What I built: /research — buyer keyword extractor with HTML gallery output The skill searches the web via Firecrawl, scrapes the top results, and produces a self-contained HTML report you can open directly in a browser. It extracts product keywords organized by category (synonyms, product names, specs, use cases, navigation labels) + an image gallery pulled from every <img> tag on the page. Trigger: /research <topic> — or just say "research X for the UK market" and it picks it up naturally. One optimization I made: The first version returned a markdown text report. After watching it run, I realized the output was all words and no visual context — I had no feel for how the product actually looks on pages buyers visit. So I added formats: ["html"] to the Firecrawl call, extracted image URLs from <img src> tags, and rewrote the output as a styled HTML file with a product gallery. Now every report shows the actual photos customers see alongside the keywords. Second smaller one: I added a capital city rule, it automatically appends "Capital city" to the search query. If you say "German market" it appends "Berlin". This anchors results to where buyers in that country actually search from, instead of getting generic global results. Skill is ~130 lines. Built iteratively in one session by running real searches and adjusting based on what came back.

What I scraped: Extracted the structured main navigation links from vistaprint.com — 12 top-level nav items (name + URL) saved to a clean CSV in one shot. One thing I learned: Claude Code didn't just call a scraper blindly — it reached for firecrawl_scrape with a json format and a structured schema instead of pulling raw HTML and parsing it manually. It also caught that the /v1/extract endpoint was deprecated mid-task and switched approaches on its own. That kind of adaptive tool selection is what makes the WAT framework click — the agent handles the how, you just define the what. One use case idea: Competitive nav monitoring — scrape the top-level navigation of 10–20 competitor sites on a schedule to track when they add new product categories or restructure their offerings. Easy signal for market moves that would otherwise take manual checking.

Got my Claude Code newsletter project up and running. Set up the repo, explored the tools for drafting and sending emails, and realized how much is happening under the hood even on day one🙌

Hey everyone! Sharing something I use every single day when working with AI. Most people don't know that AI responds to shortcut commands — you don't need to write a whole paragraph. One word changes everything. Here are 5 that actually work: ELI5: [topic] — AI explains it like you're 5. Perfect when you're learning something totally new. Zero jargon. TLDR: [paste text] — Drops a 2000-word article down to 5 key points in seconds. I use this every morning for news. JARGONIZE: [your text] — Turns simple writing into something that sounds polished and professional. Great for LinkedIn or emails. HUMANIZE: [AI text] — The opposite. Strips out the robotic tone and makes it sound like a real person wrote it. FEYNMAN: [topic] — AI teaches you step by step until you can explain it back to someone else. This one actually makes stuff stick. Bonus: You can combine them. Try → ELI5 the TLDR of [paste article]. You get a 3-line summary a 5-year-old can follow. The PDF guide is in the post as well. Questions :-> 1.Try one today and drop below what you used it for ? 2. Any specific point you're struggling with Claude or any ai tool to solve?

Being able to build is the baseline, and if I receive an offer on a partnership where the offer is "I can build". I'll be frank, there are so many people that can build, I'd wager that it's expected. So why am I bringing this up? Because the highest leverage skill is not building. It's getting clients, closing deals, and bringing in revenue. I could create a post right now and look to hire somebody, and I'd get at least 10 people wanting to work. So if that's the case that everybody can build, where is the leverage as a dev? Well, speed and reliability. Take this for example. I took over a project from a dev a few weeks back, and one look at what he built told me everything I needed to know. This would not scale or work at all, and that's why they hired me. To fix the entire system because it was not working. I rebuilt what needed to be rebuilt within a day. Used my experience to make sure that it does scale. And if they asked for extra things? I replied within an hour and fixed it. In fact, this client was so surprised by my speed and knowledge that they asked for more. Which is my goal with every single client, turn a one-off into at least 3 orders. But this is only possible if you're on top of your game. So instead of offering "I can build and you can bring leads," offer speed and reliability. Because at that point you're different, and of course the bigger portfolio you have, the better it is. And I don't mean show me the biggest, most complex thing ever made. I want to see how many different builds are under the person's belt and how they're built. Because with experience you'll realize that there is a right way to build things. And without that experience? We'll end up with a bad build that works in demo but not in the real world. And proof from the client as well :)