Stop Prompting, Start Skilling

Everyone wants to talk about agents.

Autonomous AI that runs your business while you sleep. Agents that write your emails, manage your campaigns, review your SEO. The LinkedIn drivel is constant "I replaced my marketing team with 7 AI agents."

It's sexy. It sells. And it's mostly theater.

Here's what I'm seeing in the MarketingAlec community: people who go all-in on agents spend weeks configuring, debugging, and troubleshooting brittle workflows. People who build skills shipping every day.

Shipping vs. Sexy.

I run/test 30+ AI agents. They're part of my workflow. But the thing that saves me hours every week? A boring set of markdown files that tell Claude exactly how to do repeatable tasks. No autonomy. No multi-agent orchestration. Just: here's the process, follow it, give me the output. KISS.

Those instruction files are called skills. And right now, they're the single highest-leverage thing you can implement.

I'm watching my students struggle with the same question: How do you actually build these things? Not the concept, they get the concept. The execution.

Where do I put the file? How detailed should it be? Does this work on ChatGPT or only Claude?

Let's get you leveled up on skills.

Many people use AI the same way every session: open a chat window, ask for what they need, get output, and close the window. Tomorrow, repeat.

That's the equivalent of onboarding a new hire every Monday morning. You explain the brand voice. You describe the project. You list your preferences. The AI nails it. Then the session ends, and all of that context evaporates.

Next time? Use a Skill...

This is the "cold start problem," and it's the #1 frustration I see with students. The setup and maintenance cost is "the hidden bottleneck in AI productivity tools." I feel sorry for all the n8n fanboys that are now maintaining a hundred fragile workflows. You install the tool, run a few generic prompts, get a me-too output, and go back to the old ways.

The fix isn't a better prompt. It's a saved instruction set that loads automatically, so the AI already knows who you are, what you're working on, and how you like things done before you type a single word.

That's an AI skill.

Strip away the branding, and a skill is three things in a file:

1. Context who you are, what your project is, resource docs, what conventions matter.

2. Process the step-by-step workflow the AI should follow.

3. Guardrails what NOT to do, quality checks, output format.

That's it. It's not magic. It's a markdown document.

I have a skill called "meeting processor." In plain English, it does this:

I wrote it once. It runs identically every time. Last week it processed 24 meetings. This week, it will do it again. Same quality. Same speed. No prompting.

Three things happened in March that turned skills from a niche developer concept into mainstream conversation.

Garry Tan went viral. The Y Combinator CEO open-sourced his Claude Code setup; 13 skill files that simulate an entire engineering organization. Product manager persona, spec writer, task planner, coder, reviewer. All are defined as reusable instruction files. 23,000 GitHub stars. TechCrunch wrote it up. He claims 100 PRs merged in 7 days using the system.

ChatGPT launched Skills. On March 5th, OpenAI quietly rolled out a "Skills" feature for Business and Enterprise users; reusable workflows that ChatGPT auto-applies based on context. This isn't Custom Instructions. It's a formalized system for saved process definitions. The name isn't a coincidence.

The open standard emerged. The SKILL.md file format; originally created for Claude Code, is now supported by 16+ tools: Cursor, GitHub Copilot, Gemini CLI, Windsurf, Codex, and more. Write once, run everywhere. Meanwhile, AGENTS.md is becoming the cross-platform standard for agent instructions, backed by the Linux Foundation.

Skills are how you we ship. Every single day.

This is the part nobody else is writing. Every blog post, every tutorial, every bootcamp treats skills like they only exist in Claude Code. They don't. The concept translates everywhere.

Claude Code has the most mature skill system. Your master instruction file lives at CLAUDE.md in your project root. Skills live in .claude/skills/ as packaged folders with their own instruction files, scripts, and reference docs.

The powerful part: skills auto-trigger based on what you're doing. You don't call them. Claude reads the task, scans skill descriptions, and activates the right one.

Starter CLAUDE.md (copy this):

That's 10 lines. It takes 5 minutes to write and it'll improve every session from here on out.

OpenAI launched Skills in ChatGPT on March 5th. This isn't Custom Instructions with a new name. It's a proper skill system.

A ChatGPT Skill bundles instructions, examples, and executable code into a reusable package that activates automatically when relevant; across web, mobile, and API. Sound familiar? It should. It's the same concept as Claude Code's SKILL.md, just built into ChatGPT.

How to build one: Every ChatGPT account now includes a built-in skill-creator. Ask ChatGPT to "create a skill for [your workflow]" and it scaffolds the whole thing; instructions, trigger conditions, output format. You can share skills with teammates or publish them to your workspace library.

Currently in beta for Business and Enterprise users. If you're on a Plus or free plan, you can still approximate skills using Projects (workspace-scoped instructions + file uploads) or Custom GPTs (instruction packages with knowledge files and tools).

The limitation: ChatGPT Skills live in OpenAI's UI. You can't version-control them, export them to a file, or port them to another tool. The portability isn't there yet.

If you're in the Microsoft ecosystem, it’s more limited.

GitHub Copilot uses Skills.md, where it reads a .github/copilot-instructions.md file. You can go further with .github/instructions/*.instructions.md files scoped to specific file paths using applyTo globs.

The headline feature: Copilot now natively reads CLAUDE.md, GEMINI.md, and AGENTS.md files. One instruction file, multiple tools. Write it once.

Today, you cannot just drop SKILL.md into Microsoft 365 Copilot or Copilot Studio and expect it to work the way it does in GitHub Copilot; you’d have to wrap that logic into an Agent Framework/Bot Framework skill or plugin.

Gemini CLI uses a GEMINI.md file hierarchy — global instructions in ~/.gemini/GEMINI.md, project-level in your working directory, and automatic discovery as Gemini navigates your files. It also supports a SKILL.md system nearly identical to Claude Code's format.

The unique feature: @file.md import syntax. You can modularize your instructions across files and import them. Most flexible instruction architecture of any platform.

For Gemini web users: Create Custom Gems. Each Gem is a reusable instruction preset with generous character limits. One Gem per workflow type.

A single SKILL.md file now works across Claude Code, Gemini CLI, Cursor, GitHub Copilot, and ChatGPT. The format is becoming a cross-platform standard.

You're not locked into any vendor. You're building a portable skill set.

Turn AI Into Your Most Powerful Marketing Teammate

Put AI to Work Across Your Marketing Strategy

In this hands-on course, you will learn how to confidently use tools like ChatGPT, Midjourney, and Descript to plan, create, and scale content across copy, images, and video.

You will leave with practical frameworks and real experience generating marketing assets you can apply immediately.

Led by Alec Newcomb, this course is built for marketers, leaders, and business owners who want to improve efficiency, drive growth, and stay ahead with AI.

Don't overthink this. Pick the task you explain to AI most often and save those instructions.

Step 1: Identify your repeat task. Meeting notes, blog drafts, email replies, data analysis, social posts. The one where you type the same context every session.

Step 2: Write three sections. Context, Process, Guardrails.

Step 3: Save it where your tool can find it.

Step 4: Test it 3 times. If output is consistent across all three, you have gold.

Step 5: Iterate. Add a guardrail for whatever went wrong. Skills improve the same way people do, through specific feedback.

Universal template (works on any platform):