ANASS EZ-ZOUAINE (@Ansezz)

@elghorfimedOct 30

Shopify dev heads-up: if your app uses an App Proxy, verify the signature param on every request. Sort the other params, join them as key=value with no separators, HMAC-SHA256 with your app secret. Skip it and anyone can pass a fake logged_in_customer_id to your backend.

Mohamed El-Ghorfi · Oct 29, 2026 · 8:25 PM UTC

ANASS EZ-ZOUAINE retweeted

@elghorfimedOct 29

Shopify tip: treat order webhooks as a signal, not the source of truth. They can arrive late, out of order, or twice. On receipt, refetch the order by ID from the Admin API and only overwrite your copy if its updated_at is newer. Saves a lot of weird sync bugs during BFCM.

Mohamed El-Ghorfi · Oct 28, 2026 · 7:12 PM UTC

ANASS EZ-ZOUAINE retweeted

@elghorfimedOct 28

Shopify's Fall '26 Edition mixes four statuses: live, rolling out, early access and coming soon. Custom checkout fields are coming soon, and Shopify says not before or during BFCM. Apple Pay outside Safari is coming soon too. Plan BFCM work around what is live.

ANASS EZ-ZOUAINE · Oct 28, 2026 · 10:27 AM UTC

@AnsezzOct 28

All four keep a cloud computer working after you close the laptop. The split is who says yes before a send or a payment. And whether your plan and country can get in today. Claude Duet. Grok Bot. OpenAI Dots. Meta Muse. ansezz.com/blog/grok-bot-vs-…

Claude Duet vs Grok Bot vs OpenAI Dots vs Meta Muse: pick by control and access | ansezz

Four personal agents each keep a cloud computer working after you close the laptop. Pick by who says yes before a send or a payment, and whether your plan and country can get in.

ANASS EZ-ZOUAINE · Oct 28, 2026 · 9:21 AM UTC

@AnsezzOct 28

Meta built a second agent. Its only job is to say no to the first. Muse keeps working on a Secure VM after you close the app. Sentinel sits outside the model. Muse cannot override it. That is the bar for personal agents now. ansezz.com/blog/meta-muse-pe…

Meta Muse for Solo Founders: The Guard Outside the Agent Matters More Than the Agent | ansezz

Meta's personal agent takes action on its own cloud computer and keeps working after you close the app. Sentinel sits outside the model. Price as of 28 October 2026.

ANASS EZ-ZOUAINE · Oct 28, 2026 · 4:48 AM UTC

@AnsezzOct 28

Close your laptop. Your OpenAI Dot keeps working on its own cloud computer. DM it in Slack or Teams. Same memory. You set Custom Rules for what it does alone, what needs a yes, and what it hands back. Asking for a draft is not permission to send. Pro from $100/mo (as of 28 Oct 2026). Not in EEA/UK/CH. ansezz.com/blog/openai-dots-…

OpenAI Dots for Solo Founders: An Always-On Agent in Slack and Teams, on Your Rules | ansezz

One always-on ChatGPT agent keeps working on its own cloud computer after you close the laptop. Custom Rules, Slack and Teams, and the Pro price as of 28 October 2026.

Mohamed El-Ghorfi · Oct 27, 2026 · 8:19 PM UTC

ANASS EZ-ZOUAINE retweeted

@elghorfimedOct 27

Shopify dev heads-up: if your app calls the Admin API, log the X-Shopify-API-Deprecated-Reason response header. It flags the exact calls that will break in a future API version, months before they do. One log line turns into a free migration checklist.

ANASS EZ-ZOUAINE · Oct 26, 2026 · 6:05 PM UTC

@AnsezzOct 26

Closing the laptop used to stop the work. Now the Bots keep going. Named teammates on one shared computer. They pass the work. I keep the final say. Close your laptop. Your Bots keep working. ansezz.com/blog/grok-bot-sol…

Grok Bot for Solo Founders: Named AI Teammates on One Shared Computer | ansezz

Named AI teammates share one cloud computer and keep working after you close the laptop. Parallel work, approvals, and The X Subscription price as of 27 October 2026.

