Today combined the Day 3 theory (skill anatomy, progressive loading) with a full hands-on build: a real, working Claude Code skill for a client project, built from an empty folder to a tested, functioning tool. What was built: agency-infographic Brand system (reference.md, sourced from the real platanusdesign-brand skill): PlatanusDesign's actual colors (#F47D20 orange accent, #341390 indigo, etc.), Poppins/Open Sans typography, real logo usage rules, and a "card grid" layout pattern refined through iteration — solid black background, large topic-specific icons in orange, orange card headings, white body text. Outcome: A fully working, tested skill: confirmed to research, draft, gate on user confirmation, render on-brand output, and generalize correctly across unrelated topics/industries — built and debugged hands-on rather than handed over per-made.

Let's get to know each other! Comment below sharing where you are in the world, a career goal you have, and something you like to do for fun. 😊

Goal: Connect the Firecrawl MCP server to Claude Code and use it for something real: pull structured data off the web into a clean CSV. Source: Colorado Academic Standards for Music. It's public government data, the structure is genuinely hierarchical rather than a flat list, and I already keep a similar sheet by hand. Process: Connected Firecrawl MCP, verified with /mcp, Wrote FIRECRAWL_CHEATSHEET.md — all six tools, each with when to use and when NOT to, plus a context-cost table, Set up CLAUDE.md so the project rules load automatically every session, Ran the real extraction in Plan Mode. My assumption about grade levels was wrong. Colorado's 2022 Music standards have no grades 6–12 at all — secondary uses proficiency levels instead (Novice, Intermediate, Proficient, Accomplished). Result: data/curriculum.csv — 289 rows, one per evidence outcome, seven columns. No empty cells, every outcome verified verbatim against the source, three network calls total. What it's for: Replaces a sheet I was filling in by hand, with verified official codes. Filtering by grade or standard gives a coverage list for lesson planning. And the same script runs against any of the thirteen content areas — only the source file changes. Takeaway: The most useful output wasn't the CSV. It was the cheat sheet and CLAUDE.md — they turn a one-off task into an environment that works better next time without re-explaining anything. Also worth noting: the best outcome of a scraping task was discovering that scraping wasn't the right route. The official export gave better data for less work.

I’m excited to share my Day 1 win: building my first newsletter using Claude Code. I learned a ton through the process—how Claude Code works, what goes into creating a newsletter from scratch, and where there’s room for improvement. Since this was my first time building something like this, it was a great hands-on experience. My workflow actually looks like this: · Style guide as artifact · Claude Code builds on it · MCP writes directly to WordPress — no manual export/import steps · Elementor fetches for further editing

This event has been amazing. There's not much that energizes me more than being in a room of a bunch of people who are all working hard to grow! I also got to meet a TON of you guys which makes me so happy. I didn't get a picture with everyone, but here are a few. You can even see @Duy Nguyen repping AIS. What a guy!! Montenegro is a beautiful country. Highly recommend adding it to your travel bucket list. Hope you guys are getting pumped for our first AIS in person event!! - Nate