” and “You will see a confirmation prompt in the UI detailing the skill’s name, purpose, and the directory path it will gain access to”. ↑
  21. 21Gemini CLI, Changelogs, read at source 13 Oct 2026v0.24.0, 2026-01-14, “We’ve introduced significant advancements in Agent Skills”; v0.25.0, 2026-01-20, “enabled skills by default”; v0.26.0, 2026-01-27, “enabled agent skills by default”. ↑
  22. 22Cursor, Changelog 2.4, Jan 22, 2026, read at source 13 Oct 2026“Cursor now supports Agent Skills in the editor and CLI.” ↑
  23. 23Claude Help Center, Use skills in Claude, read at source 13 Oct 2026“Skills are available for users on Free, Pro, Max, Team, and Enterprise plans. This feature requires code execution to be enabled.” ↑a“Navigate to Settings > Capabilities and ensure that Code execution and file creation is enabled.”, then “Go to Customize > Skills.” and “Toggle individual skills on or off as needed.” ↑b
  24. 24Anthropic, Tool use with Claude (Claude developer docs), read at source 1 Oct 2026“Tool use (also called function calling) lets Claude call functions that you define or that Anthropic provides. Claude determines when to call a tool based on the user’s request and the tool’s description. It then returns a structured call that your application executes (client tools) or that Anthropic executes (server tools).” ↑
  25. 25Anthropic, Define tools (Claude developer docs), read at source 1 Oct 2026“Provide extremely detailed descriptions. This is by far the most important factor in tool performance.” Descriptions should cover “When it should be used (and when it shouldn’t)”. ↑
  26. 26Model Context Protocol, What is the Model Context Protocol (MCP)?, read at source 1 Oct 2026“MCP (Model Context Protocol) is an open-source standard for connecting AI applications to external systems.” and “Think of MCP like a USB-C port for AI applications.” ↑
  27. 27OWASP GenAI Security Project, LLM01:2025 Prompt Injection, read at source 1 Oct 2026“Indirect prompt injections occur when an LLM accepts input from external sources, such as websites or files.” and “it is unclear if there are fool-proof methods of prevention for prompt injection.” ↑a“Implement human-in-the-loop controls for privileged operations to prevent unauthorized actions.” ↑b
  28. 28Claude Help Center, What are Skills?, read at source 1 Oct 2026, example uses include applying brand guidelines to documents, generating communications from templates and structuring notes with company formats.↑
  29. 29Claude Help Center, Use skills in Claude, read at source 5 Oct 2026“This feature requires code execution to be enabled.” Steps: Settings > Capabilities, “Code execution and file creation”, then Customize > Skills to “Toggle individual skills on or off as needed.” and “You don’t need to explicitly invoke them—Claude determines when each skill is needed based on your request.” ↑

Anthropic, Claude developer docs: Agent Skills, Tool use with Claude and Define tools, read at source 1 Oct 2026.

Agent Skills, Agent Skills overview (agentskills.io), read at source 1 Oct 2026.

The comparisons on this page (the recipe cards, the colleague and the written procedure, the hands and the sockets) are this site’s own ways of explaining, not claims from these sources; the onboarding-guide and USB-C comparisons are the vendors’ own.

◈ WHERE THIS SITE STANDS — Nothing here argues against AI or its development. We are messengers: concerned humans showing information and facts about how the internet is changing, and how people might adapt. Worry is not hostility.

Part of the Stay Human record. Continue: the tools atlas → · back to repos

Free printable AI cheat sheets, and AI news every week.

Sixteen one-page AI guides and checklists in one free PDF, no sign-up. And if you want to keep up, the newsletter covers what changed in AI each week, in plain English, with the sources.

SUBSCRIBE FREE ↗ GET THE CHEAT SHEET PACK