ANASS EZ-ZOUAINE · Oct 26, 2026 · 4:32 PM UTC

@AnsezzOct 26

An agent that acts alone is a risk. An agent that waits is usable. Claude Duet briefs you in the morning. Send, pay or delete? It shows you first. Worth the waitlist if you want that one agent. ansezz.com/blog/claude-duet-…

Claude Duet: a proactive agent that waits for your ok | ansezz

Claude Duet briefs you in the morning and waits before send, pay, or delete. What Anthropic has published, and why a solo founder might join the US Pro and Max waitlist.

ANASS EZ-ZOUAINE · Oct 26, 2026 · 4:32 PM UTC

@AnsezzOct 26

If you're still merging most PRs by hand, you're the bottleneck. I command. He opens the PR. I watch the result. That's the whole job now.

ANASS EZ-ZOUAINE · Oct 24, 2026 · 5:03 PM UTC

@AnsezzOct 24

your storefront shouldn't feel like a template. Hydrogen + Customer Account API = hyper-personalized journeys. show them what they need before they ask. loyalty is built on data-driven experiences, not just discount codes. time to get personal. #hydrogen #shopify #headless #ux

ANASS EZ-ZOUAINE · Oct 24, 2026 · 9:01 AM UTC

@AnsezzOct 24

stop letting heavy tracking scripts kill your LCP. Shopify Pixel API brings privacy-first, high-performance tracking to your Hydrogen storefront. clean data, fast loads, zero compromises. modernization means measuring smarter, not harder. #Shopify #Hydrogen #WebPerf #headless

ANASS EZ-ZOUAINE · Oct 23, 2026 · 6:04 PM UTC

@AnsezzOct 23

An agent that guesses is a liability. An agent that asks is a teammate. Mine fills my forms and clicks through the boring web stuff. Missing a detail? It asks me. A payment, a code or a send? I tap yes. All from my phone. My job now is answering its questions.

ANASS EZ-ZOUAINE · Oct 23, 2026 · 5:04 PM UTC

@AnsezzOct 23

is your codebase agent-readable? with laravel pao, we’re cleaning up outputs so agents don’t just read code, they validate it. better output means better automation. #laravel #aiagents #aiengineering #webdev

Mohamed El-Ghorfi · Oct 22, 2026 · 2:21 PM UTC

ANASS EZ-ZOUAINE retweeted

@elghorfimedOct 22

Pre-BFCM theme check: search theme.liquid and your snippets for code left behind by apps you uninstalled. Older apps injected script tags straight into the theme, and uninstalling doesn't always remove them. Dead scripts still cost load time on every page.

ANASS EZ-ZOUAINE · Oct 22, 2026 · 5:04 PM UTC

@AnsezzOct 22

prompt engineering was about vibes. agentic engineering is about rubrics. vibe checks are dead. rubric-based validation is the new standard. if an agent can't benchmark its own work against a strict rubric, it doesn't ship. no rubric, no deployment. i treat ai output like production code. it needs a rigorous rubric to validate tone, technical accuracy, and structure before it ever sees the light of day. stop guessing and start benchmarking. #agenticworkflows #aiengineering #softwarearchitecture #llm

Mohamed El-Ghorfi · Oct 21, 2026 · 2:22 PM UTC

ANASS EZ-ZOUAINE retweeted

@elghorfimedOct 21

Shopify dev heads-up: if your app subscribes to orders/updated, expect bursts during BFCM. Acknowledge the webhook fast with a 200 and push the work to a queue. Slow responses get retried, and a retry storm on top of a sales spike is how the same order ends up processed twice.

ANASS EZ-ZOUAINE · Oct 21, 2026 · 3:21 PM UTC

@AnsezzOct 21

PHP just patched two critical bugs. One request, no login, and an attacker can run code on your server. mod_php is the worst case. Update to 8.5.12, 8.4.27, 8.3.36 or 8.2.35. On 8.1 or older, PHP itself won't fix it. On Docker, pull the new base image and rebuild. Restarting the old one changes nothing.

133

ANASS EZ-ZOUAINE · Oct 21, 2026 · 3:50 PM UTC

@AnsezzOct 21

full breakdown, what happened and what to do now: ansezz.com/blog/php-security…

PHP 8.5.12, 8.4.27, 8.3.36 and 8.2.35: update your Laravel servers today | ansezz

PHP shipped fixes for two critical remote code execution bugs on 21 October 2026. Here is what is affected, what the interim settings do, and how to rebuild and redeploy a Laravel app in Docker.