New post ๐ข: how I used Claude Code to build a Chrome extension and an n8n workflow that automates my AI tech watch sharing. The whole thing in about 30 minutes.
www.56kode.com/posts/ai-pow...
New post ๐ข: how I used Claude Code to build a Chrome extension and an n8n workflow that automates my AI tech watch sharing. The whole thing in about 30 minutes.
www.56kode.com/posts/ai-pow...
๐ฅ Just discovered rtk-ai.app, a game-changer for Claude Code users!
It filters shell output noise before it reaches your AI:
โ
Fewer tokens = lower cost
โ
Longer sessions, no context reset needed
โ
More focused agent, less distraction
Thanks to Florian Bruniaux
Why I migrated from #Cursor to #ClaudeCode (and why you should too). โก๏ธ
I stopped typing code and started supervising agents in the terminal. The productivity gain is real. ๐ค
Read my full honest feedback here:
๐ www.56kode.com/posts/moving... #ai #agent #opus
Stop cleaning dead code manually! ๐
Discover a safe workflow combining Knip to find unused exports and an AI agent to verify and remove them without breaking production. ๐ค
๐ 56kode.com/posts/clean-...
#TypeScript #ts #ai #cursor #claudecode
Last week I asked #Gemini for a summary but forgot the URL. It gave me a summary anyway.
Apparently, prompt engineering is obsolete; weโre moving straight to telepathic engineering. Or just pure hallucinations. ๐
๐ฎ#AI
Forget โReasoning Avgโ and โCoding Avgโ. My only #AI benchmark: insults per day. #Gemini 3 Pro & Claude #Opus? So good I sometimes go 24h without calling them names. Thatโs basically AGI in my house.
After a new wave of discussion about CustomEvent vs Event subclasses, I wrote a new blog post on the topic:
TL;DR: Subclass Event instead of using CustomEvent ๐
justinfagnani.com/2025/06/25/s...
Very good, I like your implementation ๐
Our Vitest tests took 22s to start. Now? 4 seconds. ๐
How Gemini 3 Pro identified "God Modules" killing our performance and automated the fixโsaving us 15 min per CI run.
๐ www.56kode.com/posts/how-ge...
#Vitest #CI #react #reactjs
๐ Debug web apps directly from #Cursor #IDE with Browser-Tools #MCP! โก
๐น Console logs analysis
๐น Network requests inspection
๐น AI-powered debugging
๐ www.56kode.com/posts/connec...
#CursorIDE #AI #debug #chrome
I need to test this.
Yeah, thanks. I actually wanted to see the AI work on its own, but I should have added the documentation to the context.
Back from vacation โ๏ธ and straight into action!
I let AI handle a full dependency migration on my Astro blog: React 19, Astro 5 Content Layer, Tailwindโฆ all automated โก
Curious how it went?
๐ www.56kode.com/posts/automa...
#Astro #AI #cursor #cursorAI
Lost in your codebase? ๐คฏ Cursor's screenshot search feature is a game-changer! Just paste a UI screenshot and it finds the corresponding file for you. Saves so much time! #Cursor
AI agent stuck? ๐ซ Pro tip: Just open a new chat and restart! Often fixes the bug and gets things running smoothly again. #AI
๐ Tech Watch #15 is live!
๐ Compound Components in #React
๐ useSyncExternalStore demystified
๐ #AI & the future of dev jobs
๐ #TypeScript in the AI era
๐ Migrating from #ESLint to #BiomeJS
๐ open.substack.com/pub/56kode/p...
Skip the IDE-to-browser shuffle!
๐ Make Cursor talk to Jira with MCP
๐ฏ Get tickets, check specs, even auto-build test plans without leaving your editor.
๐ www.56kode.com/posts/integr... #Jira #CursorAI
๐ New article: Level Up #React - Mastering Context API
๐ Learn how React's Context API really works under the hood, whatโs new in React 19, and when to use it vs Zustand or Redux.
๐ Avoid unnecessary re-renders with proven optimization tips.
๐ Read โ www.56kode.com/posts/level-...
Mutation testing helps you find the truth:
Your tests either catch real issues โ or they don't.
Itโs not something to run on every push.
But from time to time? Itโs eye-opening ๐
So when should you use it?
๐งน When cleaning up a legacy project
๐ง When improving test quality
๐จ When you think your code is well tested โ but want proof
But be warned โ ๏ธ
Mutation testing is slow.
๐ It can multiply test time by 10x or more
๐งฑ It may require config tweaks to work with your test setup
๐ข In CI, it can make feedback loops way longer
Thatโs the magic of mutation testing:
โ
It doesnโt just check if your code is covered
โ
It checks if your tests are actually useful
And thatโs a huge difference.
Letโs say you wrote this:
return a > b;
Mutation testing might change it to:
return a < b;
If your tests still pass... do they really check the logic? ๐ค
๐ฅ Mutation testing works like this:
- The tool (like Stryker) changes your code on purpose
- It runs your test suite
- If tests fail โ good!
- If tests pass โ bad sign: the mutation slipped through
๐งช What if your tests... were being tested too?
A high code coverage doesnโt mean your app is bug-free.
It just means your code was executed by tests โ not that it was really verified.
Let me show you something better: mutation testing ๐
#testing #jest #vitest #js #ts
AI tools like ChatGPT or Claude write way better code with #TypeScript ๐ค๐ง
Your types give the AI context ๐
Well-typed code = better suggestions, safer refactors, fewer bugs.
TypeScript isnโt a nice-to-have. Itโs your AI co-pilotโs map ๐บ๏ธ
#ts #ai #cursor #windsurf
๐ Tech Watch #14 is live!
๐ Open-Closed Principle in React
๐ Pattern Matching in TanStack Query
๐ Faster Lazy Loading in React Router
๐ useEffect: Child vs Parent
๐ Cursor Rules for AI Dev in TS
๐ The Hidden Cost of AI Coding
๐ open.substack.com/pub/56kode/p...
There's nothing better than full remote work.
Want better answers from #AI?
๐ค First, ask what info itโs missing to help you better.
โ๏ธ Then, ask it to rewrite your prompt for clarity.
You donโt need to be perfect โ just collaborate smart ๐คโจ #chatpgt #claudeai #claude #gemini #deepseek