## Power Session: Learn To Build Your Own Claude Code Skill

Hosted by Sol Farahmand

Mon, Aug 10, 2026  
1:00 PM EDT (30 minutes)  
Virtual (Zoom)  
Free to join

---

### Go deeper with a course

Become a Base44 Pro Builder: From Idea to Secure, Client-Ready App

.png&w=1536&q=75)

Sol Farahmand and Shay Bar  
AI Hackathon Winner | 2X Entrepreneur | AI Workflow Builder. AI builder and educator helping teams turn ideas into secure, practical apps.

[View syllabus](/content/sol-farahmand-ai/become-a-base44-pro-builder-from-idea-to-secure-client-ready-app/index.html)

---

### What you'll learn

- Set Up Your First SKILL.md File
  - Learn to build the folder and SKILL.md file a skill needs, and what goes in the frontmatter versus the body.

- Pick a Workflow Worth Turning Into a Skill
  - Learn to spot which of your repeated workflows fit a skill, and which ones are too open-ended to be worth it.

- Write a Description That Actually Triggers
  - Learn to write the description field with real trigger phrases so Claude Code loads your skill when you need it.

- Build, Install, and Test a Commit Message Skill
  - Learn to build a working commit-message-writer skill end to end, install it, test it, and fix what does not trigger.

---

### Why this topic matters

Don't treat Claude Code like a chatbot, re-explaining the same workflow every session and getting a different answer each time. A skill is a markdown file that loads automatically and runs your workflow the same way every time. Skip it and you keep paying tokens to re-teach your own process. Skills follow an open standard, so the same file works in Copilot, Cursor, and Gemini CLI.

### You'll learn from

.png&w=1536&q=75)

**Sol Farahmand**  
AI Hackathon Winner | 2X Entrepreneur | AI Workflow Builder  
Hi, my name is Sol, most of my work is hands-on: real code, reusable Claude Skills, and agent workflows I run in my own businesses before I teach them to anyone. I came into this from eight years of running construction and retail companies, which means I care less about benchmark scores and more about whether the thing actually holds up when a real person uses it on a Tuesday.

I run AI Enabled Academy, where I teach consultants, operators, and technical folks to build with Claude Code, Claude Cowork, ChatGPT Work, Codex, and multi-Agents with LangChain, LangGraph, CrewAI, agent-framework, and MindStudio. I also discuss Agent loops and system design: planner, executor, and critic patterns, plus the finish-line definitions that keep a loop from running away or stalling.