This is the full primer behind my Agent Skills masterclass, a practical, no-fluff reference for building, evaluating, and shipping skills across Claude Code, ChatGPT and Gemini.
You'll get the open Agent Skills standard and SKILL [.] md file format; a clear mental model for skills vs MCP vs subagents vs plugins; platform-by-platform breakdowns for Claude Code, ChatGPT and Gemini, including their extended frontmatter and quirks; a full workflow for designing a new skill from scratch, sourcing real expertise, and avoiding the generic-instructions trap; techniques for writing descriptions that trigger reliably instead of silently failing; an eval-driven development framework for testing and scoring skills before you ship them.
23 pages, written for anyone (technical or non-technical) building or maintaining agent skills.
Share with friends, colleagues and other builders. Agent Skills are for everyone.
https://t.co/23mWivW9WV
@Ivan_Borovets Day 4 with 0 downloads is not a verdict, it just means nobody knows it exists yet. Research told you the problem is real, distribution is a separate job entirely. Go find the handstand corners of the internet and be useful there before you ever link the app.
@abs_uiux The checkout bug being nobody's fault is the worst part. Six vendors, six support queues, all of them technically right, and the merchant eats the lost sales. App sprawl is a real tax and most stores only notice it on the day something breaks.
@ashishpokiya010@Shopify@buildinpublic Congrats on getting it live 🚀 App Store review is its own boss fight 😅 push notifs done right is where the money hides imo 📲🔥
@gonogo_team The waiting part is the killer. Most founders know something is off way before they act, they just want one more month of data to make the decision feel safe. It never gets safer, it just gets more expensive.
@JohnEStrom71840@bookingcom This is the whole problem with support bots that can talk but cannot do anything. If it cannot actually pull your reservation and change it, it is just a slower hold music. Deflection metrics look great while the customer is stranded.
@alex_momeni Half the leaks I see are not on the page at all, they are post purchase. Wrong address, no order status, then a support ticket instead of a repeat order.
@jzazove Appreciate the follow through on this. For small app devs one unfair 1 star can move the rating more than a month of good work, so having a real appeals path matters a lot.
@Liam2307@dgalarza Cart and checkout are the honest test. Discovery is the easy half. The moment an agent has to touch a real order, address, tax, refund, you need write access with guardrails and Shopify staying the source of truth. That part is still messy.
@simonnyberg it never will be tbh. the goal is not perfect, it's that bad output gets caught before you ever see it. mine still writes nonsense sometimes, the tests just yell first
Shopify just shipped auto tax recalc when you change a shipping address.
Small thing, huge for anyone who's ever fixed an address after checkout and watched the numbers go weird.
Been living in that mess with merchants. Address edits are not edge cases. They're Tuesday.